python/python-xmldiff
SHA256
14
0
Fork 0
forked from pool/python-xmldiff
Code Pull Requests Activity
Files
ea611f4d5173c6c0131c54548c33e3ea8d7758b287b418ea832dd9755f99f672
python-xmldiff/python-xmldiff-src-upgrades.patch

11544 lines
461 KiB
Diff
Raw Blame History

From 34b810f4394965aadeca31204e6b76eb023fd11a Mon Sep 17 00:00:00 2001
From: Jeremy Lavergne <jlavergne@shoobx.com>
Date: Tue, 4 Aug 2020 08:32:14 -0400
Subject: [PATCH] src upgrades (#64)
* black
* pyupgrade
* py3+
* py3.6+
* black
* update python version deps
* travis: drop py < 3.6, use bionic
* flake fixes
* more flake8 fixups
* fixup black command
* try skipping black on pypy3
* travis: try figuring out build type by env
---
.travis.yml | 14 +-
Makefile | 5 +-
docs/source/conf.py | 158 +-
docs/source/contributing.rst | 6 +-
setup.py | 75 +-
tests/test_diff.py | 913 +++++----
tests/test_formatting.py | 394 ++--
tests/test_main.py | 83 +-
tests/test_patch.py | 169 +-
tests/test_utils.py | 116 +-
tests/testing.py | 20 +-
xmldiff/_diff_match_patch_py2.py | 3105 ++++++++++++++++--------------
xmldiff/_diff_match_patch_py3.py | 3080 +++++++++++++++--------------
xmldiff/actions.py | 22 +-
xmldiff/diff.py | 98 +-
xmldiff/diff_match_patch.py | 1 +
xmldiff/formatting.py | 246 +--
xmldiff/main.py | 170 +-
xmldiff/patch.py | 17 +-
xmldiff/utils.py | 57 +-
20 files changed, 4506 insertions(+), 4243 deletions(-)
Index: xmldiff-2.4/.travis.yml
===================================================================
--- xmldiff-2.4.orig/.travis.yml
+++ xmldiff-2.4/.travis.yml
@@ -4,17 +4,21 @@ language: python
matrix:
fast_finish: true
include:
- - python: 2.7
- - python: 3.5
- python: 3.6
+ env: MATRIX=py36
- python: 3.7
sudo: required
- dist: xenial
- - python: pypy
+ dist: bionic
+ env: MATRIX=py37
- python: pypy3
+ env: MATRIX=pypy3
+
+before_install:
+ - if [ $MATRIX != pypy3 ]; then pip install black; fi
+ - pip install coverage coveralls flake8 sphinx sphinx-argparse
install:
- - pip install . coverage coveralls flake8 sphinx sphinx-argparse
+ - pip install .
script:
- make flake
Index: xmldiff-2.4/Makefile
===================================================================
--- xmldiff-2.4.orig/Makefile
+++ xmldiff-2.4/Makefile
@@ -5,7 +5,10 @@ dfm_source_3 := "https://raw.githubuserc
all: coverage flake
flake:
- flake8 tests xmldiff --exclude *diff_match_patch*.py
+ifneq (, $(shell which black))
+ black --check .
+endif
+ flake8 tests xmldiff --exclude *diff_match_patch*.py --ignore=E231,E501,W503
coverage:
coverage run setup.py test
Index: xmldiff-2.4/docs/source/conf.py
===================================================================
--- xmldiff-2.4.orig/docs/source/conf.py
+++ xmldiff-2.4/docs/source/conf.py
@@ -1,4 +1,3 @@
-# -*- coding: utf-8 -*-
#
# xmldiff documentation build configuration file, created by
# sphinx-quickstart on Tue Sep 4 12:07:12 2018.
@@ -18,49 +17,49 @@ import os
# If extensions (or modules to document with autodoc) are in another directory,
# add these directories to sys.path here. If the directory is relative to the
# documentation root, use os.path.abspath to make it absolute, like shown here.
-#sys.path.insert(0, os.path.abspath('.'))
+# sys.path.insert(0, os.path.abspath('.'))
# -- General configuration ------------------------------------------------
# If your documentation needs a minimal Sphinx version, state it here.
-#needs_sphinx = '1.0'
+# needs_sphinx = '1.0'
# Add any Sphinx extension module names here, as strings. They can be
# extensions coming with Sphinx (named 'sphinx.ext.*') or your custom
# ones.
extensions = [
- 'sphinx.ext.doctest',
- 'sphinx.ext.coverage',
- 'sphinxarg.ext',
+ "sphinx.ext.doctest",
+ "sphinx.ext.coverage",
+ "sphinxarg.ext",
]
# Add any paths that contain templates here, relative to this directory.
-templates_path = ['_templates']
+templates_path = ["_templates"]
# The suffix(es) of source filenames.
# You can specify multiple suffix as a list of string:
# source_suffix = ['.rst', '.md']
-source_suffix = '.rst'
+source_suffix = ".rst"
# The encoding of source files.
-#source_encoding = 'utf-8-sig'
+# source_encoding = 'utf-8-sig'
# The master toctree document.
-master_doc = 'index'
+master_doc = "index"
# General information about the project.
-project = u'xmldiff'
-copyright = u'2018, Lennart Regebro'
-author = u'Lennart Regebro'
+project = "xmldiff"
+copyright = "2018, Lennart Regebro"
+author = "Lennart Regebro"
# The version info for the project you're documenting, acts as replacement for
# |version| and |release|, also used in various other places throughout the
# built documents.
#
# The short X.Y version.
-#version = u'2.0'
+# version = u'2.0'
# The full version, including alpha/beta/rc tags.
-#release = u'2.0'
+# release = u'2.0'
# The language for content autogenerated by Sphinx. Refer to documentation
# for a list of supported languages.
@@ -71,9 +70,9 @@ language = None
# There are two options for replacing |today|: either, you set today to some
# non-false value, then it is used:
-#today = ''
+# today = ''
# Else, today_fmt is used as the format for a strftime call.
-#today_fmt = '%B %d, %Y'
+# today_fmt = '%B %d, %Y'
# List of patterns, relative to source directory, that match files and
# directories to ignore when looking for source files.
@@ -81,27 +80,27 @@ exclude_patterns = []
# The reST default role (used for this markup: `text`) to use for all
# documents.
-#default_role = None
+# default_role = None
# If true, '()' will be appended to :func: etc. cross-reference text.
-#add_function_parentheses = True
+# add_function_parentheses = True
# If true, the current module name will be prepended to all description
# unit titles (such as .. function::).
-#add_module_names = True
+# add_module_names = True
# If true, sectionauthor and moduleauthor directives will be shown in the
# output. They are ignored by default.
-#show_authors = False
+# show_authors = False
# The name of the Pygments (syntax highlighting) style to use.
-pygments_style = 'sphinx'
+pygments_style = "sphinx"
# A list of ignored prefixes for module index sorting.
-#modindex_common_prefix = []
+# modindex_common_prefix = []
# If true, keep warnings as "system message" paragraphs in the built documents.
-#keep_warnings = False
+# keep_warnings = False
# If true, `todo` and `todoList` produce output, else they produce nothing.
todo_include_todos = False
@@ -111,156 +110,149 @@ todo_include_todos = False
# The theme to use for HTML and HTML Help pages. See the documentation for
# a list of builtin themes.
-html_theme = 'alabaster'
+html_theme = "alabaster"
# Theme options are theme-specific and customize the look and feel of a theme
# further. For a list of options available for each theme, see the
# documentation.
-#html_theme_options = {}
+# html_theme_options = {}
# Add any paths that contain custom themes here, relative to this directory.
-#html_theme_path = []
+# html_theme_path = []
# The name for this set of Sphinx documents. If None, it defaults to
# "<project> v<release> documentation".
-#html_title = None
+# html_title = None
# A shorter title for the navigation bar. Default is the same as html_title.
-#html_short_title = None
+# html_short_title = None
# The name of an image file (relative to this directory) to place at the top
# of the sidebar.
-#html_logo = None
+# html_logo = None
# The name of an image file (relative to this directory) to use as a favicon of
# the docs. This file should be a Windows icon file (.ico) being 16x16 or 32x32
# pixels large.
-#html_favicon = None
+# html_favicon = None
# Add any paths that contain custom static files (such as style sheets) here,
# relative to this directory. They are copied after the builtin static files,
# so a file named "default.css" will overwrite the builtin "default.css".
-html_static_path = ['static']
+html_static_path = ["static"]
# Add any extra paths that contain custom files (such as robots.txt or
# .htaccess) here, relative to this directory. These files are copied
# directly to the root of the documentation.
-#html_extra_path = []
+# html_extra_path = []
# If not '', a 'Last updated on:' timestamp is inserted at every page bottom,
# using the given strftime format.
-#html_last_updated_fmt = '%b %d, %Y'
+# html_last_updated_fmt = '%b %d, %Y'
# If true, SmartyPants will be used to convert quotes and dashes to
# typographically correct entities.
-#html_use_smartypants = True
+# html_use_smartypants = True
# Custom sidebar templates, maps document names to template names.
-#html_sidebars = {}
+# html_sidebars = {}
# Additional templates that should be rendered to pages, maps page names to
# template names.
-#html_additional_pages = {}
+# html_additional_pages = {}
# If false, no module index is generated.
-#html_domain_indices = True
+# html_domain_indices = True
# If false, no index is generated.
-#html_use_index = True
+# html_use_index = True
# If true, the index is split into individual pages for each letter.
-#html_split_index = False
+# html_split_index = False
# If true, links to the reST sources are added to the pages.
-#html_show_sourcelink = True
+# html_show_sourcelink = True
# If true, "Created using Sphinx" is shown in the HTML footer. Default is True.
-#html_show_sphinx = True
+# html_show_sphinx = True
# If true, "(C) Copyright ..." is shown in the HTML footer. Default is True.
-#html_show_copyright = True
+# html_show_copyright = True
# If true, an OpenSearch description file will be output, and all pages will
# contain a <link> tag referring to it. The value of this option must be the
# base URL from which the finished HTML is served.
-#html_use_opensearch = ''
+# html_use_opensearch = ''
# This is the file name suffix for HTML files (e.g. ".xhtml").
-#html_file_suffix = None
+# html_file_suffix = None
# Language to be used for generating the HTML full-text search index.
# Sphinx supports the following languages:
# 'da', 'de', 'en', 'es', 'fi', 'fr', 'hu', 'it', 'ja'
# 'nl', 'no', 'pt', 'ro', 'ru', 'sv', 'tr'
-#html_search_language = 'en'
+# html_search_language = 'en'
# A dictionary with options for the search language support, empty by default.
# Now only 'ja' uses this config value
-#html_search_options = {'type': 'default'}
+# html_search_options = {'type': 'default'}
# The name of a javascript file (relative to the configuration directory) that
# implements a search results scorer. If empty, the default will be used.
-#html_search_scorer = 'scorer.js'
+# html_search_scorer = 'scorer.js'
# Output file base name for HTML help builder.
-htmlhelp_basename = 'xmldiffdoc'
+htmlhelp_basename = "xmldiffdoc"
# -- Options for LaTeX output ---------------------------------------------
latex_elements = {
-# The paper size ('letterpaper' or 'a4paper').
-#'papersize': 'letterpaper',
-
-# The font size ('10pt', '11pt' or '12pt').
-#'pointsize': '10pt',
-
-# Additional stuff for the LaTeX preamble.
-#'preamble': '',
-
-# Latex figure (float) alignment
-#'figure_align': 'htbp',
+ # The paper size ('letterpaper' or 'a4paper').
+ #'papersize': 'letterpaper',
+ # The font size ('10pt', '11pt' or '12pt').
+ #'pointsize': '10pt',
+ # Additional stuff for the LaTeX preamble.
+ #'preamble': '',
+ # Latex figure (float) alignment
+ #'figure_align': 'htbp',
}
# Grouping the document tree into LaTeX files. List of tuples
# (source start file, target name, title,
# author, documentclass [howto, manual, or own class]).
latex_documents = [
- (master_doc, 'xmldiff.tex', u'xmldiff Documentation',
- u'Lennart Regebro', 'manual'),
+ (master_doc, "xmldiff.tex", "xmldiff Documentation", "Lennart Regebro", "manual"),
]
# The name of an image file (relative to this directory) to place at the top of
# the title page.
-#latex_logo = None
+# latex_logo = None
# For "manual" documents, if this is true, then toplevel headings are parts,
# not chapters.
-#latex_use_parts = False
+# latex_use_parts = False
# If true, show page references after internal links.
-#latex_show_pagerefs = False
+# latex_show_pagerefs = False
# If true, show URL addresses after external links.
-#latex_show_urls = False
+# latex_show_urls = False
# Documents to append as an appendix to all manuals.
-#latex_appendices = []
+# latex_appendices = []
# If false, no module index is generated.
-#latex_domain_indices = True
+# latex_domain_indices = True
# -- Options for manual page output ---------------------------------------
# One entry per manual page. List of tuples
# (source start file, name, description, authors, manual section).
-man_pages = [
- (master_doc, 'xmldiff', u'xmldiff Documentation',
- [author], 1)
-]
+man_pages = [(master_doc, "xmldiff", "xmldiff Documentation", [author], 1)]
# If true, show URL addresses after external links.
-#man_show_urls = False
+# man_show_urls = False
# -- Options for Texinfo output -------------------------------------------
@@ -269,19 +261,25 @@ man_pages = [
# (source start file, target name, title, author,
# dir menu entry, description, category)
texinfo_documents = [
- (master_doc, 'xmldiff', u'xmldiff Documentation',
- author, 'xmldiff', 'One line description of project.',
- 'Miscellaneous'),
+ (
+ master_doc,
+ "xmldiff",
+ "xmldiff Documentation",
+ author,
+ "xmldiff",
+ "One line description of project.",
+ "Miscellaneous",
+ ),
]
# Documents to append as an appendix to all manuals.
-#texinfo_appendices = []
+# texinfo_appendices = []
# If false, no module index is generated.
-#texinfo_domain_indices = True
+# texinfo_domain_indices = True
# How to display URL addresses: 'footnote', 'no', or 'inline'.
-#texinfo_show_urls = 'footnote'
+# texinfo_show_urls = 'footnote'
# If true, do not generate a @detailmenu in the "Top" node's menu.
-#texinfo_no_detailmenu = False
+# texinfo_no_detailmenu = False
Index: xmldiff-2.4/docs/source/contributing.rst
===================================================================
--- xmldiff-2.4.orig/docs/source/contributing.rst
+++ xmldiff-2.4/docs/source/contributing.rst
@@ -15,9 +15,9 @@ Setting Up a Development Environment
To set up a development environment you need a github account, git, and
of course Python with pip installed. You also should have the Python tools
-``coverage`` and ``flake8`` installed::
+``black``, ``coverage``, and ``flake8`` installed::
- pip install coverage flake8
+ pip install black coverage flake8
Then you need to clone the repository, and install it's dependencies::
@@ -44,8 +44,6 @@ The following test runners/commands are
* ``python setup.py test``
- * ``nosetests``
-
* ``pytest``
There is no support for ``tox`` to run test under different Python versions.
Index: xmldiff-2.4/setup.py
===================================================================
--- xmldiff-2.4.orig/setup.py
+++ xmldiff-2.4/setup.py
@@ -1,49 +1,45 @@
-from io import open
from setuptools import setup, find_packages
-version = '2.4'
+version = "2.5.dev0"
-with open('README.rst', 'rt', encoding='utf8') as readme:
+with open("README.rst", encoding="utf8") as readme:
description = readme.read()
-with open('CHANGES.rst', 'rt', encoding='utf8') as changes:
+with open("CHANGES.rst", encoding="utf8") as changes:
history = changes.read()
-setup(name='xmldiff',
- version=version,
- description="Creates diffs of XML files",
- long_description=description + '\n' + history,
- # Get strings from http://pypi.python.org/pypi?%3Aaction=list_classifiers
- classifiers=['Development Status :: 5 - Production/Stable',
- 'Topic :: Text Processing :: Markup :: XML',
- 'Operating System :: OS Independent',
- 'Programming Language :: Python :: 2',
- 'Programming Language :: Python :: 2.7',
- 'Programming Language :: Python :: 3',
- 'Programming Language :: Python :: 3.5',
- 'Programming Language :: Python :: 3.6',
- 'Programming Language :: Python :: 3.7',
- 'License :: OSI Approved :: MIT License',
- ],
- keywords='xml html diff',
- author='Lennart Regebro',
- author_email='lregebro@shoobx.com',
- url='https://github.com/Shoobx/xmldiff',
- license='MIT',
- packages=find_packages(exclude=['doc', 'tests']),
- include_package_data=True,
- zip_safe=False,
- install_requires=[
- 'setuptools',
- 'lxml>=3.1.0',
- 'six',
- ],
- test_suite='tests',
- entry_points={
- 'console_scripts': [
- 'xmldiff = xmldiff.main:diff_command',
- 'xmlpatch = xmldiff.main:patch_command',
- ],
- },
+setup(
+ name="xmldiff",
+ version=version,
+ description="Creates diffs of XML files",
+ long_description=description + "\n" + history,
+ # Get strings from http://pypi.python.org/pypi?%3Aaction=list_classifiers
+ classifiers=[
+ "Development Status :: 5 - Production/Stable",
+ "Topic :: Text Processing :: Markup :: XML",
+ "Operating System :: OS Independent",
+ "Programming Language :: Python :: 3",
+ "Programming Language :: Python :: 3.6",
+ "Programming Language :: Python :: 3.7",
+ "Programming Language :: Python :: 3.8",
+ "License :: OSI Approved :: MIT License",
+ ],
+ keywords="xml html diff",
+ author="Lennart Regebro",
+ author_email="lregebro@shoobx.com",
+ url="https://github.com/Shoobx/xmldiff",
+ license="MIT",
+ packages=find_packages(exclude=["doc", "tests"]),
+ include_package_data=True,
+ zip_safe=False,
+ install_requires=["setuptools", "lxml>=3.1.0", "six",],
+ test_suite="tests",
+ entry_points={
+ "console_scripts": [
+ "xmldiff = xmldiff.main:diff_command",
+ "xmlpatch = xmldiff.main:patch_command",
+ ],
+ },
)
+
Index: xmldiff-2.4/tests/test_diff.py
===================================================================
--- xmldiff-2.4.orig/tests/test_diff.py
+++ xmldiff-2.4/tests/test_diff.py
@@ -1,14 +1,22 @@
import os
import unittest
-from io import open
from lxml import etree
from xmldiff import utils
from xmldiff.diff import Differ
-from xmldiff.actions import (UpdateTextIn, InsertNode, MoveNode,
- DeleteNode, UpdateAttrib, InsertAttrib,
- RenameAttrib, DeleteAttrib, UpdateTextAfter,
- RenameNode, InsertComment)
+from xmldiff.actions import (
+ UpdateTextIn,
+ InsertNode,
+ MoveNode,
+ DeleteNode,
+ UpdateAttrib,
+ InsertAttrib,
+ RenameAttrib,
+ DeleteAttrib,
+ UpdateTextAfter,
+ RenameNode,
+ InsertComment,
+)
from .testing import compare_elements
@@ -16,20 +24,13 @@ from .testing import compare_elements
def dedent(string):
"""Remove the maximum common indent of the lines making up the string."""
lines = string.splitlines()
- indent = min(
- len(line) - len(line.lstrip())
- for line in lines
- if line
- )
- return "\n".join(
- line[indent:] if line else line
- for line in lines
- )
+ indent = min(len(line) - len(line.lstrip()) for line in lines if line)
+ return "\n".join(line[indent:] if line else line for line in lines)
class APITests(unittest.TestCase):
- left = u"<document><p>Text</p><p>More</p></document>"
- right = u"<document><p>Tokst</p><p>More</p></document>"
+ left = "<document><p>Text</p><p>More</p></document>"
+ right = "<document><p>Tokst</p><p>More</p></document>"
lefttree = etree.fromstring(left)
righttree = etree.fromstring(right)
differ = Differ()
@@ -107,9 +108,8 @@ class APITests(unittest.TestCase):
class NodeRatioTests(unittest.TestCase):
-
def test_compare_equal(self):
- xml = u"""<document>
+ xml = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="3" single-ref="3">
<para>First paragraph</para>
@@ -127,8 +127,10 @@ class NodeRatioTests(unittest.TestCase):
# Every node in these trees should get a 1.0 leaf_ratio,
# and if it has children, 1.0 child_ration, else None
- for left, right in zip(utils.post_order_traverse(differ.left),
- utils.post_order_traverse(differ.right)):
+ for left, right in zip(
+ utils.post_order_traverse(differ.left),
+ utils.post_order_traverse(differ.right),
+ ):
self.assertEqual(differ.leaf_ratio(left, right), 1.0)
if left.getchildren():
self.assertEqual(differ.child_ratio(left, right), 1.0)
@@ -136,7 +138,7 @@ class NodeRatioTests(unittest.TestCase):
self.assertIsNone(differ.child_ratio(left, right))
def test_compare_different_leafs(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="2" single-ref="2">
<para>This doesn't match at all</para>
@@ -151,7 +153,7 @@ class NodeRatioTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>Completely different from before</para>
@@ -172,26 +174,24 @@ class NodeRatioTests(unittest.TestCase):
# Make some choice comparisons here
# These node are exactly the same
- left = lefttree.xpath('/document/story/section[3]/para')[0]
- right = righttree.xpath('/document/story/section[3]/para')[0]
+ left = lefttree.xpath("/document/story/section[3]/para")[0]
+ right = righttree.xpath("/document/story/section[3]/para")[0]
self.assertEqual(differ.leaf_ratio(left, right), 1.0)
# These nodes have slightly different text, but no children
- left = lefttree.xpath('/document/story/section[2]/para')[0]
- right = righttree.xpath('/document/story/section[2]/para')[0]
+ left = lefttree.xpath("/document/story/section[2]/para")[0]
+ right = righttree.xpath("/document/story/section[2]/para")[0]
- self.assertAlmostEqual(differ.leaf_ratio(left, right),
- 0.75)
+ self.assertAlmostEqual(differ.leaf_ratio(left, right), 0.75)
# These nodes should not be very similar
- left = lefttree.xpath('/document/story/section[1]/para')[0]
- right = righttree.xpath('/document/story/section[1]/para')[0]
- self.assertAlmostEqual(differ.leaf_ratio(left, right),
- 0.45614035087719)
+ left = lefttree.xpath("/document/story/section[1]/para")[0]
+ right = righttree.xpath("/document/story/section[1]/para")[0]
+ self.assertAlmostEqual(differ.leaf_ratio(left, right), 0.45614035087719)
def test_compare_different_nodes(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="2" single-ref="2">
<para>First paragraph</para>
@@ -207,7 +207,7 @@ class NodeRatioTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="2" single-ref="2">
<para>First paragraph</para>
@@ -230,28 +230,28 @@ class NodeRatioTests(unittest.TestCase):
# Make some choice comparisons here. leaf_ratio will always be 1.0,
# as these leafs have the same attributes and no text, even though
# attributes may be in different order.
- left = differ.left.xpath('/document/story/section[1]')[0]
- right = differ.right.xpath('/document/story/section[1]')[0]
+ left = differ.left.xpath("/document/story/section[1]")[0]
+ right = differ.right.xpath("/document/story/section[1]")[0]
self.assertEqual(differ.leaf_ratio(left, right), 1.0)
# Only one of two matches:
self.assertEqual(differ.child_ratio(left, right), 0.5)
- left = differ.left.xpath('/document/story/section[2]')[0]
- right = differ.right.xpath('/document/story/section[2]')[0]
+ left = differ.left.xpath("/document/story/section[2]")[0]
+ right = differ.right.xpath("/document/story/section[2]")[0]
self.assertEqual(differ.leaf_ratio(left, right), 1.0)
# Only one of two matches:
self.assertEqual(differ.child_ratio(left, right), 0.5)
# These nodes should not be very similar
- left = differ.left.xpath('/document/story/section[3]')[0]
- right = differ.right.xpath('/document/story/section[3]')[0]
+ left = differ.left.xpath("/document/story/section[3]")[0]
+ right = differ.right.xpath("/document/story/section[3]")[0]
self.assertEqual(differ.leaf_ratio(left, right), 1.0)
self.assertEqual(differ.child_ratio(left, right), 1.0)
def test_compare_with_xmlid(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="1" single-ref="1">
<para>First paragraph</para>
@@ -267,7 +267,7 @@ class NodeRatioTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="newfirst" ref="1" single-ref="1">
<para>First paragraph</para>
@@ -289,8 +289,8 @@ class NodeRatioTests(unittest.TestCase):
# Make some choice comparisons here.
- left = differ.left.xpath('/document/story/section[1]')[0]
- right = differ.right.xpath('/document/story/section[1]')[0]
+ left = differ.left.xpath("/document/story/section[1]")[0]
+ right = differ.right.xpath("/document/story/section[1]")[0]
# These are very similar
self.assertEqual(differ.leaf_ratio(left, right), 0.9)
@@ -300,8 +300,8 @@ class NodeRatioTests(unittest.TestCase):
self.assertEqual(differ.node_ratio(left, right), 0)
# Here's the ones with the same id:
- left = differ.left.xpath('/document/story/section[1]')[0]
- right = differ.right.xpath('/document/story/section[2]')[0]
+ left = differ.left.xpath("/document/story/section[1]")[0]
+ right = differ.right.xpath("/document/story/section[2]")[0]
# Only one out of two children in common
self.assertEqual(differ.child_ratio(left, right), 0.5)
@@ -310,8 +310,8 @@ class NodeRatioTests(unittest.TestCase):
# The last ones are completely similar, but only one
# has an xml:id, so they do not match.
- left = differ.left.xpath('/document/story/section[3]')[0]
- right = differ.right.xpath('/document/story/section[3]')[0]
+ left = differ.left.xpath("/document/story/section[3]")[0]
+ right = differ.right.xpath("/document/story/section[3]")[0]
self.assertAlmostEqual(differ.leaf_ratio(left, right), 0.81818181818)
self.assertEqual(differ.child_ratio(left, right), 1.0)
self.assertEqual(differ.node_ratio(left, right), 0)
@@ -319,7 +319,8 @@ class NodeRatioTests(unittest.TestCase):
def test_compare_with_uniqueattrs(self):
# `uniqueattrs` can be pairs of (tag, attribute) as well as just string
# attributes.
- left = dedent(u"""\
+ left = dedent(
+ """\
<document>
<story firstPageTemplate="FirstPage">
<section name="oldfirst" ref="1" single-ref="1">
@@ -334,9 +335,11 @@ class NodeRatioTests(unittest.TestCase):
</section>
</story>
</document>
- """)
+ """
+ )
- right = dedent(u"""\
+ right = dedent(
+ """\
<document>
<story firstPageTemplate="FirstPage">
<section name="newfirst" ref="1" single-ref="1">
@@ -355,19 +358,22 @@ class NodeRatioTests(unittest.TestCase):
</subsection>
</story>
</document>
- """)
+ """
+ )
- differ = Differ(uniqueattrs=[
- ('section', 'name'),
- '{http://www.w3.org/XML/1998/namespace}id'
- ])
+ differ = Differ(
+ uniqueattrs=[
+ ("section", "name"),
+ "{http://www.w3.org/XML/1998/namespace}id",
+ ]
+ )
differ.set_trees(etree.fromstring(left), etree.fromstring(right))
differ.match()
# Make some choice comparisons here.
- left = differ.left.xpath('/document/story/section[1]')[0]
- right = differ.right.xpath('/document/story/section[1]')[0]
+ left = differ.left.xpath("/document/story/section[1]")[0]
+ right = differ.right.xpath("/document/story/section[1]")[0]
# These are very similar
self.assertEqual(differ.leaf_ratio(left, right), 0.90625)
@@ -377,8 +383,8 @@ class NodeRatioTests(unittest.TestCase):
self.assertEqual(differ.node_ratio(left, right), 0)
# Here's the ones with the same tag and name attribute:
- left = differ.left.xpath('/document/story/section[1]')[0]
- right = differ.right.xpath('/document/story/section[2]')[0]
+ left = differ.left.xpath("/document/story/section[1]")[0]
+ right = differ.right.xpath("/document/story/section[2]")[0]
# Only one out of two children in common
self.assertEqual(differ.child_ratio(left, right), 0)
@@ -387,29 +393,29 @@ class NodeRatioTests(unittest.TestCase):
# The last ones are completely similar, but only one
# has an name, so they do not match.
- left = differ.left.xpath('/document/story/section[3]')[0]
- right = differ.right.xpath('/document/story/section[3]')[0]
+ left = differ.left.xpath("/document/story/section[3]")[0]
+ right = differ.right.xpath("/document/story/section[3]")[0]
self.assertAlmostEqual(differ.leaf_ratio(left, right), 0.78260869565)
self.assertEqual(differ.child_ratio(left, right), 1.0)
self.assertEqual(differ.node_ratio(left, right), 0)
# Now these are structurally similar, have the same name, but
# one of them is not a section, so the uniqueattr does not match
- left = differ.left.xpath('/document/story/section[1]')[0]
- right = differ.right.xpath('/document/story/subsection[1]')[0]
+ left = differ.left.xpath("/document/story/section[1]")[0]
+ right = differ.right.xpath("/document/story/subsection[1]")[0]
self.assertAlmostEqual(differ.leaf_ratio(left, right), 1.0)
self.assertEqual(differ.child_ratio(left, right), 0.5)
self.assertAlmostEqual(differ.node_ratio(left, right), 0.75)
def test_compare_node_rename(self):
- left = u"""<document>
+ left = """<document>
<para>First paragraph</para>
<para attr="value">Second paragraph</para>
<para attr="value">Third paragraph</para>
</document>
"""
- right = u"""<document>
+ right = """<document>
<section>First paragraph</section>
<section attr="something else">Second paragraph</section>
<section attr="something else">A different text</section>
@@ -421,38 +427,35 @@ class NodeRatioTests(unittest.TestCase):
differ.match()
# Make some choice comparisons here.
- left = differ.left.xpath('/document/para[1]')[0]
- right = differ.right.xpath('/document/section[1]')[0]
+ left = differ.left.xpath("/document/para[1]")[0]
+ right = differ.right.xpath("/document/section[1]")[0]
# These have different tags, but should still match
- self.assertEqual(differ.leaf_ratio(left, right),
- 1.0)
+ self.assertEqual(differ.leaf_ratio(left, right), 1.0)
# These have different tags, and different attribute value,
# but still similar enough
- left = differ.left.xpath('/document/para[2]')[0]
- right = differ.right.xpath('/document/section[2]')[0]
+ left = differ.left.xpath("/document/para[2]")[0]
+ right = differ.right.xpath("/document/section[2]")[0]
# These have different tags, but should still match
- self.assertAlmostEqual(differ.leaf_ratio(left, right),
- 0.76190476190476)
+ self.assertAlmostEqual(differ.leaf_ratio(left, right), 0.76190476190476)
# These have different tags, and different attribute value,
# but still similar enough
- left = differ.left.xpath('/document/para[3]')[0]
- right = differ.right.xpath('/document/section[3]')[0]
+ left = differ.left.xpath("/document/para[3]")[0]
+ right = differ.right.xpath("/document/section[3]")[0]
# These are too different
- self.assertAlmostEqual(differ.leaf_ratio(left, right),
- 0.45161290322580)
+ self.assertAlmostEqual(differ.leaf_ratio(left, right), 0.45161290322580)
def test_compare_namespaces(self):
- left = u"""<document>
+ left = """<document>
<foo:para xmlns:foo="someuri">First paragraph</foo:para>
</document>
"""
- right = u"""<document>
+ right = """<document>
<foo:para xmlns:foo="otheruri">First paragraph</foo:para>
</document>
"""
@@ -462,46 +465,46 @@ class NodeRatioTests(unittest.TestCase):
differ.match()
# Make some choice comparisons here.
- left = differ.left.xpath('/document/foo:para[1]',
- namespaces={'foo': 'someuri'})[0]
- right = differ.right.xpath('/document/foo:para[1]',
- namespaces={'foo': 'otheruri'})[0]
+ left = differ.left.xpath(
+ "/document/foo:para[1]", namespaces={"foo": "someuri"}
+ )[0]
+ right = differ.right.xpath(
+ "/document/foo:para[1]", namespaces={"foo": "otheruri"}
+ )[0]
# These have different namespaces, but should still match
- self.assertEqual(differ.leaf_ratio(left, right),
- 1.0)
+ self.assertEqual(differ.leaf_ratio(left, right), 1.0)
def test_different_ratio_modes(self):
- node1 = etree.Element('para')
+ node1 = etree.Element("para")
node1.text = "This doesn't match at all"
- node2 = etree.Element('para')
+ node2 = etree.Element("para")
node2.text = "It's completely different"
- node3 = etree.Element('para')
+ node3 = etree.Element("para")
node3.text = "Completely different from before"
# These texts are very different
- differ = Differ(ratio_mode='accurate')
+ differ = Differ(ratio_mode="accurate")
self.assertAlmostEqual(differ.leaf_ratio(node1, node2), 0.24)
# However, the quick_ratio doesn't catch that, and think they match
- differ = Differ(ratio_mode='fast')
+ differ = Differ(ratio_mode="fast")
self.assertAlmostEqual(differ.leaf_ratio(node1, node2), 0.64)
# It still realizes these sentences are different, though.
- differ = Differ(ratio_mode='fast')
+ differ = Differ(ratio_mode="fast")
self.assertAlmostEqual(differ.leaf_ratio(node1, node3), 0.4561403508)
# Faster thinks the first two are the same!
- differ = Differ(ratio_mode='faster')
+ differ = Differ(ratio_mode="faster")
self.assertAlmostEqual(differ.leaf_ratio(node1, node2), 1.0)
# And that the third is almost the same
- differ = Differ(ratio_mode='faster')
+ differ = Differ(ratio_mode="faster")
self.assertAlmostEqual(differ.leaf_ratio(node1, node3), 0.8771929824)
# Invalid modes raise error:
with self.assertRaises(ValueError):
- differ = Differ(ratio_mode='allezlebleus')
+ differ = Differ(ratio_mode="allezlebleus")
class MatchTests(unittest.TestCase):
-
def _match(self, left, right):
left_tree = etree.fromstring(left)
right_tree = etree.fromstring(right)
@@ -513,7 +516,7 @@ class MatchTests(unittest.TestCase):
return [(lpath(item[0]), rpath(item[1])) for item in matches]
def test_same_tree(self):
- xml = u"""<document>
+ xml = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="3" single-ref="3">
<para>First paragraph</para>
@@ -537,7 +540,7 @@ class MatchTests(unittest.TestCase):
# If the number of similar attributes are few it works fine, the
# differing content of the ref="3" section means it's detected to
# be an insert.
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -551,7 +554,7 @@ class MatchTests(unittest.TestCase):
# We even detect that the first section is an insert without
# xmlid, but that's less reliable.
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>New paragraph</para>
@@ -567,27 +570,24 @@ class MatchTests(unittest.TestCase):
"""
result = self._match(left, right)
- self.assertEqual(result, [
- ('/document/story/section[1]/para',
- '/document/story/section[2]/para'),
- ('/document/story/section[1]',
- '/document/story/section[2]'),
- ('/document/story/section[2]/para',
- '/document/story/section[3]/para'),
- ('/document/story/section[2]',
- '/document/story/section[3]'),
- ('/document/story',
- '/document/story'),
- ('/document',
- '/document')
- ])
+ self.assertEqual(
+ result,
+ [
+ ("/document/story/section[1]/para", "/document/story/section[2]/para"),
+ ("/document/story/section[1]", "/document/story/section[2]"),
+ ("/document/story/section[2]/para", "/document/story/section[3]/para"),
+ ("/document/story/section[2]", "/document/story/section[3]"),
+ ("/document/story", "/document/story"),
+ ("/document", "/document"),
+ ],
+ )
def test_with_xmlid(self):
# This first section contains attributes that are similar (and longer
# than the content text. That would trick the matcher into matching
# the oldfirst and the newfirst section to match, except that we
# this time also have xml:id's, and they trump everything else!
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3" xml:id="oldfirst"
description="This is to trick the differ">
@@ -605,7 +605,7 @@ class MatchTests(unittest.TestCase):
# We even detect that the first section is an insert without
# xmlid, but that's less reliable.
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3" xml:id="newfirst"
description="This is to trick the differ">
@@ -625,28 +625,23 @@ class MatchTests(unittest.TestCase):
"""
result = self._match(left, right)
- self.assertEqual(result, [
- ('/document/story/section[1]/para',
- '/document/story/section[2]/para'),
- ('/document/story/section[1]',
- '/document/story/section[2]'),
- ('/document/story/section[2]/para',
- '/document/story/section[3]/para'),
- ('/document/story/section[2]',
- '/document/story/section[3]'),
- ('/document/story/section[3]/para',
- '/document/story/section[4]/para'),
- ('/document/story/section[3]',
- '/document/story/section[4]'),
- ('/document/story',
- '/document/story'),
- ('/document',
- '/document')
- ])
+ self.assertEqual(
+ result,
+ [
+ ("/document/story/section[1]/para", "/document/story/section[2]/para"),
+ ("/document/story/section[1]", "/document/story/section[2]"),
+ ("/document/story/section[2]/para", "/document/story/section[3]/para"),
+ ("/document/story/section[2]", "/document/story/section[3]"),
+ ("/document/story/section[3]/para", "/document/story/section[4]/para"),
+ ("/document/story/section[3]", "/document/story/section[4]"),
+ ("/document/story", "/document/story"),
+ ("/document", "/document"),
+ ],
+ )
def test_change_attribs(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="3" single-ref="3">
<para>First</para>
@@ -658,7 +653,7 @@ class MatchTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="4" single-ref="4">
<para>First</para>
@@ -672,23 +667,20 @@ class MatchTests(unittest.TestCase):
# It matches everything straight, which means the attrib changes
# should become updates, which makes sense.
result = self._match(left, right)
- self.assertEqual(result, [
- ('/document/story/section[1]/para',
- '/document/story/section[1]/para'),
- ('/document/story/section[1]',
- '/document/story/section[1]'),
- ('/document/story/section[2]/para',
- '/document/story/section[2]/para'),
- ('/document/story/section[2]',
- '/document/story/section[2]'),
- ('/document/story',
- '/document/story'),
- ('/document',
- '/document')
- ])
+ self.assertEqual(
+ result,
+ [
+ ("/document/story/section[1]/para", "/document/story/section[1]/para"),
+ ("/document/story/section[1]", "/document/story/section[1]"),
+ ("/document/story/section[2]/para", "/document/story/section[2]/para"),
+ ("/document/story/section[2]", "/document/story/section[2]"),
+ ("/document/story", "/document/story"),
+ ("/document", "/document"),
+ ],
+ )
def test_move_paragraph(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -701,7 +693,7 @@ class MatchTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -714,18 +706,27 @@ class MatchTests(unittest.TestCase):
</document>
"""
result = self._match(left, right)
- self.assertEqual(result, [
- ('/document/story/section[1]/para[1]',
- '/document/story/section[1]/para'),
- ('/document/story/section[1]/para[2]',
- '/document/story/section[2]/para[1]'),
- ('/document/story/section[1]', '/document/story/section[1]'),
- ('/document/story/section[2]/para',
- '/document/story/section[2]/para[2]'),
- ('/document/story/section[2]', '/document/story/section[2]'),
- ('/document/story', '/document/story'),
- ('/document', '/document')
- ])
+ self.assertEqual(
+ result,
+ [
+ (
+ "/document/story/section[1]/para[1]",
+ "/document/story/section[1]/para",
+ ),
+ (
+ "/document/story/section[1]/para[2]",
+ "/document/story/section[2]/para[1]",
+ ),
+ ("/document/story/section[1]", "/document/story/section[1]"),
+ (
+ "/document/story/section[2]/para",
+ "/document/story/section[2]/para[2]",
+ ),
+ ("/document/story/section[2]", "/document/story/section[2]"),
+ ("/document/story", "/document/story"),
+ ("/document", "/document"),
+ ],
+ )
def test_match_complex_text(self):
left = """<wrap id="1533728456.41"><para>
@@ -769,50 +770,50 @@ class MatchTests(unittest.TestCase):
</para></wrap>"""
result = self._match(left, right)
- self.assertEqual(result, [
- ('/wrap/para/b', '/wrap/para/b'),
- ('/wrap/para', '/wrap/para'),
- ('/wrap', '/wrap')
- ])
+ self.assertEqual(
+ result,
+ [
+ ("/wrap/para/b", "/wrap/para/b"),
+ ("/wrap/para", "/wrap/para"),
+ ("/wrap", "/wrap"),
+ ],
+ )
def test_match_insert_node(self):
- left = u'''<document title="insert-node">
+ left = """<document title="insert-node">
<story id="id">
</story>
</document>
-'''
- right = u'''<document title="insert-node">
+"""
+ right = """<document title="insert-node">
<story id="id">
<h1>Inserted <i>Node</i></h1>
</story>
-</document>'''
+</document>"""
result = self._match(left, right)
- self.assertEqual(result, [
- ('/document/story', '/document/story'),
- ('/document', '/document'),
- ])
+ self.assertEqual(
+ result,
+ [("/document/story", "/document/story"), ("/document", "/document"),],
+ )
def test_entirely_different(self):
- left = u'''<document title="insert-node">
+ left = """<document title="insert-node">
<story id="id">
</story>
</document>
-'''
- right = u'''<document title="something else">
+"""
+ right = """<document title="something else">
<h1>Inserted <i>Node</i></h1>
-</document>'''
+</document>"""
result = self._match(left, right)
- self.assertEqual(result, [
- ('/document', '/document'),
- ])
+ self.assertEqual(result, [("/document", "/document"),])
class FastMatchTests(unittest.TestCase):
-
def _match(self, left, right, fast_match):
left_tree = etree.fromstring(left)
right_tree = etree.fromstring(right)
@@ -824,7 +825,7 @@ class FastMatchTests(unittest.TestCase):
return [(lpath(item[0]), rpath(item[1])) for item in matches]
def test_move_paragraph(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -837,7 +838,7 @@ class FastMatchTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -858,7 +859,7 @@ class FastMatchTests(unittest.TestCase):
def test_move_children(self):
# Here the paragraphs are all so similar that that each paragraph
# will match any other.
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -869,7 +870,7 @@ class FastMatchTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>Second paragraph</para>
@@ -882,31 +883,31 @@ class FastMatchTests(unittest.TestCase):
# The slow match will match the nodes that match *best*, so it will
# find that paragraphs have moved around.
slow_result = sorted(self._match(left, right, False))
- self.assertEqual(slow_result, [
- ('/document', '/document'),
- ('/document/story', '/document/story'),
- ('/document/story/section', '/document/story/section'),
- ('/document/story/section/para[1]',
- '/document/story/section/para[3]'),
- ('/document/story/section/para[2]',
- '/document/story/section/para[1]'),
- ('/document/story/section/para[3]',
- '/document/story/section/para[2]')
- ])
+ self.assertEqual(
+ slow_result,
+ [
+ ("/document", "/document"),
+ ("/document/story", "/document/story"),
+ ("/document/story/section", "/document/story/section"),
+ ("/document/story/section/para[1]", "/document/story/section/para[3]"),
+ ("/document/story/section/para[2]", "/document/story/section/para[1]"),
+ ("/document/story/section/para[3]", "/document/story/section/para[2]"),
+ ],
+ )
# But the fast match will just pick any that matches.
fast_result = sorted(self._match(left, right, True))
- self.assertEqual(fast_result, [
- ('/document', '/document'),
- ('/document/story', '/document/story'),
- ('/document/story/section', '/document/story/section'),
- ('/document/story/section/para[1]',
- '/document/story/section/para[1]'),
- ('/document/story/section/para[2]',
- '/document/story/section/para[2]'),
- ('/document/story/section/para[3]',
- '/document/story/section/para[3]')
- ])
+ self.assertEqual(
+ fast_result,
+ [
+ ("/document", "/document"),
+ ("/document/story", "/document/story"),
+ ("/document/story/section", "/document/story/section"),
+ ("/document/story/section/para[1]", "/document/story/section/para[1]"),
+ ("/document/story/section/para[2]", "/document/story/section/para[2]"),
+ ("/document/story/section/para[3]", "/document/story/section/para[3]"),
+ ],
+ )
class UpdateNodeTests(unittest.TestCase):
@@ -926,7 +927,7 @@ class UpdateNodeTests(unittest.TestCase)
return steps
def test_same_tree(self):
- xml = u"""<document>
+ xml = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="3" single-ref="3">
<para>First paragraph</para>
@@ -942,24 +943,28 @@ class UpdateNodeTests(unittest.TestCase)
self.assertEqual(result, [])
def test_attribute_changes(self):
- left = u"""<root><node attr1="ohyeah" attr2="ohno" attr3="maybe" """\
- u"""attr0="del">The contained text</node>And a tail!</root>"""
+ left = (
+ """<root><node attr1="ohyeah" attr2="ohno" attr3="maybe" """
+ """attr0="del">The contained text</node>And a tail!</root>"""
+ )
- right = u"""<root><node attr4="ohyeah" attr2="uhhuh" attr3="maybe" """\
- u"""attr5="new">The new text</node>Also a tail!</root>"""
+ right = (
+ """<root><node attr4="ohyeah" attr2="uhhuh" attr3="maybe" """
+ """attr5="new">The new text</node>Also a tail!</root>"""
+ )
result = self._match(left, right)
self.assertEqual(
result,
[
- UpdateAttrib('/root/node[1]', 'attr2', 'uhhuh'),
- RenameAttrib('/root/node[1]', 'attr1', 'attr4'),
- InsertAttrib('/root/node[1]', 'attr5', 'new'),
- DeleteAttrib('/root/node[1]', 'attr0'),
- UpdateTextIn('/root/node[1]', 'The new text'),
- UpdateTextAfter('/root/node[1]', 'Also a tail!'),
- ]
+ UpdateAttrib("/root/node[1]", "attr2", "uhhuh"),
+ RenameAttrib("/root/node[1]", "attr1", "attr4"),
+ InsertAttrib("/root/node[1]", "attr5", "new"),
+ DeleteAttrib("/root/node[1]", "attr0"),
+ UpdateTextIn("/root/node[1]", "The new text"),
+ UpdateTextAfter("/root/node[1]", "Also a tail!"),
+ ],
)
@@ -978,7 +983,7 @@ class AlignChildrenTests(unittest.TestCa
return steps
def test_same_tree(self):
- xml = u"""<document>
+ xml = """<document>
<story firstPageTemplate="FirstPage">
<section xml:id="oldfirst" ref="3" single-ref="3">
<para>First paragraph</para>
@@ -994,7 +999,7 @@ class AlignChildrenTests(unittest.TestCa
self.assertEqual(result, [])
def test_move_paragraph(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -1007,7 +1012,7 @@ class AlignChildrenTests(unittest.TestCa
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -1024,7 +1029,7 @@ class AlignChildrenTests(unittest.TestCa
self.assertEqual(result, [])
def test_move_children(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -1035,7 +1040,7 @@ class AlignChildrenTests(unittest.TestCa
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>Second paragraph</para>
@@ -1046,9 +1051,14 @@ class AlignChildrenTests(unittest.TestCa
</document>
"""
result = self._align(left, right)
- self.assertEqual(result,
- [MoveNode('/document/story/section/para[1]',
- '/document/story/section[1]', 2)])
+ self.assertEqual(
+ result,
+ [
+ MoveNode(
+ "/document/story/section/para[1]", "/document/story/section[1]", 2
+ )
+ ],
+ )
class DiffTests(unittest.TestCase):
@@ -1065,7 +1075,7 @@ class DiffTests(unittest.TestCase):
return editscript
def test_process(self):
- left = u"""<document>
+ left = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -1079,7 +1089,7 @@ class DiffTests(unittest.TestCase):
</document>
"""
- right = u"""<document>
+ right = """<document>
<story firstPageTemplate="FirstPage">
<section ref="3" single-ref="3">
<para>First paragraph</para>
@@ -1096,17 +1106,19 @@ class DiffTests(unittest.TestCase):
self.assertEqual(
result,
[
- InsertNode('/document/story[1]', 'section', 1),
- InsertAttrib('/document/story/section[2]', 'ref', '4'),
- InsertAttrib('/document/story/section[2]', 'single-ref', '4'),
- MoveNode('/document/story/section[1]/para[3]',
- '/document/story/section[2]', 0),
- InsertNode('/document/story/section[2]', 'para', 1),
- UpdateTextIn('/document/story/section[2]/para[2]',
- 'Fourth paragraph'),
- DeleteNode('/document/story/deleteme/para[1]'),
- DeleteNode('/document/story/deleteme[1]'),
- ]
+ InsertNode("/document/story[1]", "section", 1),
+ InsertAttrib("/document/story/section[2]", "ref", "4"),
+ InsertAttrib("/document/story/section[2]", "single-ref", "4"),
+ MoveNode(
+ "/document/story/section[1]/para[3]",
+ "/document/story/section[2]",
+ 0,
+ ),
+ InsertNode("/document/story/section[2]", "para", 1),
+ UpdateTextIn("/document/story/section[2]/para[2]", "Fourth paragraph"),
+ DeleteNode("/document/story/deleteme/para[1]"),
+ DeleteNode("/document/story/deleteme[1]"),
+ ],
)
def test_needs_align(self):
@@ -1116,34 +1128,36 @@ class DiffTests(unittest.TestCase):
self.assertEqual(
result,
[
- MoveNode('/root/n[1]', '/root[1]', 1),
- MoveNode('/root/n[2]/p[2]', '/root/n[1]', 0),
- ]
+ MoveNode("/root/n[1]", "/root[1]", 1),
+ MoveNode("/root/n[2]/p[2]", "/root/n[1]", 0),
+ ],
)
def test_no_root_match(self):
- left = '<root attr="val"><root><n><p>1</p><p>2</p><p>3</p></n>'\
- '<n><p>4</p></n></root></root>'
- right = '<root><n><p>2</p><p>4</p></n><n><p>1</p><p>3</p></n></root>'
+ left = (
+ '<root attr="val"><root><n><p>1</p><p>2</p><p>3</p></n>'
+ "<n><p>4</p></n></root></root>"
+ )
+ right = "<root><n><p>2</p><p>4</p></n><n><p>1</p><p>3</p></n></root>"
result = self._diff(left, right)
self.assertEqual(
result,
[
- DeleteAttrib('/root[1]', 'attr'),
- MoveNode('/root/root/n[2]', '/root[1]', 0),
- MoveNode('/root/root/n[1]', '/root[1]', 1),
- MoveNode('/root/n[2]/p[2]', '/root/n[1]', 0),
- DeleteNode('/root/root[1]')
- ]
+ DeleteAttrib("/root[1]", "attr"),
+ MoveNode("/root/root/n[2]", "/root[1]", 0),
+ MoveNode("/root/root/n[1]", "/root[1]", 1),
+ MoveNode("/root/n[2]/p[2]", "/root/n[1]", 0),
+ DeleteNode("/root/root[1]"),
+ ],
)
def test_rmldoc(self):
here = os.path.split(__file__)[0]
- lfile = os.path.join(here, 'test_data', 'rmldoc.left.xml')
- rfile = os.path.join(here, 'test_data', 'rmldoc.right.xml')
- with open(lfile, 'rt', encoding='utf8') as infile:
+ lfile = os.path.join(here, "test_data", "rmldoc.left.xml")
+ rfile = os.path.join(here, "test_data", "rmldoc.right.xml")
+ with open(lfile, encoding="utf8") as infile:
left = infile.read()
- with open(rfile, 'rt', encoding='utf8') as infile:
+ with open(rfile, encoding="utf8") as infile:
right = infile.read()
result = self._diff(left, right)
@@ -1151,240 +1165,197 @@ class DiffTests(unittest.TestCase):
result,
[
InsertNode(
- '/document/story[1]',
- '{http://namespaces.shoobx.com/application}section',
- 4),
- InsertAttrib(
- '/document/story/app:section[4]', 'hidden', 'false'),
- InsertAttrib(
- '/document/story/app:section[4]', 'name', 'sign'),
- InsertAttrib(
- '/document/story/app:section[4]', 'ref', '3'),
- InsertAttrib(
- '/document/story/app:section[4]', 'removed', 'false'),
- InsertAttrib(
- '/document/story/app:section[4]', 'single-ref', '3'),
+ "/document/story[1]",
+ "{http://namespaces.shoobx.com/application}section",
+ 4,
+ ),
+ InsertAttrib("/document/story/app:section[4]", "hidden", "false"),
+ InsertAttrib("/document/story/app:section[4]", "name", "sign"),
+ InsertAttrib("/document/story/app:section[4]", "ref", "3"),
+ InsertAttrib("/document/story/app:section[4]", "removed", "false"),
+ InsertAttrib("/document/story/app:section[4]", "single-ref", "3"),
InsertAttrib(
- '/document/story/app:section[4]', 'title', 'Signing Bonus'),
- UpdateAttrib('/document/story/app:section[5]', 'ref', '4'),
- UpdateAttrib(
- '/document/story/app:section[5]', 'single-ref', '4'),
- UpdateAttrib('/document/story/app:section[6]', 'ref', '5'),
- UpdateAttrib(
- '/document/story/app:section[6]', 'single-ref', '5'),
- UpdateAttrib('/document/story/app:section[7]', 'ref', '6'),
- UpdateAttrib(
- '/document/story/app:section[7]', 'single-ref', '6'),
- UpdateAttrib('/document/story/app:section[8]', 'ref', '7'),
- UpdateAttrib(
- '/document/story/app:section[8]', 'single-ref', '7'),
- UpdateAttrib('/document/story/app:section[9]', 'ref', '8'),
- UpdateAttrib(
- '/document/story/app:section[9]', 'single-ref', '8'),
- UpdateAttrib('/document/story/app:section[10]', 'ref', '9'),
- UpdateAttrib(
- '/document/story/app:section[10]', 'single-ref', '9'),
- UpdateAttrib('/document/story/app:section[11]', 'ref', '10'),
- UpdateAttrib(
- '/document/story/app:section[11]', 'single-ref', '10'),
- UpdateAttrib('/document/story/app:section[12]', 'ref', '11'),
- UpdateAttrib(
- '/document/story/app:section[12]', 'single-ref', '11'),
- UpdateAttrib('/document/story/app:section[14]', 'ref', '12'),
- UpdateAttrib(
- '/document/story/app:section[14]', 'single-ref', '12'),
+ "/document/story/app:section[4]", "title", "Signing Bonus"
+ ),
+ UpdateAttrib("/document/story/app:section[5]", "ref", "4"),
+ UpdateAttrib("/document/story/app:section[5]", "single-ref", "4"),
+ UpdateAttrib("/document/story/app:section[6]", "ref", "5"),
+ UpdateAttrib("/document/story/app:section[6]", "single-ref", "5"),
+ UpdateAttrib("/document/story/app:section[7]", "ref", "6"),
+ UpdateAttrib("/document/story/app:section[7]", "single-ref", "6"),
+ UpdateAttrib("/document/story/app:section[8]", "ref", "7"),
+ UpdateAttrib("/document/story/app:section[8]", "single-ref", "7"),
+ UpdateAttrib("/document/story/app:section[9]", "ref", "8"),
+ UpdateAttrib("/document/story/app:section[9]", "single-ref", "8"),
+ UpdateAttrib("/document/story/app:section[10]", "ref", "9"),
+ UpdateAttrib("/document/story/app:section[10]", "single-ref", "9"),
+ UpdateAttrib("/document/story/app:section[11]", "ref", "10"),
+ UpdateAttrib("/document/story/app:section[11]", "single-ref", "10"),
+ UpdateAttrib("/document/story/app:section[12]", "ref", "11"),
+ UpdateAttrib("/document/story/app:section[12]", "single-ref", "11"),
+ UpdateAttrib("/document/story/app:section[14]", "ref", "12"),
+ UpdateAttrib("/document/story/app:section[14]", "single-ref", "12"),
InsertNode(
- '/document/story/app:section[4]',
- '{http://namespaces.shoobx.com/application}term',
- 0),
- InsertAttrib(
- '/document/story/app:section[4]/app:term[1]', 'name',
- 'sign_bonus'),
+ "/document/story/app:section[4]",
+ "{http://namespaces.shoobx.com/application}term",
+ 0,
+ ),
InsertAttrib(
- '/document/story/app:section[4]/app:term[1]', 'set', 'ol'),
- InsertNode('/document/story/app:section[4]', 'para', 1),
+ "/document/story/app:section[4]/app:term[1]", "name", "sign_bonus"
+ ),
+ InsertAttrib("/document/story/app:section[4]/app:term[1]", "set", "ol"),
+ InsertNode("/document/story/app:section[4]", "para", 1),
UpdateTextIn(
- '/document/story/app:section[1]/para[2]/'
- 'app:placeholder[1]',
- 'consectetur'),
+ "/document/story/app:section[1]/para[2]/" "app:placeholder[1]",
+ "consectetur",
+ ),
InsertNode(
- '/document/story/app:section[4]/para[1]',
- '{http://namespaces.shoobx.com/application}ref',
- 0),
+ "/document/story/app:section[4]/para[1]",
+ "{http://namespaces.shoobx.com/application}ref",
+ 0,
+ ),
InsertAttrib(
- '/document/story/app:section[4]/para/app:ref[1]', 'name',
- 'sign'),
+ "/document/story/app:section[4]/para/app:ref[1]", "name", "sign"
+ ),
InsertAttrib(
- '/document/story/app:section[4]/para/app:ref[1]',
- '{http://namespaces.shoobx.com/preview}body',
- '<Ref>'),
- UpdateTextIn(
- '/document/story/app:section[4]/para/app:ref[1]', '3'),
- UpdateTextAfter(
- '/document/story/app:section[4]/para/app:ref[1]', 'eu'),
- InsertNode('/document/story/app:section[4]/para[1]', 'u', 1),
+ "/document/story/app:section[4]/para/app:ref[1]",
+ "{http://namespaces.shoobx.com/preview}body",
+ "<Ref>",
+ ),
+ UpdateTextIn("/document/story/app:section[4]/para/app:ref[1]", "3"),
+ UpdateTextAfter("/document/story/app:section[4]/para/app:ref[1]", "eu"),
+ InsertNode("/document/story/app:section[4]/para[1]", "u", 1),
UpdateTextAfter(
- '/document/story/app:section[4]/para/u[1]',
- 'ntum augue.\n\nAliquam nec tortor diam. Ph'),
+ "/document/story/app:section[4]/para/u[1]",
+ "ntum augue.\n\nAliquam nec tortor diam. Ph",
+ ),
InsertNode(
- '/document/story/app:section[4]/para[1]',
- '{http://namespaces.shoobx.com/application}placeholder',
- 2),
+ "/document/story/app:section[4]/para[1]",
+ "{http://namespaces.shoobx.com/application}placeholder",
+ 2,
+ ),
InsertAttrib(
- '/document/story/app:section[4]/para/app:placeholder[1]',
- 'field',
- 'ol.sign_bonus_include_amt'),
+ "/document/story/app:section[4]/para/app:placeholder[1]",
+ "field",
+ "ol.sign_bonus_include_amt",
+ ),
InsertAttrib(
- '/document/story/app:section[4]/para/app:placeholder[1]',
- 'missing',
- 'Signing Bonus Amount'),
+ "/document/story/app:section[4]/para/app:placeholder[1]",
+ "missing",
+ "Signing Bonus Amount",
+ ),
UpdateTextAfter(
- '/document/story/app:section[4]/para/app:placeholder[1]',
- 'asellus congue accumsan tempor. Donec vel risus se'
+ "/document/story/app:section[4]/para/app:placeholder[1]",
+ "asellus congue accumsan tempor. Donec vel risus se",
),
+ UpdateTextIn("/document/story/app:section[5]/para/app:ref[1]", "4"),
+ UpdateTextIn("/document/story/app:section[6]/para/app:ref[1]", "5"),
+ UpdateTextIn("/document/story/app:section[7]/para/app:ref[1]", "6"),
+ UpdateTextIn("/document/story/app:section[8]/para/app:ref[1]", "7"),
+ UpdateTextIn("/document/story/app:section[9]/para/app:ref[1]", "8"),
+ UpdateTextIn("/document/story/app:section[10]/para/app:ref[1]", "9"),
+ UpdateTextIn("/document/story/app:section[11]/para/app:ref[1]", "10"),
+ UpdateTextIn("/document/story/app:section[12]/para/app:ref[1]", "11"),
+ InsertNode("/document/story/app:section[4]/para/u[1]", "b", 0),
UpdateTextIn(
- '/document/story/app:section[5]/para/app:ref[1]',
- '4'),
- UpdateTextIn(
- '/document/story/app:section[6]/para/app:ref[1]',
- '5'),
- UpdateTextIn(
- '/document/story/app:section[7]/para/app:ref[1]',
- '6'),
- UpdateTextIn(
- '/document/story/app:section[8]/para/app:ref[1]',
- '7'),
- UpdateTextIn(
- '/document/story/app:section[9]/para/app:ref[1]',
- '8'),
- UpdateTextIn(
- '/document/story/app:section[10]/para/app:ref[1]',
- '9'),
- UpdateTextIn(
- '/document/story/app:section[11]/para/app:ref[1]',
- '10'),
- UpdateTextIn(
- '/document/story/app:section[12]/para/app:ref[1]',
- '11'),
- InsertNode('/document/story/app:section[4]/para/u[1]', 'b', 0),
- UpdateTextIn(
- '/document/story/app:section[4]/para/u/b[1]',
- 'ger nec ferme'),
- ]
+ "/document/story/app:section[4]/para/u/b[1]", "ger nec ferme"
+ ),
+ ],
)
def test_sbt_template(self):
here = os.path.split(__file__)[0]
- lfile = os.path.join(here, 'test_data', 'sbt_template.left.xml')
- rfile = os.path.join(here, 'test_data', 'sbt_template.right.xml')
- with open(lfile, 'rt', encoding='utf8') as infile:
+ lfile = os.path.join(here, "test_data", "sbt_template.left.xml")
+ rfile = os.path.join(here, "test_data", "sbt_template.right.xml")
+ with open(lfile, encoding="utf8") as infile:
left = infile.read()
- with open(rfile, 'rt', encoding='utf8') as infile:
+ with open(rfile, encoding="utf8") as infile:
right = infile.read()
result = self._diff(left, right)
- # Most lines get too long and flake8 complains because of this part:
- bm_bm_bm = '/metal:block/metal:block/metal:block'
+ bm_bm_bm = "/metal:block/metal:block/metal:block"
self.assertEqual(
result,
[
InsertNode(
- bm_bm_bm + '[1]',
- '{http://namespaces.shoobx.com/application}section',
- 0),
- InsertAttrib(
- bm_bm_bm + '/app:section[1]',
- 'allowCustom',
- 'False'),
- InsertAttrib(
- bm_bm_bm + '/app:section[1]',
- 'hidden',
- "advisor.payment_type == 'none'"),
- InsertAttrib(
- bm_bm_bm + '/app:section[1]',
- 'name',
- 'payment'),
+ bm_bm_bm + "[1]",
+ "{http://namespaces.shoobx.com/application}section",
+ 0,
+ ),
+ InsertAttrib(bm_bm_bm + "/app:section[1]", "allowCustom", "False"),
InsertAttrib(
- bm_bm_bm + '/app:section[1]',
- 'title',
- 'Payment'),
+ bm_bm_bm + "/app:section[1]",
+ "hidden",
+ "advisor.payment_type == 'none'",
+ ),
+ InsertAttrib(bm_bm_bm + "/app:section[1]", "name", "payment"),
+ InsertAttrib(bm_bm_bm + "/app:section[1]", "title", "Payment"),
InsertNode(
- bm_bm_bm + '/app:section[1]',
- '{http://xml.zope.org/namespaces/tal}if',
- 0),
+ bm_bm_bm + "/app:section[1]",
+ "{http://xml.zope.org/namespaces/tal}if",
+ 0,
+ ),
InsertAttrib(
- bm_bm_bm + '/app:section[1]/tal:if[1]',
- 'condition',
- "python: advisor.payment_type == 'stock_award'"),
+ bm_bm_bm + "/app:section[1]/tal:if[1]",
+ "condition",
+ "python: advisor.payment_type == 'stock_award'",
+ ),
InsertNode(
- bm_bm_bm + '/app:section[1]',
- '{http://xml.zope.org/namespaces/tal}if',
- 1),
+ bm_bm_bm + "/app:section[1]",
+ "{http://xml.zope.org/namespaces/tal}if",
+ 1,
+ ),
InsertAttrib(
- bm_bm_bm + '/app:section[1]/tal:if[2]',
- 'condition',
- "python: advisor.payment_type == 'cash'"),
+ bm_bm_bm + "/app:section[1]/tal:if[2]",
+ "condition",
+ "python: advisor.payment_type == 'cash'",
+ ),
InsertNode(
- bm_bm_bm + '/app:section[1]',
- '{http://xml.zope.org/namespaces/tal}if',
- 2),
+ bm_bm_bm + "/app:section[1]",
+ "{http://xml.zope.org/namespaces/tal}if",
+ 2,
+ ),
InsertAttrib(
- bm_bm_bm + '/app:section[1]/tal:if[3]',
- 'condition',
- "python: advisor.payment_type == 'stock_award_and_cash'"),
- InsertNode(
- bm_bm_bm + '/app:section[1]/tal:if[1]',
- 'para',
- 0),
- UpdateTextIn(
- bm_bm_bm + '/app:section[1]/tal:if[1]/para[1]',
- '\n A '),
- InsertNode(
- bm_bm_bm + '/app:section[1]/tal:if[2]',
- 'para',
- 0),
+ bm_bm_bm + "/app:section[1]/tal:if[3]",
+ "condition",
+ "python: advisor.payment_type == 'stock_award_and_cash'",
+ ),
+ InsertNode(bm_bm_bm + "/app:section[1]/tal:if[1]", "para", 0),
UpdateTextIn(
- bm_bm_bm + '/app:section[1]/tal:if[2]/para[1]',
- '\n More text for diffing purposes\n '),
- InsertNode(
- bm_bm_bm + '/app:section[1]/tal:if[3]',
- 'para',
- 0),
+ bm_bm_bm + "/app:section[1]/tal:if[1]/para[1]", "\n A "
+ ),
+ InsertNode(bm_bm_bm + "/app:section[1]/tal:if[2]", "para", 0),
UpdateTextIn(
- bm_bm_bm + '/app:section[1]/tal:if[3]/para[1]',
- '\n Lorem hipster ipso facto\n '),
- InsertNode(
- bm_bm_bm + '/app:section[1]/tal:if[1]/para[1]',
- 'i',
- 0),
+ bm_bm_bm + "/app:section[1]/tal:if[2]/para[1]",
+ "\n More text for diffing purposes\n ",
+ ),
+ InsertNode(bm_bm_bm + "/app:section[1]/tal:if[3]", "para", 0),
UpdateTextIn(
- bm_bm_bm + '/app:section[1]/tal:if[1]/para/i[1]',
- 'whole'),
+ bm_bm_bm + "/app:section[1]/tal:if[3]/para[1]",
+ "\n Lorem hipster ipso facto\n ",
+ ),
+ InsertNode(bm_bm_bm + "/app:section[1]/tal:if[1]/para[1]", "i", 0),
+ UpdateTextIn(bm_bm_bm + "/app:section[1]/tal:if[1]/para/i[1]", "whole"),
UpdateTextAfter(
- bm_bm_bm + '/app:section[1]/tal:if[1]/para/i[1]',
- ' load of formatted text and '),
- InsertNode(
- bm_bm_bm + '/app:section[1]/tal:if[1]/para[1]',
- 'br',
- 1),
+ bm_bm_bm + "/app:section[1]/tal:if[1]/para/i[1]",
+ " load of formatted text and ",
+ ),
+ InsertNode(bm_bm_bm + "/app:section[1]/tal:if[1]/para[1]", "br", 1),
UpdateTextAfter(
- bm_bm_bm + '/app:section[1]/tal:if[1]/para/br[1]',
- ' other stuff.\n '),
- DeleteNode(
- bm_bm_bm + '/app:section[2]/tal:if/para/b[1]'),
- DeleteNode(
- bm_bm_bm + '/app:section[2]/tal:if/para[1]'),
- DeleteNode(
- bm_bm_bm + '/app:section[2]/tal:if[1]'),
- DeleteNode(
- bm_bm_bm + '/app:section[2]')
- ]
+ bm_bm_bm + "/app:section[1]/tal:if[1]/para/br[1]",
+ " other stuff.\n ",
+ ),
+ DeleteNode(bm_bm_bm + "/app:section[2]/tal:if/para/b[1]"),
+ DeleteNode(bm_bm_bm + "/app:section[2]/tal:if/para[1]"),
+ DeleteNode(bm_bm_bm + "/app:section[2]/tal:if[1]"),
+ DeleteNode(bm_bm_bm + "/app:section[2]"),
+ ],
)
def test_namespace(self):
# Test changing nodes and attributes with namespaces
- left = u"""<document xmlns:app="someuri">
+ left = """<document xmlns:app="someuri">
<story app:foo="FirstPage">
<app:section>
<foo:para xmlns:foo="otheruri">Lorem ipsum dolor sit amet,
@@ -1407,7 +1378,7 @@ class DiffTests(unittest.TestCase):
</document>
"""
- right = u"""<document xmlns:app="someuri">
+ right = """<document xmlns:app="someuri">
<story app:foo="FirstPage">
<app:section>
<app:para>Lorem ipsum dolor sit amet,
@@ -1433,17 +1404,17 @@ class DiffTests(unittest.TestCase):
self.assertEqual(
result,
[
- RenameNode(
- '/document/story/app:section/foo:para[1]',
- '{someuri}para'),
+ RenameNode("/document/story/app:section/foo:para[1]", "{someuri}para"),
InsertAttrib(
- '/document/story/app:section/app:para[3]',
- '{someuri}attrib', 'value'),
- ]
+ "/document/story/app:section/app:para[3]",
+ "{someuri}attrib",
+ "value",
+ ),
+ ],
)
def test_multiple_tag_deletes(self):
- left = u"""<document title="delte-node-ul">
+ left = """<document title="delte-node-ul">
<story id="id">
<ul>
@@ -1455,7 +1426,7 @@ class DiffTests(unittest.TestCase):
</story>
</document>"""
- right = u"""<document title="delte-node-ul">
+ right = """<document title="delte-node-ul">
<story id="id">
</story>
</document>"""
@@ -1463,23 +1434,21 @@ class DiffTests(unittest.TestCase):
result = self._diff(left, right)
self.assertEqual(
result,
- [UpdateTextIn('/document/story[1]', '\n '),
- DeleteNode('/document/story/ul/li[3]'),
- DeleteNode('/document/story/ul/li[2]'),
- DeleteNode('/document/story/ul/li[1]'),
- DeleteNode('/document/story/ul[1]'),
- ]
+ [
+ UpdateTextIn("/document/story[1]", "\n "),
+ DeleteNode("/document/story/ul/li[3]"),
+ DeleteNode("/document/story/ul/li[2]"),
+ DeleteNode("/document/story/ul/li[1]"),
+ DeleteNode("/document/story/ul[1]"),
+ ],
)
def test_insert_comment(self):
- left = u"<doc><body>Something</body></doc>"
- right = u"<doc><!-- New comment! --><body>Something</body></doc>"
+ left = "<doc><body>Something</body></doc>"
+ right = "<doc><!-- New comment! --><body>Something</body></doc>"
result = self._diff(left, right)
- self.assertEqual(
- result,
- [InsertComment('/doc[1]', 0, ' New comment! ')]
- )
+ self.assertEqual(result, [InsertComment("/doc[1]", 0, " New comment! ")])
def test_issue_21_default_namespaces(self):
# When you have a default namespace you get "*" instead of the
@@ -1488,4 +1457,4 @@ class DiffTests(unittest.TestCase):
left = '<tag xmlns="ns">old</tag>'
right = '<tag xmlns="ns">new</tag>'
result = self._diff(left, right)
- self.assertEqual(result[0].node, '/*[1]')
+ self.assertEqual(result[0].node, "/*[1]")
Index: xmldiff-2.4/tests/test_formatting.py
===================================================================
--- xmldiff-2.4.orig/tests/test_formatting.py
+++ xmldiff-2.4/tests/test_formatting.py
@@ -1,4 +1,3 @@
-# -*- coding: UTF-8 -*-
import os
import sys
import unittest
@@ -8,89 +7,80 @@ from xmldiff import formatting, main, ac
from .testing import generate_filebased_cases
-START = u'<document xmlns:diff="http://namespaces.shoobx.com/diff"><node'
-END = u'</node></document>'
+START = '<document xmlns:diff="http://namespaces.shoobx.com/diff"><node'
+END = "</node></document>"
class PlaceholderMakerTests(unittest.TestCase):
-
def test_get_placeholder(self):
replacer = formatting.PlaceholderMaker()
# Get a placeholder:
- ph = replacer.get_placeholder(
- etree.Element('tag'), formatting.T_OPEN, None)
- self.assertEqual(ph, u'\ue005')
+ ph = replacer.get_placeholder(etree.Element("tag"), formatting.T_OPEN, None)
+ self.assertEqual(ph, "\ue005")
# Do it again:
- ph = replacer.get_placeholder(
- etree.Element('tag'), formatting.T_OPEN, None)
- self.assertEqual(ph, u'\ue005')
+ ph = replacer.get_placeholder(etree.Element("tag"), formatting.T_OPEN, None)
+ self.assertEqual(ph, "\ue005")
# Get another one
- ph = replacer.get_placeholder(
- etree.Element('tag'), formatting.T_CLOSE, ph)
- self.assertEqual(ph, u'\ue006')
+ ph = replacer.get_placeholder(etree.Element("tag"), formatting.T_CLOSE, ph)
+ self.assertEqual(ph, "\ue006")
def test_do_element(self):
- replacer = formatting.PlaceholderMaker(['p'], ['b'])
+ replacer = formatting.PlaceholderMaker(["p"], ["b"])
# Formatting tags get replaced, and the content remains
- text = u'<p>This is a tag with <b>formatted</b> text.</p>'
+ text = "<p>This is a tag with <b>formatted</b> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
self.assertEqual(
etree.tounicode(element),
- u'<p>This is a tag with \ue006formatted\ue005 text.</p>')
+ "<p>This is a tag with \ue006formatted\ue005 text.</p>",
+ )
replacer.undo_element(element)
self.assertEqual(etree.tounicode(element), text)
# Non formatting tags get replaced with content
- text = u'<p>This is a tag with <foo>formatted</foo> text.</p>'
+ text = "<p>This is a tag with <foo>formatted</foo> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
result = etree.tounicode(element)
- self.assertEqual(
- result,
- u'<p>This is a tag with \ue007 text.</p>')
+ self.assertEqual(result, "<p>This is a tag with \ue007 text.</p>")
# Single formatting tags still get two placeholders.
- text = u'<p>This is a <b/> with <foo/> text.</p>'
+ text = "<p>This is a <b/> with <foo/> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
result = etree.tounicode(element)
- self.assertEqual(
- result,
- u'<p>This is a \ue009\ue008 with \ue00a text.</p>')
+ self.assertEqual(result, "<p>This is a \ue009\ue008 with \ue00a text.</p>")
def test_do_undo_element(self):
- replacer = formatting.PlaceholderMaker(['p'], ['b'])
+ replacer = formatting.PlaceholderMaker(["p"], ["b"])
# Formatting tags get replaced, and the content remains
- text = u'<p>This <is/> a <f>tag</f> with <b>formatted</b> text.</p>'
+ text = "<p>This <is/> a <f>tag</f> with <b>formatted</b> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
self.assertEqual(
- element.text,
- u'This \ue005 a \ue006 with \ue008formatted'
- u'\ue007 text.')
+ element.text, "This \ue005 a \ue006 with \ue008formatted" "\ue007 text."
+ )
replacer.undo_element(element)
result = etree.tounicode(element)
self.assertEqual(result, text)
def test_do_undo_element_double_format(self):
- replacer = formatting.PlaceholderMaker(['p'], ['b', 'u'])
+ replacer = formatting.PlaceholderMaker(["p"], ["b", "u"])
# Formatting tags get replaced, and the content remains
- text = u'<p>This is <u>doubly <b>formatted</b></u> text.</p>'
+ text = "<p>This is <u>doubly <b>formatted</b></u> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
self.assertEqual(
- element.text,
- u'This is \ue006doubly \ue008formatted\ue007'
- u'\ue005 text.')
+ element.text, "This is \ue006doubly \ue008formatted\ue007" "\ue005 text."
+ )
replacer.undo_element(element)
result = etree.tounicode(element)
@@ -98,7 +88,7 @@ class PlaceholderMakerTests(unittest.Tes
def test_rml_bug(self):
etree.register_namespace(formatting.DIFF_PREFIX, formatting.DIFF_NS)
- before_diff = u"""<document xmlns:diff="http://namespaces.shoobx.com/diff">
+ before_diff = """<document xmlns:diff="http://namespaces.shoobx.com/diff">
<section>
<para>
<ref>4</ref>.
@@ -109,9 +99,10 @@ class PlaceholderMakerTests(unittest.Tes
</document>"""
tree = etree.fromstring(before_diff)
replacer = formatting.PlaceholderMaker(
- text_tags=('para',), formatting_tags=('b', 'u', 'i',))
+ text_tags=("para",), formatting_tags=("b", "u", "i",)
+ )
replacer.do_tree(tree)
- after_diff = u"""<document xmlns:diff="http://namespaces.shoobx.com/diff">
+ after_diff = """<document xmlns:diff="http://namespaces.shoobx.com/diff">
<section>
<para>
<insert>\ue005</insert>.
@@ -122,15 +113,13 @@ class PlaceholderMakerTests(unittest.Tes
</document>"""
# The diff formatting will find some text to insert.
- delete_attrib = u'{%s}delete-format' % formatting.DIFF_NS
- replacer.placeholder2tag[u'\ue006'
- ].element.attrib[delete_attrib] = ''
- replacer.placeholder2tag[u'\ue007'
- ].element.attrib[delete_attrib] = ''
+ delete_attrib = "{%s}delete-format" % formatting.DIFF_NS
+ replacer.placeholder2tag["\ue006"].element.attrib[delete_attrib] = ""
+ replacer.placeholder2tag["\ue007"].element.attrib[delete_attrib] = ""
tree = etree.fromstring(after_diff)
replacer.undo_tree(tree)
result = etree.tounicode(tree)
- expected = u"""<document xmlns:diff="http://namespaces.shoobx.com/diff">
+ expected = """<document xmlns:diff="http://namespaces.shoobx.com/diff">
<section>
<para>
<insert><ref>4</ref></insert>.
@@ -150,18 +139,17 @@ class PlaceholderMakerTests(unittest.Tes
# This is the last character of the Private use area
formatting.PLACEHOLDER_START = 0xF8FF
- replacer = formatting.PlaceholderMaker(['p'], ['b'])
+ replacer = formatting.PlaceholderMaker(["p"], ["b"])
# Formatting tags get replaced, and the content remains
- text = u'<p>This <is/> a <f>tag</f> with <b>some</b> text.</p>'
+ text = "<p>This <is/> a <f>tag</f> with <b>some</b> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
#
self.assertEqual(
- element.text,
- u'This \uf904 a \uf905 with \uf907some'
- u'\uf906 text.')
+ element.text, "This \uf904 a \uf905 with \uf907some" "\uf906 text."
+ )
try:
# If this is a wide build, also test what happens if we
@@ -169,18 +157,19 @@ class PlaceholderMakerTests(unittest.Tes
# (On narrow builds this will give an error)
formatting.PLACEHOLDER_START = 0xFFFF
- replacer = formatting.PlaceholderMaker(['p'], ['b'])
+ replacer = formatting.PlaceholderMaker(["p"], ["b"])
# Formatting tags get replaced, and the content remains
- text = u'<p>This <is/> a <f>tag</f> with <b>some</b> text.</p>'
+ text = "<p>This <is/> a <f>tag</f> with <b>some</b> text.</p>"
element = etree.fromstring(text)
replacer.do_element(element)
# This should raise an error on a narrow build
self.assertEqual(
element.text,
- u'This \U00010004 a \U00010005 with \U00010007some'
- u'\U00010006 text.')
+ "This \U00010004 a \U00010005 with \U00010007some"
+ "\U00010006 text.",
+ )
except ValueError:
if sys.maxunicode > 0x10000:
# This is a wide build, we should NOT get an error
@@ -192,229 +181,224 @@ class PlaceholderMakerTests(unittest.Tes
class XMLFormatTests(unittest.TestCase):
-
def _format_test(self, left, action, expected):
formatter = formatting.XMLFormatter(pretty_print=False)
result = formatter.format([action], etree.fromstring(left))
self.assertEqual(result, expected)
def test_incorrect_xpaths(self):
- left = u'<document><node a="v"/><node>Text</node></document>'
- expected = START + u' diff:delete-attr="a">Text' + END
+ left = '<document><node a="v"/><node>Text</node></document>'
+ expected = START + ' diff:delete-attr="a">Text' + END
with self.assertRaises(ValueError):
- action = actions.DeleteAttrib('/document/node', 'a')
+ action = actions.DeleteAttrib("/document/node", "a")
self._format_test(left, action, expected)
with self.assertRaises(ValueError):
- action = actions.DeleteAttrib('/document/ummagumma', 'a')
+ action = actions.DeleteAttrib("/document/ummagumma", "a")
self._format_test(left, action, expected)
def test_del_attr(self):
- left = u'<document><node a="v">Text</node></document>'
- action = actions.DeleteAttrib('/document/node', 'a')
- expected = START + u' diff:delete-attr="a">Text' + END
+ left = '<document><node a="v">Text</node></document>'
+ action = actions.DeleteAttrib("/document/node", "a")
+ expected = START + ' diff:delete-attr="a">Text' + END
self._format_test(left, action, expected)
def test_del_node(self):
- left = u'<document><node attr="val">Text</node></document>'
- action = actions.DeleteNode('/document/node')
- expected = START + u' attr="val" diff:delete="">Text' + END
+ left = '<document><node attr="val">Text</node></document>'
+ action = actions.DeleteNode("/document/node")
+ expected = START + ' attr="val" diff:delete="">Text' + END
self._format_test(left, action, expected)
def test_del_text(self):
- left = u'<document><node attr="val">Text</node></document>'
- action = actions.UpdateTextIn('/document/node', None)
- expected = START + u' attr="val"><diff:delete>Text</diff:delete>' + END
+ left = '<document><node attr="val">Text</node></document>'
+ action = actions.UpdateTextIn("/document/node", None)
+ expected = START + ' attr="val"><diff:delete>Text</diff:delete>' + END
self._format_test(left, action, expected)
def test_insert_attr(self):
- left = u'<document><node>We need more text</node></document>'
- action = actions.InsertAttrib('/document/node', 'attr', 'val')
- expected = START + u' attr="val" diff:add-attr="attr">'\
- u'We need more text' + END
+ left = "<document><node>We need more text</node></document>"
+ action = actions.InsertAttrib("/document/node", "attr", "val")
+ expected = START + ' attr="val" diff:add-attr="attr">' "We need more text" + END
self._format_test(left, action, expected)
def test_insert_node(self):
- left = u'<document></document>'
- action = actions.InsertNode('/document', 'node', 0)
- expected = START + u' diff:insert=""/></document>'
+ left = "<document></document>"
+ action = actions.InsertNode("/document", "node", 0)
+ expected = START + ' diff:insert=""/></document>'
self._format_test(left, action, expected)
def test_move_attr(self):
# The library currently only uses move attr for when attributes are
# renamed:
- left = u'<document><node attr="val">Text</node></document>'
- action = actions.RenameAttrib('/document/node', 'attr', 'bottr')
- expected = START + u' bottr="val" diff:rename-attr="attr:bottr"'\
- u'>Text' + END
+ left = '<document><node attr="val">Text</node></document>'
+ action = actions.RenameAttrib("/document/node", "attr", "bottr")
+ expected = START + ' bottr="val" diff:rename-attr="attr:bottr"' ">Text" + END
self._format_test(left, action, expected)
def test_move_node(self):
# Move 1 down
- left = u'<document><node id="1" /><node id="2" /></document>'
- action = actions.MoveNode('/document/node[1]', '/document', 1)
- expected = START + u' id="1" diff:delete=""/><node id="2"/><node '\
- u'id="1" diff:insert=""/></document>'
+ left = '<document><node id="1" /><node id="2" /></document>'
+ action = actions.MoveNode("/document/node[1]", "/document", 1)
+ expected = (
+ START + ' id="1" diff:delete=""/><node id="2"/><node '
+ 'id="1" diff:insert=""/></document>'
+ )
self._format_test(left, action, expected)
# Move 2 up (same result, different diff)
- left = u'<document><node id="1" /><node id="2" /></document>'
- action = actions.MoveNode('/document/node[2]', '/document', 0)
- expected = START + u' id="2" diff:insert=""/><node id="1"/><node '\
- u'id="2" diff:delete=""/></document>'
+ left = '<document><node id="1" /><node id="2" /></document>'
+ action = actions.MoveNode("/document/node[2]", "/document", 0)
+ expected = (
+ START + ' id="2" diff:insert=""/><node id="1"/><node '
+ 'id="2" diff:delete=""/></document>'
+ )
self._format_test(left, action, expected)
def test_rename_node(self):
- left = u'<document><node><para>Content</para>Tail</node></document>'
- action = actions.RenameNode('/document/node[1]/para[1]', 'newtag')
- expected = START + u'><newtag diff:rename="para">Content'\
- '</newtag>Tail' + END
+ left = "<document><node><para>Content</para>Tail</node></document>"
+ action = actions.RenameNode("/document/node[1]/para[1]", "newtag")
+ expected = START + '><newtag diff:rename="para">Content' "</newtag>Tail" + END
self._format_test(left, action, expected)
def test_update_attr(self):
- left = u'<document><node attr="val"/></document>'
- action = actions.UpdateAttrib('/document/node', 'attr', 'newval')
- expected = START + u' attr="newval" diff:update-attr="attr:val"/>'\
- u'</document>'
+ left = '<document><node attr="val"/></document>'
+ action = actions.UpdateAttrib("/document/node", "attr", "newval")
+ expected = START + ' attr="newval" diff:update-attr="attr:val"/>' "</document>"
self._format_test(left, action, expected)
def test_update_text_in(self):
- left = u'<document><node attr="val"/></document>'
- action = actions.UpdateTextIn('/document/node', 'Text')
- expected = START + u' attr="val"><diff:insert>Text</diff:insert>' + END
+ left = '<document><node attr="val"/></document>'
+ action = actions.UpdateTextIn("/document/node", "Text")
+ expected = START + ' attr="val"><diff:insert>Text</diff:insert>' + END
self._format_test(left, action, expected)
- left = u'<document><node>This is a bit of text, right' + END
- action = actions.UpdateTextIn('/document/node',
- 'Also a bit of text, rick')
- expected = START + u'><diff:delete>This is</diff:delete><diff:insert>'\
- u'Also</diff:insert> a bit of text, ri<diff:delete>ght'\
- u'</diff:delete><diff:insert>ck</diff:insert>' + END
+ left = "<document><node>This is a bit of text, right" + END
+ action = actions.UpdateTextIn("/document/node", "Also a bit of text, rick")
+ expected = (
+ START + "><diff:delete>This is</diff:delete><diff:insert>"
+ "Also</diff:insert> a bit of text, ri<diff:delete>ght"
+ "</diff:delete><diff:insert>ck</diff:insert>" + END
+ )
self._format_test(left, action, expected)
def test_update_text_after_1(self):
- left = u'<document><node/><node/></document>'
- action = actions.UpdateTextAfter('/document/node[1]', 'Text')
- expected = START + u'/><diff:insert>Text</diff:insert>'\
- u'<node/></document>'
+ left = "<document><node/><node/></document>"
+ action = actions.UpdateTextAfter("/document/node[1]", "Text")
+ expected = START + "/><diff:insert>Text</diff:insert>" "<node/></document>"
self._format_test(left, action, expected)
def test_update_text_after_2(self):
- left = u'<document><node/>This is a bit of text, right</document>'
- action = actions.UpdateTextAfter('/document/node',
- 'Also a bit of text, rick')
- expected = START + u'/><diff:delete>This is</diff:delete>'\
- u'<diff:insert>Also</diff:insert> a bit of text, ri<diff:delete>'\
- u'ght</diff:delete><diff:insert>ck</diff:insert></document>'
+ left = "<document><node/>This is a bit of text, right</document>"
+ action = actions.UpdateTextAfter("/document/node", "Also a bit of text, rick")
+ expected = (
+ START + "/><diff:delete>This is</diff:delete>"
+ "<diff:insert>Also</diff:insert> a bit of text, ri<diff:delete>"
+ "ght</diff:delete><diff:insert>ck</diff:insert></document>"
+ )
self._format_test(left, action, expected)
class DiffFormatTests(unittest.TestCase):
-
def _format_test(self, action, expected):
formatter = formatting.DiffFormatter()
result = formatter.format([action], None)
self.assertEqual(result, expected)
def test_del_attr(self):
- action = actions.DeleteAttrib('/document/node', 'a')
- expected = '[delete-attribute, /document/node, a]'
+ action = actions.DeleteAttrib("/document/node", "a")
+ expected = "[delete-attribute, /document/node, a]"
self._format_test(action, expected)
def test_del_node(self):
- action = actions.DeleteNode('/document/node')
- expected = '[delete, /document/node]'
+ action = actions.DeleteNode("/document/node")
+ expected = "[delete, /document/node]"
self._format_test(action, expected)
def test_del_text(self):
- action = actions.UpdateTextIn('/document/node', None)
- expected = '[update-text, /document/node, null]'
+ action = actions.UpdateTextIn("/document/node", None)
+ expected = "[update-text, /document/node, null]"
self._format_test(action, expected)
def test_insert_attr(self):
- action = actions.InsertAttrib('/document/node', 'attr', 'val')
+ action = actions.InsertAttrib("/document/node", "attr", "val")
expected = '[insert-attribute, /document/node, attr, "val"]'
self._format_test(action, expected)
def test_insert_node(self):
- action = actions.InsertNode('/document', 'node', 0)
- expected = '[insert, /document, node, 0]'
+ action = actions.InsertNode("/document", "node", 0)
+ expected = "[insert, /document, node, 0]"
self._format_test(action, expected)
def test_rename_attr(self):
- action = actions.RenameAttrib('/document/node', 'attr', 'bottr')
- expected = '[rename-attribute, /document/node, attr, bottr]'
+ action = actions.RenameAttrib("/document/node", "attr", "bottr")
+ expected = "[rename-attribute, /document/node, attr, bottr]"
self._format_test(action, expected)
def test_move_node(self):
# Move 1 down
- action = actions.MoveNode('/document/node[1]', '/document', 1)
- expected = '[move, /document/node[1], /document, 1]'
+ action = actions.MoveNode("/document/node[1]", "/document", 1)
+ expected = "[move, /document/node[1], /document, 1]"
self._format_test(action, expected)
# Move 2 up (same result, different diff)
- action = actions.MoveNode('/document/node[2]', '/document', 0)
- expected = '[move, /document/node[2], /document, 0]'
+ action = actions.MoveNode("/document/node[2]", "/document", 0)
+ expected = "[move, /document/node[2], /document, 0]"
self._format_test(action, expected)
def test_rename_node(self):
# Move 1 down
- action = actions.RenameNode('/document/node[1]', 'newtag')
- expected = '[rename, /document/node[1], newtag]'
+ action = actions.RenameNode("/document/node[1]", "newtag")
+ expected = "[rename, /document/node[1], newtag]"
self._format_test(action, expected)
# Move 2 up (same result, different diff)
- action = actions.MoveNode('/document/node[2]', '/document', 0)
- expected = '[move, /document/node[2], /document, 0]'
+ action = actions.MoveNode("/document/node[2]", "/document", 0)
+ expected = "[move, /document/node[2], /document, 0]"
self._format_test(action, expected)
def test_update_attr(self):
- action = actions.UpdateAttrib('/document/node', 'attr', 'newval')
+ action = actions.UpdateAttrib("/document/node", "attr", "newval")
expected = '[update-attribute, /document/node, attr, "newval"]'
self._format_test(action, expected)
def test_update_text_in(self):
- action = actions.UpdateTextIn('/document/node', 'Text')
+ action = actions.UpdateTextIn("/document/node", "Text")
expected = '[update-text, /document/node, "Text"]'
self._format_test(action, expected)
- action = actions.UpdateTextIn('/document/node',
- 'Also a bit of text, "rick"')
- expected = '[update-text, /document/node, '\
- u'"Also a bit of text, \\"rick\\""]'
+ action = actions.UpdateTextIn("/document/node", 'Also a bit of text, "rick"')
+ expected = "[update-text, /document/node, " '"Also a bit of text, \\"rick\\""]'
self._format_test(action, expected)
def test_update_text_after_1(self):
- action = actions.UpdateTextAfter('/document/node[1]', 'Text')
+ action = actions.UpdateTextAfter("/document/node[1]", "Text")
expected = '[update-text-after, /document/node[1], "Text"]'
self._format_test(action, expected)
def test_update_text_after_2(self):
- action = actions.UpdateTextAfter('/document/node',
- 'Also a bit of text, rick')
- expected = '[update-text-after, /document/node, '\
- u'"Also a bit of text, rick"]'
+ action = actions.UpdateTextAfter("/document/node", "Also a bit of text, rick")
+ expected = "[update-text-after, /document/node, " '"Also a bit of text, rick"]'
self._format_test(action, expected)
def test_insert_comment(self):
- action = actions.InsertComment('/document/node', 2, 'Commentary')
+ action = actions.InsertComment("/document/node", 2, "Commentary")
expected = '[insert-comment, /document/node, 2, "Commentary"]'
self._format_test(action, expected)
@@ -430,97 +414,95 @@ class XmlDiffFormatTests(unittest.TestCa
self.assertEqual(result, expected)
def test_del_attr(self):
- action = actions.DeleteAttrib('/document/node', 'a')
- expected = '[remove, /document/node/@a]'
+ action = actions.DeleteAttrib("/document/node", "a")
+ expected = "[remove, /document/node/@a]"
self._format_test(action, expected)
def test_del_node(self):
- action = actions.DeleteNode('/document/node')
- expected = '[remove, /document/node]'
+ action = actions.DeleteNode("/document/node")
+ expected = "[remove, /document/node]"
self._format_test(action, expected)
def test_del_text(self):
- action = actions.UpdateTextIn('/document/node', None)
- expected = '[update, /document/node/text()[1], null]'
+ action = actions.UpdateTextIn("/document/node", None)
+ expected = "[update, /document/node/text()[1], null]"
self._format_test(action, expected)
def test_insert_attr(self):
- action = actions.InsertAttrib('/document/node', 'attr', 'val')
- expected = '[insert, /document/node, \n<@attr>\nval\n</@attr>]'
+ action = actions.InsertAttrib("/document/node", "attr", "val")
+ expected = "[insert, /document/node, \n<@attr>\nval\n</@attr>]"
self._format_test(action, expected)
def test_insert_node(self):
- action = actions.InsertNode('/document', 'node', 0)
- expected = '[insert-first, /document, \n<node/>]'
+ action = actions.InsertNode("/document", "node", 0)
+ expected = "[insert-first, /document, \n<node/>]"
self._format_test(action, expected)
def test_rename_node(self):
# Move 1 down
- action = actions.RenameNode('/document/node[1]', 'newtag')
- expected = '[rename, /document/node[1], newtag]'
+ action = actions.RenameNode("/document/node[1]", "newtag")
+ expected = "[rename, /document/node[1], newtag]"
self._format_test(action, expected)
# Move 2 up (same result, different diff)
- action = actions.MoveNode('/document/node[2]', '/document', 0)
- expected = '[move-first, /document/node[2], /document]'
+ action = actions.MoveNode("/document/node[2]", "/document", 0)
+ expected = "[move-first, /document/node[2], /document]"
self._format_test(action, expected)
def test_update_attr(self):
- action = actions.UpdateAttrib('/document/node', 'attr', 'newval')
+ action = actions.UpdateAttrib("/document/node", "attr", "newval")
expected = '[update, /document/node/@attr, "newval"]'
self._format_test(action, expected)
def test_update_text_in(self):
- action = actions.UpdateTextIn('/document/node', 'Text')
+ action = actions.UpdateTextIn("/document/node", "Text")
expected = '[update, /document/node/text()[1], "Text"]'
self._format_test(action, expected)
- action = actions.UpdateTextIn('/document/node',
- 'Also a bit of text, "rick"')
- expected = '[update, /document/node/text()[1], '\
- u'"Also a bit of text, \\"rick\\""]'
+ action = actions.UpdateTextIn("/document/node", 'Also a bit of text, "rick"')
+ expected = (
+ "[update, /document/node/text()[1], " '"Also a bit of text, \\"rick\\""]'
+ )
self._format_test(action, expected)
def test_update_text_after_1(self):
- action = actions.UpdateTextAfter('/document/node[1]', 'Text')
+ action = actions.UpdateTextAfter("/document/node[1]", "Text")
expected = '[update, /document/node[1]/text()[2], "Text"]'
self._format_test(action, expected)
def test_update_text_after_2(self):
- action = actions.UpdateTextAfter('/document/node',
- 'Also a bit of text, rick')
- expected = '[update, /document/node/text()[2], '\
- u'"Also a bit of text, rick"]'
+ action = actions.UpdateTextAfter("/document/node", "Also a bit of text, rick")
+ expected = "[update, /document/node/text()[2], " '"Also a bit of text, rick"]'
self._format_test(action, expected)
def test_all_actions(self):
here = os.path.split(__file__)[0]
- lfile = os.path.join(here, 'test_data', 'all_actions.left.xml')
- rfile = os.path.join(here, 'test_data', 'all_actions.right.xml')
+ lfile = os.path.join(here, "test_data", "all_actions.left.xml")
+ rfile = os.path.join(here, "test_data", "all_actions.right.xml")
formatter = formatting.XmlDiffFormatter()
result = main.diff_files(lfile, rfile, formatter=formatter)
expected = (
- u'[move-after, /document/node[2], /document/tag[1]]\n'
- u'[insert-comment, /document[1], 0, Insert a new comment ]\n'
- u'[update, /document/node[1]/@name, "was updated"]\n'
- u'[remove, /document/node[1]/@attribute]\n'
- u'[insert, /document/node[1], \n'
- u'<@newtribute>\n'
- u'renamed\n'
- u'</@newtribute>]\n'
- u'[insert, /document/node[1], \n'
- u'<@this>\n'
- u'is new\n'
- u'</@this>]\n'
- u'[remove, /document/node[1]/@attr]\n'
- u'[update, /document/node[1]/text()[1], "\\n Modified\\n "]\n'
- u'[update, /document/node[1]/text()[2], "\\n '
- u'New tail content\\n "]\n'
- u'[rename, /document/node[2], nod]\n'
- u'[insert-after, /document/tail[1], \n'
- u'<new/>]\n'
- u'[remove, /document/tail[1]]'
+ "[move-after, /document/node[2], /document/tag[1]]\n"
+ "[insert-comment, /document[1], 0, Insert a new comment ]\n"
+ '[update, /document/node[1]/@name, "was updated"]\n'
+ "[remove, /document/node[1]/@attribute]\n"
+ "[insert, /document/node[1], \n"
+ "<@newtribute>\n"
+ "renamed\n"
+ "</@newtribute>]\n"
+ "[insert, /document/node[1], \n"
+ "<@this>\n"
+ "is new\n"
+ "</@this>]\n"
+ "[remove, /document/node[1]/@attr]\n"
+ '[update, /document/node[1]/text()[1], "\\n Modified\\n "]\n'
+ '[update, /document/node[1]/text()[2], "\\n '
+ 'New tail content\\n "]\n'
+ "[rename, /document/node[2], nod]\n"
+ "[insert-after, /document/tail[1], \n"
+ "<new/>]\n"
+ "[remove, /document/tail[1]]"
)
self.assertEqual(result, expected)
@@ -537,12 +519,14 @@ class FormatterFileTests(unittest.TestCa
class XMLFormatterFileTests(FormatterFileTests):
# The XMLFormatter has no text or formatting tags, so
- formatter = formatting.XMLFormatter(pretty_print=False,
- normalize=formatting.WS_TEXT)
+ formatter = formatting.XMLFormatter(
+ pretty_print=False, normalize=formatting.WS_TEXT
+ )
# Also test the bits that handle text tags:
+
class HTMLFormatterFileTests(FormatterFileTests):
# We use a few tags for the placeholder tests.
@@ -551,15 +535,27 @@ class HTMLFormatterFileTests(FormatterFi
formatter = formatting.XMLFormatter(
normalize=formatting.WS_BOTH,
pretty_print=True,
- text_tags=('p', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'li'),
- formatting_tags=('b', 'u', 'i', 'strike', 'em', 'super',
- 'sup', 'sub', 'link', 'a', 'span'))
+ text_tags=("p", "h1", "h2", "h3", "h4", "h5", "h6", "li"),
+ formatting_tags=(
+ "b",
+ "u",
+ "i",
+ "strike",
+ "em",
+ "super",
+ "sup",
+ "sub",
+ "link",
+ "a",
+ "span",
+ ),
+ )
# Add tests that use no placeholder replacement (ie plain XML)
-data_dir = os.path.join(os.path.dirname(__file__), 'test_data')
+data_dir = os.path.join(os.path.dirname(__file__), "test_data")
generate_filebased_cases(data_dir, XMLFormatterFileTests)
# Add tests that use placeholder replacement (ie HTML)
-data_dir = os.path.join(os.path.dirname(__file__), 'test_data')
-generate_filebased_cases(data_dir, HTMLFormatterFileTests, suffix='html')
+data_dir = os.path.join(os.path.dirname(__file__), "test_data")
+generate_filebased_cases(data_dir, HTMLFormatterFileTests, suffix="html")
Index: xmldiff-2.4/tests/test_main.py
===================================================================
--- xmldiff-2.4.orig/tests/test_main.py
+++ xmldiff-2.4/tests/test_main.py
@@ -7,44 +7,42 @@ from lxml import etree
from xmldiff import main, formatting
CURDIR = os.path.split(__file__)[0]
-LEFT_FILE = os.path.join(CURDIR, 'test_data', 'rmldoc.left.xml')
-RIGHT_FILE = os.path.join(CURDIR, 'test_data', 'rmldoc.right.xml')
-EXPECTED_FILE = os.path.join(CURDIR, 'test_data', 'rmldoc.expected.xml')
+LEFT_FILE = os.path.join(CURDIR, "test_data", "rmldoc.left.xml")
+RIGHT_FILE = os.path.join(CURDIR, "test_data", "rmldoc.right.xml")
+EXPECTED_FILE = os.path.join(CURDIR, "test_data", "rmldoc.expected.xml")
class MainAPITests(unittest.TestCase):
-
def test_api_diff_files(self):
# diff_files can take filenames
result1 = main.diff_files(LEFT_FILE, RIGHT_FILE)
# Or open file streams:
- with open(LEFT_FILE, 'rb') as linfile:
- with open(RIGHT_FILE, 'rb') as rinfile:
+ with open(LEFT_FILE, "rb") as linfile:
+ with open(RIGHT_FILE, "rb") as rinfile:
result2 = main.diff_files(linfile, rinfile)
self.assertEqual(result1, result2)
# Give something else, and it fails:
with self.assertRaises(IOError):
- main.diff_files('<xml1/>', '<xml2/>')
+ main.diff_files("<xml1/>", "<xml2/>")
def test_api_diff_texts(self):
# diff_text can take bytes
- with open(LEFT_FILE, 'rb') as linfile:
- with open(RIGHT_FILE, 'rb') as rinfile:
+ with open(LEFT_FILE, "rb") as linfile:
+ with open(RIGHT_FILE, "rb") as rinfile:
left = linfile.read()
right = rinfile.read()
result1 = main.diff_texts(left, right)
# And unicode
- result2 = main.diff_texts(left.decode('utf8'),
- right.decode('utf8'))
+ result2 = main.diff_texts(left.decode("utf8"), right.decode("utf8"))
self.assertEqual(result1, result2)
- with open(LEFT_FILE, 'rb') as infile:
- with open(RIGHT_FILE, 'rb') as infile:
+ with open(LEFT_FILE, "rb") as infile:
+ with open(RIGHT_FILE, "rb") as infile:
# Give something else, and it fails:
with self.assertRaises(ValueError):
main.diff_texts(infile, infile)
@@ -72,7 +70,6 @@ class MainAPITests(unittest.TestCase):
class MainCLITests(unittest.TestCase):
-
def call_run(self, args, command=main.diff_command):
output = six.StringIO()
errors = six.StringIO()
@@ -97,75 +94,75 @@ class MainCLITests(unittest.TestCase):
def test_diff_cli_simple(self):
curdir = os.path.dirname(__file__)
- filepath = os.path.join(curdir, 'test_data')
- file1 = os.path.join(filepath, 'insert-node.left.html')
- file2 = os.path.join(filepath, 'insert-node.right.html')
+ filepath = os.path.join(curdir, "test_data")
+ file1 = os.path.join(filepath, "insert-node.left.html")
+ file2 = os.path.join(filepath, "insert-node.right.html")
output, errors = self.call_run([file1, file2])
self.assertEqual(len(output.splitlines()), 3)
# This should default to the diff formatter:
- self.assertEqual(output[0], '[')
+ self.assertEqual(output[0], "[")
def test_diff_cli_args(self):
curdir = os.path.dirname(__file__)
- filepath = os.path.join(curdir, 'test_data')
- file1 = os.path.join(filepath, 'insert-node.left.html')
- file2 = os.path.join(filepath, 'insert-node.right.html')
+ filepath = os.path.join(curdir, "test_data")
+ file1 = os.path.join(filepath, "insert-node.left.html")
+ file2 = os.path.join(filepath, "insert-node.right.html")
# Select a formatter:
- output, errors = self.call_run([file1, file2, '--formatter', 'xml'])
+ output, errors = self.call_run([file1, file2, "--formatter", "xml"])
# It gives a very compact output
self.assertEqual(len(output.splitlines()), 1)
# Now it's XML
- self.assertEqual(output[0], '<')
+ self.assertEqual(output[0], "<")
# Don't strip the whitespace keeps the formatting from the source:
- output, errors = self.call_run([file1, file2, '--keep-whitespace',
- '--formatter', 'xml'])
+ output, errors = self.call_run(
+ [file1, file2, "--keep-whitespace", "--formatter", "xml"]
+ )
self.assertEqual(len(output.splitlines()), 5)
# And stripping and pretty printing gives a longer readable output
- output, errors = self.call_run([file1, file2, '--pretty-print',
- '--formatter', 'xml'])
+ output, errors = self.call_run(
+ [file1, file2, "--pretty-print", "--formatter", "xml"]
+ )
self.assertEqual(len(output.splitlines()), 6)
# The default output gives three lines for three actions
- output, errors = self.call_run([file1, file2, '--ratio-mode', 'fast'])
+ output, errors = self.call_run([file1, file2, "--ratio-mode", "fast"])
self.assertEqual(len(output.splitlines()), 3)
# 'fast' is default, so it's the same output
- output2, errors = self.call_run([file1, file2, '--ratio-mode', 'fast'])
+ output2, errors = self.call_run([file1, file2, "--ratio-mode", "fast"])
self.assertEqual(output, output2)
# Accurate is the same in this case, although sometimes it isn't
- output2, errors = self.call_run([file1, file2, '--ratio-mode',
- 'accurate'])
+ output2, errors = self.call_run([file1, file2, "--ratio-mode", "accurate"])
self.assertEqual(output, output2)
# But "faster" gives six actions instead of three
- output, errors = self.call_run([file1, file2, '--ratio-mode',
- 'faster'])
+ output, errors = self.call_run([file1, file2, "--ratio-mode", "faster"])
self.assertEqual(len(output.splitlines()), 6)
# You can specify unique attributes:
- output, errors = self.call_run([file1, file2, '--unique-attributes',
- 'id,foo,frotz'])
+ output, errors = self.call_run(
+ [file1, file2, "--unique-attributes", "id,foo,frotz"]
+ )
self.assertEqual(len(output.splitlines()), 3)
# Or none
- output, errors = self.call_run([file1, file2, '--unique-attributes'])
+ output, errors = self.call_run([file1, file2, "--unique-attributes"])
self.assertEqual(len(output.splitlines()), 3)
def test_patch_cli_simple(self):
curdir = os.path.dirname(__file__)
- filepath = os.path.join(curdir, 'test_data')
- patchfile = os.path.join(filepath, 'insert-node.diff')
- xmlfile = os.path.join(filepath, 'insert-node.left.html')
+ filepath = os.path.join(curdir, "test_data")
+ patchfile = os.path.join(filepath, "insert-node.diff")
+ xmlfile = os.path.join(filepath, "insert-node.left.html")
- output, errors = self.call_run([patchfile, xmlfile],
- command=main.patch_command)
+ output, errors = self.call_run([patchfile, xmlfile], command=main.patch_command)
- expectedfile = os.path.join(filepath, 'insert-node.right.html')
- with open(expectedfile, 'rt') as f:
+ expectedfile = os.path.join(filepath, "insert-node.right.html")
+ with open(expectedfile) as f:
expected = f.read()
self.assertEqual(output, expected)
Index: xmldiff-2.4/tests/test_patch.py
===================================================================
--- xmldiff-2.4.orig/tests/test_patch.py
+++ xmldiff-2.4/tests/test_patch.py
@@ -5,10 +5,19 @@ from lxml import etree
from xmldiff.formatting import DiffFormatter, WS_NONE
from xmldiff.main import diff_trees, diff_texts, patch_text, patch_file
from xmldiff.patch import Patcher, DiffParser
-from xmldiff.actions import (UpdateTextIn, InsertNode, MoveNode,
- DeleteNode, UpdateAttrib, InsertAttrib,
- RenameAttrib, DeleteAttrib, UpdateTextAfter,
- RenameNode, InsertComment)
+from xmldiff.actions import (
+ UpdateTextIn,
+ InsertNode,
+ MoveNode,
+ DeleteNode,
+ UpdateAttrib,
+ InsertAttrib,
+ RenameAttrib,
+ DeleteAttrib,
+ UpdateTextAfter,
+ RenameNode,
+ InsertComment,
+)
from .testing import compare_elements
@@ -23,67 +32,84 @@ class PatcherTests(unittest.TestCase):
self.assertEqual(etree.tounicode(tree), end)
def test_delete_node(self):
- self._test('<root><deleteme/></root>',
- DeleteNode('/root/deleteme'),
- '<root/>')
+ self._test("<root><deleteme/></root>", DeleteNode("/root/deleteme"), "<root/>")
def test_insert_node(self):
- self._test('<root><anode/></root>',
- InsertNode('/root/anode', 'newnode', 0),
- '<root><anode><newnode/></anode></root>')
+ self._test(
+ "<root><anode/></root>",
+ InsertNode("/root/anode", "newnode", 0),
+ "<root><anode><newnode/></anode></root>",
+ )
def test_rename_node(self):
- self._test('<root><oldname/></root>',
- RenameNode('/root/oldname', 'newname'),
- '<root><newname/></root>')
+ self._test(
+ "<root><oldname/></root>",
+ RenameNode("/root/oldname", "newname"),
+ "<root><newname/></root>",
+ )
def test_move_node(self):
- self._test('<root><anode><moveme/></anode></root>',
- MoveNode('/root/anode/moveme', '/root', 1),
- '<root><anode/><moveme/></root>')
+ self._test(
+ "<root><anode><moveme/></anode></root>",
+ MoveNode("/root/anode/moveme", "/root", 1),
+ "<root><anode/><moveme/></root>",
+ )
def test_update_text_in(self):
- self._test('<root><anode/></root>',
- UpdateTextIn('/root/anode', 'New text'),
- '<root><anode>New text</anode></root>')
+ self._test(
+ "<root><anode/></root>",
+ UpdateTextIn("/root/anode", "New text"),
+ "<root><anode>New text</anode></root>",
+ )
def test_update_text_after(self):
- self._test('<root><anode/></root>',
- UpdateTextAfter('/root/anode', 'New text'),
- '<root><anode/>New text</root>')
+ self._test(
+ "<root><anode/></root>",
+ UpdateTextAfter("/root/anode", "New text"),
+ "<root><anode/>New text</root>",
+ )
def test_update_attrib(self):
- self._test('<root><anode attrib="oldvalue" /></root>',
- UpdateAttrib('/root/anode', 'attrib', 'newvalue'),
- '<root><anode attrib="newvalue"/></root>')
+ self._test(
+ '<root><anode attrib="oldvalue" /></root>',
+ UpdateAttrib("/root/anode", "attrib", "newvalue"),
+ '<root><anode attrib="newvalue"/></root>',
+ )
def test_delete_attrib(self):
- self._test('<root><anode attrib="oldvalue" /></root>',
- DeleteAttrib('/root/anode', 'attrib'),
- '<root><anode/></root>')
+ self._test(
+ '<root><anode attrib="oldvalue" /></root>',
+ DeleteAttrib("/root/anode", "attrib"),
+ "<root><anode/></root>",
+ )
def test_insert_attrib(self):
- self._test('<root><anode/></root>',
- InsertAttrib('/root/anode', 'attrib', 'value'),
- '<root><anode attrib="value"/></root>')
+ self._test(
+ "<root><anode/></root>",
+ InsertAttrib("/root/anode", "attrib", "value"),
+ '<root><anode attrib="value"/></root>',
+ )
def test_rename_attrib(self):
- self._test('<root><anode oldname="value"/></root>',
- RenameAttrib('/root/anode', 'oldname', 'newname'),
- '<root><anode newname="value"/></root>')
+ self._test(
+ '<root><anode oldname="value"/></root>',
+ RenameAttrib("/root/anode", "oldname", "newname"),
+ '<root><anode newname="value"/></root>',
+ )
def test_insert_comment(self):
- self._test('<root><anode/></root>',
- InsertComment('/root', 1, "This is a new comment"),
- '<root><anode/><!--This is a new comment--></root>')
+ self._test(
+ "<root><anode/></root>",
+ InsertComment("/root", 1, "This is a new comment"),
+ "<root><anode/><!--This is a new comment--></root>",
+ )
class DiffPatch(unittest.TestCase):
-
def test_diff_patch(self):
here = os.path.split(__file__)[0]
- lfile = os.path.join(here, 'test_data', 'all_actions.left.xml')
- rfile = os.path.join(here, 'test_data', 'all_actions.right.xml')
+ lfile = os.path.join(here, "test_data", "all_actions.left.xml")
+ rfile = os.path.join(here, "test_data", "all_actions.right.xml")
left = etree.parse(lfile)
right = etree.parse(rfile)
@@ -96,7 +122,7 @@ class DiffPatch(unittest.TestCase):
compare_elements(result.getroot(), right.getroot())
-TEST_DIFF = '''[delete, node]
+TEST_DIFF = """[delete, node]
[insert, target, tag, 0]
[rename, node, tag]
[move, node, target, 0]
@@ -107,67 +133,61 @@ TEST_DIFF = '''[delete, node]
[insert-attribute, node, name, "value"]
[rename-attribute, node, oldname, newname]
[insert-comment, target, 0, "text"]
-'''
+"""
class ParserTests(unittest.TestCase):
-
def test_make_action(self):
parser = DiffParser()
- self.assertEqual(
- parser.make_action('[delete, node]'),
- DeleteNode('node')
- )
+ self.assertEqual(parser.make_action("[delete, node]"), DeleteNode("node"))
self.assertEqual(
- parser.make_action('[insert, target, tag, 0]'),
- InsertNode('target', 'tag', 0)
+ parser.make_action("[insert, target, tag, 0]"),
+ InsertNode("target", "tag", 0),
)
self.assertEqual(
- parser.make_action('[rename, node, tag]'),
- RenameNode('node', 'tag')
+ parser.make_action("[rename, node, tag]"), RenameNode("node", "tag")
)
self.assertEqual(
- parser.make_action('[move, node, target, 0]'),
- MoveNode('node', 'target', 0)
+ parser.make_action("[move, node, target, 0]"), MoveNode("node", "target", 0)
)
self.assertEqual(
parser.make_action('[update-text, node, "text"]'),
- UpdateTextIn('node', 'text')
+ UpdateTextIn("node", "text"),
)
self.assertEqual(
parser.make_action('[update-text-after, node, "text"]'),
- UpdateTextAfter('node', 'text')
+ UpdateTextAfter("node", "text"),
)
self.assertEqual(
parser.make_action('[update-attribute, node, name, "value"]'),
- UpdateAttrib('node', 'name', 'value')
+ UpdateAttrib("node", "name", "value"),
)
self.assertEqual(
- parser.make_action('[delete-attribute, node, name]'),
- DeleteAttrib('node', 'name')
+ parser.make_action("[delete-attribute, node, name]"),
+ DeleteAttrib("node", "name"),
)
self.assertEqual(
parser.make_action('[insert-attribute, node, name, "value"]'),
- InsertAttrib('node', 'name', 'value')
+ InsertAttrib("node", "name", "value"),
)
self.assertEqual(
- parser.make_action('[rename-attribute, node, oldname, newname]'),
- RenameAttrib('node', 'oldname', 'newname')
+ parser.make_action("[rename-attribute, node, oldname, newname]"),
+ RenameAttrib("node", "oldname", "newname"),
)
self.assertEqual(
parser.make_action('[insert-comment, target, 0, "text"]'),
- InsertComment('target', 0, 'text')
+ InsertComment("target", 0, "text"),
)
def test_parse(self):
@@ -180,43 +200,42 @@ class ParserTests(unittest.TestCase):
parser = DiffParser()
# Empty file, nothing happens
- actions = list(parser.parse(''))
+ actions = list(parser.parse(""))
self.assertEqual(actions, [])
# Not a diff raises error
with self.assertRaises(ValueError):
- actions = list(parser.parse('Not a diff'))
+ actions = list(parser.parse("Not a diff"))
# It should handle lines that have been broken, say in an email
actions = list(parser.parse('[insert-comment, target,\n 0, "text"]'))
- self.assertEqual(actions, [InsertComment('target', 0, 'text')])
+ self.assertEqual(actions, [InsertComment("target", 0, "text")])
# It should not handle broken files
with self.assertRaises(ValueError):
- actions = list(parser.parse('[insert-comment, target,\n'))
+ actions = list(parser.parse("[insert-comment, target,\n"))
def test_diff_patch(self):
here = os.path.split(__file__)[0]
- lfile = os.path.join(here, 'test_data', 'all_actions.left.xml')
- rfile = os.path.join(here, 'test_data', 'all_actions.right.xml')
+ lfile = os.path.join(here, "test_data", "all_actions.left.xml")
+ rfile = os.path.join(here, "test_data", "all_actions.right.xml")
with open(lfile) as f:
left = f.read()
with open(rfile) as f:
right = f.read()
- diff = diff_texts(left, right,
- formatter=DiffFormatter(normalize=WS_NONE))
+ diff = diff_texts(left, right, formatter=DiffFormatter(normalize=WS_NONE))
result = patch_text(diff, left)
compare_elements(etree.fromstring(result), etree.fromstring(right))
def test_patch_stream(self):
- here = os.path.join(os.path.split(__file__)[0], 'test_data')
- xmlfile = os.path.join(here, 'insert-node.left.html')
- patchfile = os.path.join(here, 'insert-node.diff')
+ here = os.path.join(os.path.split(__file__)[0], "test_data")
+ xmlfile = os.path.join(here, "insert-node.left.html")
+ patchfile = os.path.join(here, "insert-node.diff")
result = patch_file(patchfile, xmlfile)
- expectedfile = os.path.join(here, 'insert-node.right.html')
- with open(expectedfile, 'rt') as f:
+ expectedfile = os.path.join(here, "insert-node.right.html")
+ with open(expectedfile) as f:
expected = f.read()
# lxml.etree.parse() will strip ending whitespace
self.assertEqual(result, expected.rstrip())
Index: xmldiff-2.4/tests/test_utils.py
===================================================================
--- xmldiff-2.4.orig/tests/test_utils.py
+++ xmldiff-2.4/tests/test_utils.py
@@ -5,9 +5,8 @@ from xmldiff import utils
class TraverseTests(unittest.TestCase):
-
def test_post_order(self):
- xml = u'''<document>
+ xml = """<document>
<story firstPageTemplate='FirstPage'>
<section xml:id='oldfirst' ref='3' single-ref='3'>
<para>First paragraph</para>
@@ -17,19 +16,24 @@ class TraverseTests(unittest.TestCase):
</section>
</story>
</document>
-'''
+"""
root = etree.fromstring(xml)
tree = root.getroottree()
res = [tree.getpath(x) for x in utils.post_order_traverse(root)]
- self.assertEqual(res, ['/document/story/section[1]/para',
- '/document/story/section[1]',
- '/document/story/section[2]/para',
- '/document/story/section[2]',
- '/document/story',
- '/document'])
+ self.assertEqual(
+ res,
+ [
+ "/document/story/section[1]/para",
+ "/document/story/section[1]",
+ "/document/story/section[2]/para",
+ "/document/story/section[2]",
+ "/document/story",
+ "/document",
+ ],
+ )
def test_reverse_post_order(self):
- xml = u'''<document>
+ xml = """<document>
<story firstPageTemplate='FirstPage'>
<section xml:id='oldfirst' ref='3' single-ref='3'>
<para>First paragraph</para>
@@ -39,20 +43,24 @@ class TraverseTests(unittest.TestCase):
</section>
</story>
</document>
-'''
+"""
root = etree.fromstring(xml)
tree = root.getroottree()
- res = [tree.getpath(x) for x in
- utils.reverse_post_order_traverse(root)]
- self.assertEqual(res, ['/document/story/section[2]/para',
- '/document/story/section[2]',
- '/document/story/section[1]/para',
- '/document/story/section[1]',
- '/document/story',
- '/document'])
+ res = [tree.getpath(x) for x in utils.reverse_post_order_traverse(root)]
+ self.assertEqual(
+ res,
+ [
+ "/document/story/section[2]/para",
+ "/document/story/section[2]",
+ "/document/story/section[1]/para",
+ "/document/story/section[1]",
+ "/document/story",
+ "/document",
+ ],
+ )
def test_breadth_first(self):
- xml = u'''<document>
+ xml = """<document>
<story>
<section>
<para>First <i>paragraph</i></para>
@@ -69,68 +77,70 @@ class TraverseTests(unittest.TestCase):
</section>
</story>
</document>
-'''
+"""
root = etree.fromstring(xml)
tree = root.getroottree()
res = [tree.getpath(x) for x in utils.breadth_first_traverse(root)]
- self.assertEqual(res, ['/document',
- '/document/story[1]',
- '/document/story[2]',
- '/document/story[1]/section[1]',
- '/document/story[1]/section[2]',
- '/document/story[2]/section',
- '/document/story[1]/section[1]/para[1]',
- '/document/story[1]/section[1]/para[2]',
- '/document/story[1]/section[2]/para[1]',
- '/document/story[1]/section[2]/para[2]',
- '/document/story[2]/section/para',
- '/document/story[1]/section[1]/para[1]/i',
- '/document/story[1]/section[2]/para[2]/b',
- ])
+ self.assertEqual(
+ res,
+ [
+ "/document",
+ "/document/story[1]",
+ "/document/story[2]",
+ "/document/story[1]/section[1]",
+ "/document/story[1]/section[2]",
+ "/document/story[2]/section",
+ "/document/story[1]/section[1]/para[1]",
+ "/document/story[1]/section[1]/para[2]",
+ "/document/story[1]/section[2]/para[1]",
+ "/document/story[1]/section[2]/para[2]",
+ "/document/story[2]/section/para",
+ "/document/story[1]/section[1]/para[1]/i",
+ "/document/story[1]/section[2]/para[2]/b",
+ ],
+ )
class LongestCommonSubsequenceTests(unittest.TestCase):
-
def _diff(self, left, right, result):
res = []
for x, y in utils.longest_common_subsequence(left, right):
self.assertEqual(left[x], right[y])
res.append(left[x])
- self.assertEqual(''.join(res), result)
+ self.assertEqual("".join(res), result)
def test_lcs(self):
- self._diff('ABCDEF', 'ABCDEF', 'ABCDEF')
+ self._diff("ABCDEF", "ABCDEF", "ABCDEF")
- self._diff('ABCDEF', 'GHIJKL', '')
+ self._diff("ABCDEF", "GHIJKL", "")
- self._diff('ABCDEF', 'ACDQRB', 'ACD')
+ self._diff("ABCDEF", "ACDQRB", "ACD")
- self._diff('CXCDEFX', 'CDEFX', 'CDEFX')
+ self._diff("CXCDEFX", "CDEFX", "CDEFX")
- self._diff('HUMAN', 'CHIMPANZEE', 'HMAN')
+ self._diff("HUMAN", "CHIMPANZEE", "HMAN")
- self._diff('ABCDEF', 'A', 'A')
+ self._diff("ABCDEF", "A", "A")
- self._diff('123AAAAAAAAA', '123BBBBBBBBB', '123')
+ self._diff("123AAAAAAAAA", "123BBBBBBBBB", "123")
- self._diff('AAAAAAAAA123', 'BBBBBBBBB123', '123')
+ self._diff("AAAAAAAAA123", "BBBBBBBBB123", "123")
- self._diff('ABCDE1', '1FGHIJK', '1')
+ self._diff("ABCDE1", "1FGHIJK", "1")
# There are several correct options here, make sure that doesn't
# confuse it, we want just one, and don't care which.
- self._diff('HORSEBACK', 'SNOWFLAKE', 'SAK')
+ self._diff("HORSEBACK", "SNOWFLAKE", "SAK")
# Empty sequences:
- self._diff('', '', '')
+ self._diff("", "", "")
class MakeAsciiTreeTests(unittest.TestCase):
-
def test_make_ascii_tree(self):
- xml = u'''<document xmlns:diff="http://namespaces.shoobx.com/diff">
+ xml = """<document xmlns:diff="http://namespaces.shoobx.com/diff">
<story firstPageTemplate='FirstPage'>
<section xml:id='oldfirst' ref='3' single-ref='3'>
<para diff:delete="">First paragraph</para>
@@ -140,11 +150,11 @@ class MakeAsciiTreeTests(unittest.TestCa
</section>
</story>
</document>
-'''
+"""
root = etree.fromstring(xml)
tree = utils.make_ascii_tree(root)
self.assertEqual(
tree,
- ' document \n story \n section \n para (delete)\n'
- ' section \n para \n diff:insert '
+ " document \n story \n section \n para (delete)\n"
+ " section \n para \n diff:insert ",
)
Index: xmldiff-2.4/tests/testing.py
===================================================================
--- xmldiff-2.4.orig/tests/testing.py
+++ xmldiff-2.4/tests/testing.py
@@ -1,39 +1,37 @@
import os
-from io import open
-
def make_case_function(left_filename):
- right_filename = left_filename.replace('.left.', '.right.')
- expected_filename = left_filename.replace('.left.', '.expected.')
+ right_filename = left_filename.replace(".left.", ".right.")
+ expected_filename = left_filename.replace(".left.", ".expected.")
def test(self):
- with open(expected_filename, 'rt', encoding='utf8') as input_file:
+ with open(expected_filename, encoding="utf8") as input_file:
expected_xml = input_file.read()
try:
result_xml = self.process(left_filename, right_filename)
except Exception as err:
- if u'.err' not in left_filename:
+ if ".err" not in left_filename:
raise
- result_xml = u'%s: %s' % (err.__class__.__name__, err)
+ result_xml = f"{err.__class__.__name__}: {err}"
self.assertEqual(expected_xml.strip(), result_xml.strip())
return test
-def generate_filebased_cases(data_dir, test_class, suffix='xml', ignore=()):
+def generate_filebased_cases(data_dir, test_class, suffix="xml", ignore=()):
for left_filename in os.listdir(data_dir):
- if not left_filename.endswith('.left.' + suffix):
+ if not left_filename.endswith(".left." + suffix):
continue
if left_filename in ignore:
continue
left_filename = os.path.join(data_dir, left_filename)
test_function = make_case_function(left_filename)
- function_name = os.path.split(left_filename)[-1].replace('.', '-')
- test_name = 'test_' + function_name
+ function_name = os.path.split(left_filename)[-1].replace(".", "-")
+ test_name = "test_" + function_name
setattr(test_class, test_name, test_function)
Index: xmldiff-2.4/xmldiff/_diff_match_patch_py2.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/_diff_match_patch_py2.py
+++ xmldiff-2.4/xmldiff/_diff_match_patch_py2.py
@@ -1,6 +1,5 @@
#!/usr/bin/python2.4
-from __future__ import division
"""Diff Match and Patch
Copyright 2018 The diff-match-patch Authors.
@@ -25,7 +24,7 @@ Computes the difference between two text
Applies the patch onto another text, allowing for errors.
"""
-__author__ = 'fraser@google.com (Neil Fraser)'
+__author__ = "fraser@google.com (Neil Fraser)"
import re
import sys
@@ -34,51 +33,51 @@ import urllib
class diff_match_patch:
- """Class containing the diff, match and patch methods.
+ """Class containing the diff, match and patch methods.
Also contains the behaviour settings.
"""
- def __init__(self):
- """Inits a diff_match_patch object with default settings.
+ def __init__(self):
+ """Inits a diff_match_patch object with default settings.
Redefine these in your program to override the defaults.
"""
- # Number of seconds to map a diff before giving up (0 for infinity).
- self.Diff_Timeout = 1.0
- # Cost of an empty edit operation in terms of edit characters.
- self.Diff_EditCost = 4
- # At what point is no match declared (0.0 = perfection, 1.0 = very loose).
- self.Match_Threshold = 0.5
- # How far to search for a match (0 = exact location, 1000+ = broad match).
- # A match this many characters away from the expected location will add
- # 1.0 to the score (0.0 is a perfect match).
- self.Match_Distance = 1000
- # When deleting a large block of text (over ~64 characters), how close do
- # the contents have to be to match the expected contents. (0.0 = perfection,
- # 1.0 = very loose). Note that Match_Threshold controls how closely the
- # end points of a delete need to match.
- self.Patch_DeleteThreshold = 0.5
- # Chunk size for context length.
- self.Patch_Margin = 4
-
- # The number of bits in an int.
- # Python has no maximum, thus to disable patch splitting set to 0.
- # However to avoid long patches in certain pathological cases, use 32.
- # Multiple short patches (using native ints) are much faster than long ones.
- self.Match_MaxBits = 32
-
- # DIFF FUNCTIONS
-
- # The data structure representing a diff is an array of tuples:
- # [(DIFF_DELETE, "Hello"), (DIFF_INSERT, "Goodbye"), (DIFF_EQUAL, " world.")]
- # which means: delete "Hello", add "Goodbye" and keep " world."
- DIFF_DELETE = -1
- DIFF_INSERT = 1
- DIFF_EQUAL = 0
+ # Number of seconds to map a diff before giving up (0 for infinity).
+ self.Diff_Timeout = 1.0
+ # Cost of an empty edit operation in terms of edit characters.
+ self.Diff_EditCost = 4
+ # At what point is no match declared (0.0 = perfection, 1.0 = very loose).
+ self.Match_Threshold = 0.5
+ # How far to search for a match (0 = exact location, 1000+ = broad match).
+ # A match this many characters away from the expected location will add
+ # 1.0 to the score (0.0 is a perfect match).
+ self.Match_Distance = 1000
+ # When deleting a large block of text (over ~64 characters), how close do
+ # the contents have to be to match the expected contents. (0.0 = perfection,
+ # 1.0 = very loose). Note that Match_Threshold controls how closely the
+ # end points of a delete need to match.
+ self.Patch_DeleteThreshold = 0.5
+ # Chunk size for context length.
+ self.Patch_Margin = 4
+
+ # The number of bits in an int.
+ # Python has no maximum, thus to disable patch splitting set to 0.
+ # However to avoid long patches in certain pathological cases, use 32.
+ # Multiple short patches (using native ints) are much faster than long ones.
+ self.Match_MaxBits = 32
+
+ # DIFF FUNCTIONS
+
+ # The data structure representing a diff is an array of tuples:
+ # [(DIFF_DELETE, "Hello"), (DIFF_INSERT, "Goodbye"), (DIFF_EQUAL, " world.")]
+ # which means: delete "Hello", add "Goodbye" and keep " world."
+ DIFF_DELETE = -1
+ DIFF_INSERT = 1
+ DIFF_EQUAL = 0
- def diff_main(self, text1, text2, checklines=True, deadline=None):
- """Find the differences between two texts. Simplifies the problem by
+ def diff_main(self, text1, text2, checklines=True, deadline=None):
+ """Find the differences between two texts. Simplifies the problem by
stripping any common prefix or suffix off the texts before diffing.
Args:
@@ -93,52 +92,52 @@ class diff_match_patch:
Returns:
Array of changes.
"""
- # Set a deadline by which time the diff must be complete.
- if deadline == None:
- # Unlike in most languages, Python counts time in seconds.
- if self.Diff_Timeout <= 0:
- deadline = sys.maxint
- else:
- deadline = time.time() + self.Diff_Timeout
-
- # Check for null inputs.
- if text1 == None or text2 == None:
- raise ValueError("Null inputs. (diff_main)")
-
- # Check for equality (speedup).
- if text1 == text2:
- if text1:
- return [(self.DIFF_EQUAL, text1)]
- return []
-
- # Trim off common prefix (speedup).
- commonlength = self.diff_commonPrefix(text1, text2)
- commonprefix = text1[:commonlength]
- text1 = text1[commonlength:]
- text2 = text2[commonlength:]
-
- # Trim off common suffix (speedup).
- commonlength = self.diff_commonSuffix(text1, text2)
- if commonlength == 0:
- commonsuffix = ''
- else:
- commonsuffix = text1[-commonlength:]
- text1 = text1[:-commonlength]
- text2 = text2[:-commonlength]
-
- # Compute the diff on the middle block.
- diffs = self.diff_compute(text1, text2, checklines, deadline)
-
- # Restore the prefix and suffix.
- if commonprefix:
- diffs[:0] = [(self.DIFF_EQUAL, commonprefix)]
- if commonsuffix:
- diffs.append((self.DIFF_EQUAL, commonsuffix))
- self.diff_cleanupMerge(diffs)
- return diffs
+ # Set a deadline by which time the diff must be complete.
+ if deadline == None:
+ # Unlike in most languages, Python counts time in seconds.
+ if self.Diff_Timeout <= 0:
+ deadline = sys.maxint
+ else:
+ deadline = time.time() + self.Diff_Timeout
+
+ # Check for null inputs.
+ if text1 == None or text2 == None:
+ raise ValueError("Null inputs. (diff_main)")
+
+ # Check for equality (speedup).
+ if text1 == text2:
+ if text1:
+ return [(self.DIFF_EQUAL, text1)]
+ return []
+
+ # Trim off common prefix (speedup).
+ commonlength = self.diff_commonPrefix(text1, text2)
+ commonprefix = text1[:commonlength]
+ text1 = text1[commonlength:]
+ text2 = text2[commonlength:]
+
+ # Trim off common suffix (speedup).
+ commonlength = self.diff_commonSuffix(text1, text2)
+ if commonlength == 0:
+ commonsuffix = ""
+ else:
+ commonsuffix = text1[-commonlength:]
+ text1 = text1[:-commonlength]
+ text2 = text2[:-commonlength]
+
+ # Compute the diff on the middle block.
+ diffs = self.diff_compute(text1, text2, checklines, deadline)
+
+ # Restore the prefix and suffix.
+ if commonprefix:
+ diffs[:0] = [(self.DIFF_EQUAL, commonprefix)]
+ if commonsuffix:
+ diffs.append((self.DIFF_EQUAL, commonsuffix))
+ self.diff_cleanupMerge(diffs)
+ return diffs
- def diff_compute(self, text1, text2, checklines, deadline):
- """Find the differences between two texts. Assumes that the texts do not
+ def diff_compute(self, text1, text2, checklines, deadline):
+ """Find the differences between two texts. Assumes that the texts do not
have any common prefix or suffix.
Args:
@@ -152,52 +151,55 @@ class diff_match_patch:
Returns:
Array of changes.
"""
- if not text1:
- # Just add some text (speedup).
- return [(self.DIFF_INSERT, text2)]
-
- if not text2:
- # Just delete some text (speedup).
- return [(self.DIFF_DELETE, text1)]
-
- if len(text1) > len(text2):
- (longtext, shorttext) = (text1, text2)
- else:
- (shorttext, longtext) = (text1, text2)
- i = longtext.find(shorttext)
- if i != -1:
- # Shorter text is inside the longer text (speedup).
- diffs = [(self.DIFF_INSERT, longtext[:i]), (self.DIFF_EQUAL, shorttext),
- (self.DIFF_INSERT, longtext[i + len(shorttext):])]
- # Swap insertions for deletions if diff is reversed.
- if len(text1) > len(text2):
- diffs[0] = (self.DIFF_DELETE, diffs[0][1])
- diffs[2] = (self.DIFF_DELETE, diffs[2][1])
- return diffs
-
- if len(shorttext) == 1:
- # Single character string.
- # After the previous speedup, the character can't be an equality.
- return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
-
- # Check to see if the problem can be split in two.
- hm = self.diff_halfMatch(text1, text2)
- if hm:
- # A half-match was found, sort out the return data.
- (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
- # Send both pairs off for separate processing.
- diffs_a = self.diff_main(text1_a, text2_a, checklines, deadline)
- diffs_b = self.diff_main(text1_b, text2_b, checklines, deadline)
- # Merge the results.
- return diffs_a + [(self.DIFF_EQUAL, mid_common)] + diffs_b
+ if not text1:
+ # Just add some text (speedup).
+ return [(self.DIFF_INSERT, text2)]
+
+ if not text2:
+ # Just delete some text (speedup).
+ return [(self.DIFF_DELETE, text1)]
- if checklines and len(text1) > 100 and len(text2) > 100:
- return self.diff_lineMode(text1, text2, deadline)
+ if len(text1) > len(text2):
+ (longtext, shorttext) = (text1, text2)
+ else:
+ (shorttext, longtext) = (text1, text2)
+ i = longtext.find(shorttext)
+ if i != -1:
+ # Shorter text is inside the longer text (speedup).
+ diffs = [
+ (self.DIFF_INSERT, longtext[:i]),
+ (self.DIFF_EQUAL, shorttext),
+ (self.DIFF_INSERT, longtext[i + len(shorttext) :]),
+ ]
+ # Swap insertions for deletions if diff is reversed.
+ if len(text1) > len(text2):
+ diffs[0] = (self.DIFF_DELETE, diffs[0][1])
+ diffs[2] = (self.DIFF_DELETE, diffs[2][1])
+ return diffs
+
+ if len(shorttext) == 1:
+ # Single character string.
+ # After the previous speedup, the character can't be an equality.
+ return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
+
+ # Check to see if the problem can be split in two.
+ hm = self.diff_halfMatch(text1, text2)
+ if hm:
+ # A half-match was found, sort out the return data.
+ (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
+ # Send both pairs off for separate processing.
+ diffs_a = self.diff_main(text1_a, text2_a, checklines, deadline)
+ diffs_b = self.diff_main(text1_b, text2_b, checklines, deadline)
+ # Merge the results.
+ return diffs_a + [(self.DIFF_EQUAL, mid_common)] + diffs_b
+
+ if checklines and len(text1) > 100 and len(text2) > 100:
+ return self.diff_lineMode(text1, text2, deadline)
- return self.diff_bisect(text1, text2, deadline)
+ return self.diff_bisect(text1, text2, deadline)
- def diff_lineMode(self, text1, text2, deadline):
- """Do a quick line-level diff on both strings, then rediff the parts for
+ def diff_lineMode(self, text1, text2, deadline):
+ """Do a quick line-level diff on both strings, then rediff the parts for
greater accuracy.
This speedup can produce non-minimal diffs.
@@ -210,51 +212,51 @@ class diff_match_patch:
Array of changes.
"""
- # Scan the text on a line-by-line basis first.
- (text1, text2, linearray) = self.diff_linesToChars(text1, text2)
+ # Scan the text on a line-by-line basis first.
+ (text1, text2, linearray) = self.diff_linesToChars(text1, text2)
- diffs = self.diff_main(text1, text2, False, deadline)
+ diffs = self.diff_main(text1, text2, False, deadline)
- # Convert the diff back to original text.
- self.diff_charsToLines(diffs, linearray)
- # Eliminate freak matches (e.g. blank lines)
- self.diff_cleanupSemantic(diffs)
-
- # Rediff any replacement blocks, this time character-by-character.
- # Add a dummy entry at the end.
- diffs.append((self.DIFF_EQUAL, ''))
- pointer = 0
- count_delete = 0
- count_insert = 0
- text_delete = ''
- text_insert = ''
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_INSERT:
- count_insert += 1
- text_insert += diffs[pointer][1]
- elif diffs[pointer][0] == self.DIFF_DELETE:
- count_delete += 1
- text_delete += diffs[pointer][1]
- elif diffs[pointer][0] == self.DIFF_EQUAL:
- # Upon reaching an equality, check for prior redundancies.
- if count_delete >= 1 and count_insert >= 1:
- # Delete the offending records and add the merged ones.
- subDiff = self.diff_main(text_delete, text_insert, False, deadline)
- diffs[pointer - count_delete - count_insert : pointer] = subDiff
- pointer = pointer - count_delete - count_insert + len(subDiff)
- count_insert = 0
+ # Convert the diff back to original text.
+ self.diff_charsToLines(diffs, linearray)
+ # Eliminate freak matches (e.g. blank lines)
+ self.diff_cleanupSemantic(diffs)
+
+ # Rediff any replacement blocks, this time character-by-character.
+ # Add a dummy entry at the end.
+ diffs.append((self.DIFF_EQUAL, ""))
+ pointer = 0
count_delete = 0
- text_delete = ''
- text_insert = ''
+ count_insert = 0
+ text_delete = ""
+ text_insert = ""
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_INSERT:
+ count_insert += 1
+ text_insert += diffs[pointer][1]
+ elif diffs[pointer][0] == self.DIFF_DELETE:
+ count_delete += 1
+ text_delete += diffs[pointer][1]
+ elif diffs[pointer][0] == self.DIFF_EQUAL:
+ # Upon reaching an equality, check for prior redundancies.
+ if count_delete >= 1 and count_insert >= 1:
+ # Delete the offending records and add the merged ones.
+ subDiff = self.diff_main(text_delete, text_insert, False, deadline)
+ diffs[pointer - count_delete - count_insert : pointer] = subDiff
+ pointer = pointer - count_delete - count_insert + len(subDiff)
+ count_insert = 0
+ count_delete = 0
+ text_delete = ""
+ text_insert = ""
- pointer += 1
+ pointer += 1
- diffs.pop() # Remove the dummy entry at the end.
+ diffs.pop() # Remove the dummy entry at the end.
- return diffs
+ return diffs
- def diff_bisect(self, text1, text2, deadline):
- """Find the 'middle snake' of a diff, split the problem in two
+ def diff_bisect(self, text1, text2, deadline):
+ """Find the 'middle snake' of a diff, split the problem in two
and return the recursively constructed diff.
See Myers 1986 paper: An O(ND) Difference Algorithm and Its Variations.
@@ -267,96 +269,98 @@ class diff_match_patch:
Array of diff tuples.
"""
- # Cache the text lengths to prevent multiple calls.
- text1_length = len(text1)
- text2_length = len(text2)
- max_d = (text1_length + text2_length + 1) // 2
- v_offset = max_d
- v_length = 2 * max_d
- v1 = [-1] * v_length
- v1[v_offset + 1] = 0
- v2 = v1[:]
- delta = text1_length - text2_length
- # If the total number of characters is odd, then the front path will
- # collide with the reverse path.
- front = (delta % 2 != 0)
- # Offsets for start and end of k loop.
- # Prevents mapping of space beyond the grid.
- k1start = 0
- k1end = 0
- k2start = 0
- k2end = 0
- for d in xrange(max_d):
- # Bail out if deadline is reached.
- if time.time() > deadline:
- break
-
- # Walk the front path one step.
- for k1 in xrange(-d + k1start, d + 1 - k1end, 2):
- k1_offset = v_offset + k1
- if k1 == -d or (k1 != d and
- v1[k1_offset - 1] < v1[k1_offset + 1]):
- x1 = v1[k1_offset + 1]
- else:
- x1 = v1[k1_offset - 1] + 1
- y1 = x1 - k1
- while (x1 < text1_length and y1 < text2_length and
- text1[x1] == text2[y1]):
- x1 += 1
- y1 += 1
- v1[k1_offset] = x1
- if x1 > text1_length:
- # Ran off the right of the graph.
- k1end += 2
- elif y1 > text2_length:
- # Ran off the bottom of the graph.
- k1start += 2
- elif front:
- k2_offset = v_offset + delta - k1
- if k2_offset >= 0 and k2_offset < v_length and v2[k2_offset] != -1:
- # Mirror x2 onto top-left coordinate system.
- x2 = text1_length - v2[k2_offset]
- if x1 >= x2:
- # Overlap detected.
- return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
-
- # Walk the reverse path one step.
- for k2 in xrange(-d + k2start, d + 1 - k2end, 2):
- k2_offset = v_offset + k2
- if k2 == -d or (k2 != d and
- v2[k2_offset - 1] < v2[k2_offset + 1]):
- x2 = v2[k2_offset + 1]
- else:
- x2 = v2[k2_offset - 1] + 1
- y2 = x2 - k2
- while (x2 < text1_length and y2 < text2_length and
- text1[-x2 - 1] == text2[-y2 - 1]):
- x2 += 1
- y2 += 1
- v2[k2_offset] = x2
- if x2 > text1_length:
- # Ran off the left of the graph.
- k2end += 2
- elif y2 > text2_length:
- # Ran off the top of the graph.
- k2start += 2
- elif not front:
- k1_offset = v_offset + delta - k2
- if k1_offset >= 0 and k1_offset < v_length and v1[k1_offset] != -1:
- x1 = v1[k1_offset]
- y1 = v_offset + x1 - k1_offset
- # Mirror x2 onto top-left coordinate system.
- x2 = text1_length - x2
- if x1 >= x2:
- # Overlap detected.
- return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
-
- # Diff took too long and hit the deadline or
- # number of diffs equals number of characters, no commonality at all.
- return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
+ # Cache the text lengths to prevent multiple calls.
+ text1_length = len(text1)
+ text2_length = len(text2)
+ max_d = (text1_length + text2_length + 1) // 2
+ v_offset = max_d
+ v_length = 2 * max_d
+ v1 = [-1] * v_length
+ v1[v_offset + 1] = 0
+ v2 = v1[:]
+ delta = text1_length - text2_length
+ # If the total number of characters is odd, then the front path will
+ # collide with the reverse path.
+ front = delta % 2 != 0
+ # Offsets for start and end of k loop.
+ # Prevents mapping of space beyond the grid.
+ k1start = 0
+ k1end = 0
+ k2start = 0
+ k2end = 0
+ for d in xrange(max_d):
+ # Bail out if deadline is reached.
+ if time.time() > deadline:
+ break
+
+ # Walk the front path one step.
+ for k1 in xrange(-d + k1start, d + 1 - k1end, 2):
+ k1_offset = v_offset + k1
+ if k1 == -d or (k1 != d and v1[k1_offset - 1] < v1[k1_offset + 1]):
+ x1 = v1[k1_offset + 1]
+ else:
+ x1 = v1[k1_offset - 1] + 1
+ y1 = x1 - k1
+ while (
+ x1 < text1_length and y1 < text2_length and text1[x1] == text2[y1]
+ ):
+ x1 += 1
+ y1 += 1
+ v1[k1_offset] = x1
+ if x1 > text1_length:
+ # Ran off the right of the graph.
+ k1end += 2
+ elif y1 > text2_length:
+ # Ran off the bottom of the graph.
+ k1start += 2
+ elif front:
+ k2_offset = v_offset + delta - k1
+ if k2_offset >= 0 and k2_offset < v_length and v2[k2_offset] != -1:
+ # Mirror x2 onto top-left coordinate system.
+ x2 = text1_length - v2[k2_offset]
+ if x1 >= x2:
+ # Overlap detected.
+ return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
+
+ # Walk the reverse path one step.
+ for k2 in xrange(-d + k2start, d + 1 - k2end, 2):
+ k2_offset = v_offset + k2
+ if k2 == -d or (k2 != d and v2[k2_offset - 1] < v2[k2_offset + 1]):
+ x2 = v2[k2_offset + 1]
+ else:
+ x2 = v2[k2_offset - 1] + 1
+ y2 = x2 - k2
+ while (
+ x2 < text1_length
+ and y2 < text2_length
+ and text1[-x2 - 1] == text2[-y2 - 1]
+ ):
+ x2 += 1
+ y2 += 1
+ v2[k2_offset] = x2
+ if x2 > text1_length:
+ # Ran off the left of the graph.
+ k2end += 2
+ elif y2 > text2_length:
+ # Ran off the top of the graph.
+ k2start += 2
+ elif not front:
+ k1_offset = v_offset + delta - k2
+ if k1_offset >= 0 and k1_offset < v_length and v1[k1_offset] != -1:
+ x1 = v1[k1_offset]
+ y1 = v_offset + x1 - k1_offset
+ # Mirror x2 onto top-left coordinate system.
+ x2 = text1_length - x2
+ if x1 >= x2:
+ # Overlap detected.
+ return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
+
+ # Diff took too long and hit the deadline or
+ # number of diffs equals number of characters, no commonality at all.
+ return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
- def diff_bisectSplit(self, text1, text2, x, y, deadline):
- """Given the location of the 'middle snake', split the diff in two parts
+ def diff_bisectSplit(self, text1, text2, x, y, deadline):
+ """Given the location of the 'middle snake', split the diff in two parts
and recurse.
Args:
@@ -369,19 +373,19 @@ class diff_match_patch:
Returns:
Array of diff tuples.
"""
- text1a = text1[:x]
- text2a = text2[:y]
- text1b = text1[x:]
- text2b = text2[y:]
-
- # Compute both diffs serially.
- diffs = self.diff_main(text1a, text2a, False, deadline)
- diffsb = self.diff_main(text1b, text2b, False, deadline)
+ text1a = text1[:x]
+ text2a = text2[:y]
+ text1b = text1[x:]
+ text2b = text2[y:]
+
+ # Compute both diffs serially.
+ diffs = self.diff_main(text1a, text2a, False, deadline)
+ diffsb = self.diff_main(text1b, text2b, False, deadline)
- return diffs + diffsb
+ return diffs + diffsb
- def diff_linesToChars(self, text1, text2):
- """Split two texts into an array of strings. Reduce the texts to a string
+ def diff_linesToChars(self, text1, text2):
+ """Split two texts into an array of strings. Reduce the texts to a string
of hashes where each Unicode character represents one line.
Args:
@@ -393,15 +397,15 @@ class diff_match_patch:
the array of unique strings. The zeroth element of the array of unique
strings is intentionally blank.
"""
- lineArray = [] # e.g. lineArray[4] == "Hello\n"
- lineHash = {} # e.g. lineHash["Hello\n"] == 4
+ lineArray = [] # e.g. lineArray[4] == "Hello\n"
+ lineHash = {} # e.g. lineHash["Hello\n"] == 4
- # "\x00" is a valid character, but various debuggers don't like it.
- # So we'll insert a junk entry to avoid generating a null character.
- lineArray.append('')
+ # "\x00" is a valid character, but various debuggers don't like it.
+ # So we'll insert a junk entry to avoid generating a null character.
+ lineArray.append("")
- def diff_linesToCharsMunge(text):
- """Split a text into an array of strings. Reduce the texts to a string
+ def diff_linesToCharsMunge(text):
+ """Split a text into an array of strings. Reduce the texts to a string
of hashes where each Unicode character represents one line.
Modifies linearray and linehash through being a closure.
@@ -411,54 +415,54 @@ class diff_match_patch:
Returns:
Encoded string.
"""
- chars = []
- # Walk the text, pulling out a substring for each line.
- # text.split('\n') would would temporarily double our memory footprint.
- # Modifying text would create many large strings to garbage collect.
- lineStart = 0
- lineEnd = -1
- while lineEnd < len(text) - 1:
- lineEnd = text.find('\n', lineStart)
- if lineEnd == -1:
- lineEnd = len(text) - 1
- line = text[lineStart:lineEnd + 1]
-
- if line in lineHash:
- chars.append(unichr(lineHash[line]))
- else:
- if len(lineArray) == maxLines:
- # Bail out at 65535 because unichr(65536) throws.
- line = text[lineStart:]
- lineEnd = len(text)
- lineArray.append(line)
- lineHash[line] = len(lineArray) - 1
- chars.append(unichr(len(lineArray) - 1))
- lineStart = lineEnd + 1
- return "".join(chars)
-
- # Allocate 2/3rds of the space for text1, the rest for text2.
- maxLines = 40000
- chars1 = diff_linesToCharsMunge(text1)
- maxLines = 65535
- chars2 = diff_linesToCharsMunge(text2)
- return (chars1, chars2, lineArray)
+ chars = []
+ # Walk the text, pulling out a substring for each line.
+ # text.split('\n') would would temporarily double our memory footprint.
+ # Modifying text would create many large strings to garbage collect.
+ lineStart = 0
+ lineEnd = -1
+ while lineEnd < len(text) - 1:
+ lineEnd = text.find("\n", lineStart)
+ if lineEnd == -1:
+ lineEnd = len(text) - 1
+ line = text[lineStart : lineEnd + 1]
+
+ if line in lineHash:
+ chars.append(unichr(lineHash[line]))
+ else:
+ if len(lineArray) == maxLines:
+ # Bail out at 65535 because unichr(65536) throws.
+ line = text[lineStart:]
+ lineEnd = len(text)
+ lineArray.append(line)
+ lineHash[line] = len(lineArray) - 1
+ chars.append(unichr(len(lineArray) - 1))
+ lineStart = lineEnd + 1
+ return "".join(chars)
+
+ # Allocate 2/3rds of the space for text1, the rest for text2.
+ maxLines = 40000
+ chars1 = diff_linesToCharsMunge(text1)
+ maxLines = 65535
+ chars2 = diff_linesToCharsMunge(text2)
+ return (chars1, chars2, lineArray)
- def diff_charsToLines(self, diffs, lineArray):
- """Rehydrate the text in a diff from a string of line hashes to real lines
+ def diff_charsToLines(self, diffs, lineArray):
+ """Rehydrate the text in a diff from a string of line hashes to real lines
of text.
Args:
diffs: Array of diff tuples.
lineArray: Array of unique strings.
"""
- for i in xrange(len(diffs)):
- text = []
- for char in diffs[i][1]:
- text.append(lineArray[ord(char)])
- diffs[i] = (diffs[i][0], "".join(text))
+ for i in xrange(len(diffs)):
+ text = []
+ for char in diffs[i][1]:
+ text.append(lineArray[ord(char)])
+ diffs[i] = (diffs[i][0], "".join(text))
- def diff_commonPrefix(self, text1, text2):
- """Determine the common prefix of two strings.
+ def diff_commonPrefix(self, text1, text2):
+ """Determine the common prefix of two strings.
Args:
text1: First string.
@@ -467,26 +471,26 @@ class diff_match_patch:
Returns:
The number of characters common to the start of each string.
"""
- # Quick check for common null cases.
- if not text1 or not text2 or text1[0] != text2[0]:
- return 0
- # Binary search.
- # Performance analysis: https://neil.fraser.name/news/2007/10/09/
- pointermin = 0
- pointermax = min(len(text1), len(text2))
- pointermid = pointermax
- pointerstart = 0
- while pointermin < pointermid:
- if text1[pointerstart:pointermid] == text2[pointerstart:pointermid]:
- pointermin = pointermid
- pointerstart = pointermin
- else:
- pointermax = pointermid
- pointermid = (pointermax - pointermin) // 2 + pointermin
- return pointermid
+ # Quick check for common null cases.
+ if not text1 or not text2 or text1[0] != text2[0]:
+ return 0
+ # Binary search.
+ # Performance analysis: https://neil.fraser.name/news/2007/10/09/
+ pointermin = 0
+ pointermax = min(len(text1), len(text2))
+ pointermid = pointermax
+ pointerstart = 0
+ while pointermin < pointermid:
+ if text1[pointerstart:pointermid] == text2[pointerstart:pointermid]:
+ pointermin = pointermid
+ pointerstart = pointermin
+ else:
+ pointermax = pointermid
+ pointermid = (pointermax - pointermin) // 2 + pointermin
+ return pointermid
- def diff_commonSuffix(self, text1, text2):
- """Determine the common suffix of two strings.
+ def diff_commonSuffix(self, text1, text2):
+ """Determine the common suffix of two strings.
Args:
text1: First string.
@@ -495,27 +499,29 @@ class diff_match_patch:
Returns:
The number of characters common to the end of each string.
"""
- # Quick check for common null cases.
- if not text1 or not text2 or text1[-1] != text2[-1]:
- return 0
- # Binary search.
- # Performance analysis: https://neil.fraser.name/news/2007/10/09/
- pointermin = 0
- pointermax = min(len(text1), len(text2))
- pointermid = pointermax
- pointerend = 0
- while pointermin < pointermid:
- if (text1[-pointermid:len(text1) - pointerend] ==
- text2[-pointermid:len(text2) - pointerend]):
- pointermin = pointermid
- pointerend = pointermin
- else:
- pointermax = pointermid
- pointermid = (pointermax - pointermin) // 2 + pointermin
- return pointermid
+ # Quick check for common null cases.
+ if not text1 or not text2 or text1[-1] != text2[-1]:
+ return 0
+ # Binary search.
+ # Performance analysis: https://neil.fraser.name/news/2007/10/09/
+ pointermin = 0
+ pointermax = min(len(text1), len(text2))
+ pointermid = pointermax
+ pointerend = 0
+ while pointermin < pointermid:
+ if (
+ text1[-pointermid : len(text1) - pointerend]
+ == text2[-pointermid : len(text2) - pointerend]
+ ):
+ pointermin = pointermid
+ pointerend = pointermin
+ else:
+ pointermax = pointermid
+ pointermid = (pointermax - pointermin) // 2 + pointermin
+ return pointermid
- def diff_commonOverlap(self, text1, text2):
- """Determine if the suffix of one string is the prefix of another.
+ def diff_commonOverlap(self, text1, text2):
+ """Determine if the suffix of one string is the prefix of another.
Args:
text1 First string.
@@ -525,39 +531,39 @@ class diff_match_patch:
The number of characters common to the end of the first
string and the start of the second string.
"""
- # Cache the text lengths to prevent multiple calls.
- text1_length = len(text1)
- text2_length = len(text2)
- # Eliminate the null case.
- if text1_length == 0 or text2_length == 0:
- return 0
- # Truncate the longer string.
- if text1_length > text2_length:
- text1 = text1[-text2_length:]
- elif text1_length < text2_length:
- text2 = text2[:text1_length]
- text_length = min(text1_length, text2_length)
- # Quick check for the worst case.
- if text1 == text2:
- return text_length
-
- # Start by looking for a single character match
- # and increase length until no match is found.
- # Performance analysis: https://neil.fraser.name/news/2010/11/04/
- best = 0
- length = 1
- while True:
- pattern = text1[-length:]
- found = text2.find(pattern)
- if found == -1:
- return best
- length += found
- if found == 0 or text1[-length:] == text2[:length]:
- best = length
- length += 1
+ # Cache the text lengths to prevent multiple calls.
+ text1_length = len(text1)
+ text2_length = len(text2)
+ # Eliminate the null case.
+ if text1_length == 0 or text2_length == 0:
+ return 0
+ # Truncate the longer string.
+ if text1_length > text2_length:
+ text1 = text1[-text2_length:]
+ elif text1_length < text2_length:
+ text2 = text2[:text1_length]
+ text_length = min(text1_length, text2_length)
+ # Quick check for the worst case.
+ if text1 == text2:
+ return text_length
+
+ # Start by looking for a single character match
+ # and increase length until no match is found.
+ # Performance analysis: https://neil.fraser.name/news/2010/11/04/
+ best = 0
+ length = 1
+ while True:
+ pattern = text1[-length:]
+ found = text2.find(pattern)
+ if found == -1:
+ return best
+ length += found
+ if found == 0 or text1[-length:] == text2[:length]:
+ best = length
+ length += 1
- def diff_halfMatch(self, text1, text2):
- """Do the two texts share a substring which is at least half the length of
+ def diff_halfMatch(self, text1, text2):
+ """Do the two texts share a substring which is at least half the length of
the longer text?
This speedup can produce non-minimal diffs.
@@ -570,18 +576,18 @@ class diff_match_patch:
the prefix of text2, the suffix of text2 and the common middle. Or None
if there was no match.
"""
- if self.Diff_Timeout <= 0:
- # Don't risk returning a non-optimal diff if we have unlimited time.
- return None
- if len(text1) > len(text2):
- (longtext, shorttext) = (text1, text2)
- else:
- (shorttext, longtext) = (text1, text2)
- if len(longtext) < 4 or len(shorttext) * 2 < len(longtext):
- return None # Pointless.
+ if self.Diff_Timeout <= 0:
+ # Don't risk returning a non-optimal diff if we have unlimited time.
+ return None
+ if len(text1) > len(text2):
+ (longtext, shorttext) = (text1, text2)
+ else:
+ (shorttext, longtext) = (text1, text2)
+ if len(longtext) < 4 or len(shorttext) * 2 < len(longtext):
+ return None # Pointless.
- def diff_halfMatchI(longtext, shorttext, i):
- """Does a substring of shorttext exist within longtext such that the
+ def diff_halfMatchI(longtext, shorttext, i):
+ """Does a substring of shorttext exist within longtext such that the
substring is at least half the length of longtext?
Closure, but does not reference any external variables.
@@ -595,148 +601,181 @@ class diff_match_patch:
longtext, the prefix of shorttext, the suffix of shorttext and the
common middle. Or None if there was no match.
"""
- seed = longtext[i:i + len(longtext) // 4]
- best_common = ''
- j = shorttext.find(seed)
- while j != -1:
- prefixLength = self.diff_commonPrefix(longtext[i:], shorttext[j:])
- suffixLength = self.diff_commonSuffix(longtext[:i], shorttext[:j])
- if len(best_common) < suffixLength + prefixLength:
- best_common = (shorttext[j - suffixLength:j] +
- shorttext[j:j + prefixLength])
- best_longtext_a = longtext[:i - suffixLength]
- best_longtext_b = longtext[i + prefixLength:]
- best_shorttext_a = shorttext[:j - suffixLength]
- best_shorttext_b = shorttext[j + prefixLength:]
- j = shorttext.find(seed, j + 1)
-
- if len(best_common) * 2 >= len(longtext):
- return (best_longtext_a, best_longtext_b,
- best_shorttext_a, best_shorttext_b, best_common)
- else:
- return None
-
- # First check if the second quarter is the seed for a half-match.
- hm1 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 3) // 4)
- # Check again based on the third quarter.
- hm2 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 1) // 2)
- if not hm1 and not hm2:
- return None
- elif not hm2:
- hm = hm1
- elif not hm1:
- hm = hm2
- else:
- # Both matched. Select the longest.
- if len(hm1[4]) > len(hm2[4]):
- hm = hm1
- else:
- hm = hm2
-
- # A half-match was found, sort out the return data.
- if len(text1) > len(text2):
- (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
- else:
- (text2_a, text2_b, text1_a, text1_b, mid_common) = hm
- return (text1_a, text1_b, text2_a, text2_b, mid_common)
+ seed = longtext[i : i + len(longtext) // 4]
+ best_common = ""
+ j = shorttext.find(seed)
+ while j != -1:
+ prefixLength = self.diff_commonPrefix(longtext[i:], shorttext[j:])
+ suffixLength = self.diff_commonSuffix(longtext[:i], shorttext[:j])
+ if len(best_common) < suffixLength + prefixLength:
+ best_common = (
+ shorttext[j - suffixLength : j]
+ + shorttext[j : j + prefixLength]
+ )
+ best_longtext_a = longtext[: i - suffixLength]
+ best_longtext_b = longtext[i + prefixLength :]
+ best_shorttext_a = shorttext[: j - suffixLength]
+ best_shorttext_b = shorttext[j + prefixLength :]
+ j = shorttext.find(seed, j + 1)
+
+ if len(best_common) * 2 >= len(longtext):
+ return (
+ best_longtext_a,
+ best_longtext_b,
+ best_shorttext_a,
+ best_shorttext_b,
+ best_common,
+ )
+ else:
+ return None
+
+ # First check if the second quarter is the seed for a half-match.
+ hm1 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 3) // 4)
+ # Check again based on the third quarter.
+ hm2 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 1) // 2)
+ if not hm1 and not hm2:
+ return None
+ elif not hm2:
+ hm = hm1
+ elif not hm1:
+ hm = hm2
+ else:
+ # Both matched. Select the longest.
+ if len(hm1[4]) > len(hm2[4]):
+ hm = hm1
+ else:
+ hm = hm2
- def diff_cleanupSemantic(self, diffs):
- """Reduce the number of edits by eliminating semantically trivial
+ # A half-match was found, sort out the return data.
+ if len(text1) > len(text2):
+ (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
+ else:
+ (text2_a, text2_b, text1_a, text1_b, mid_common) = hm
+ return (text1_a, text1_b, text2_a, text2_b, mid_common)
+
+ def diff_cleanupSemantic(self, diffs):
+ """Reduce the number of edits by eliminating semantically trivial
equalities.
Args:
diffs: Array of diff tuples.
"""
- changes = False
- equalities = [] # Stack of indices where equalities are found.
- lastEquality = None # Always equal to diffs[equalities[-1]][1]
- pointer = 0 # Index of current position.
- # Number of chars that changed prior to the equality.
- length_insertions1, length_deletions1 = 0, 0
- # Number of chars that changed after the equality.
- length_insertions2, length_deletions2 = 0, 0
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
- equalities.append(pointer)
- length_insertions1, length_insertions2 = length_insertions2, 0
- length_deletions1, length_deletions2 = length_deletions2, 0
- lastEquality = diffs[pointer][1]
- else: # An insertion or deletion.
- if diffs[pointer][0] == self.DIFF_INSERT:
- length_insertions2 += len(diffs[pointer][1])
- else:
- length_deletions2 += len(diffs[pointer][1])
- # Eliminate an equality that is smaller or equal to the edits on both
- # sides of it.
- if (lastEquality and (len(lastEquality) <=
- max(length_insertions1, length_deletions1)) and
- (len(lastEquality) <= max(length_insertions2, length_deletions2))):
- # Duplicate record.
- diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
- # Change second copy to insert.
- diffs[equalities[-1] + 1] = (self.DIFF_INSERT,
- diffs[equalities[-1] + 1][1])
- # Throw away the equality we just deleted.
- equalities.pop()
- # Throw away the previous equality (it needs to be reevaluated).
- if len(equalities):
- equalities.pop()
- if len(equalities):
- pointer = equalities[-1]
- else:
- pointer = -1
- # Reset the counters.
- length_insertions1, length_deletions1 = 0, 0
- length_insertions2, length_deletions2 = 0, 0
- lastEquality = None
- changes = True
- pointer += 1
-
- # Normalize the diff.
- if changes:
- self.diff_cleanupMerge(diffs)
- self.diff_cleanupSemanticLossless(diffs)
-
- # Find any overlaps between deletions and insertions.
- # e.g: <del>abcxxx</del><ins>xxxdef</ins>
- # -> <del>abc</del>xxx<ins>def</ins>
- # e.g: <del>xxxabc</del><ins>defxxx</ins>
- # -> <ins>def</ins>xxx<del>abc</del>
- # Only extract an overlap if it is as big as the edit ahead or behind it.
- pointer = 1
- while pointer < len(diffs):
- if (diffs[pointer - 1][0] == self.DIFF_DELETE and
- diffs[pointer][0] == self.DIFF_INSERT):
- deletion = diffs[pointer - 1][1]
- insertion = diffs[pointer][1]
- overlap_length1 = self.diff_commonOverlap(deletion, insertion)
- overlap_length2 = self.diff_commonOverlap(insertion, deletion)
- if overlap_length1 >= overlap_length2:
- if (overlap_length1 >= len(deletion) / 2.0 or
- overlap_length1 >= len(insertion) / 2.0):
- # Overlap found. Insert an equality and trim the surrounding edits.
- diffs.insert(pointer, (self.DIFF_EQUAL,
- insertion[:overlap_length1]))
- diffs[pointer - 1] = (self.DIFF_DELETE,
- deletion[:len(deletion) - overlap_length1])
- diffs[pointer + 1] = (self.DIFF_INSERT,
- insertion[overlap_length1:])
+ changes = False
+ equalities = [] # Stack of indices where equalities are found.
+ lastEquality = None # Always equal to diffs[equalities[-1]][1]
+ pointer = 0 # Index of current position.
+ # Number of chars that changed prior to the equality.
+ length_insertions1, length_deletions1 = 0, 0
+ # Number of chars that changed after the equality.
+ length_insertions2, length_deletions2 = 0, 0
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
+ equalities.append(pointer)
+ length_insertions1, length_insertions2 = length_insertions2, 0
+ length_deletions1, length_deletions2 = length_deletions2, 0
+ lastEquality = diffs[pointer][1]
+ else: # An insertion or deletion.
+ if diffs[pointer][0] == self.DIFF_INSERT:
+ length_insertions2 += len(diffs[pointer][1])
+ else:
+ length_deletions2 += len(diffs[pointer][1])
+ # Eliminate an equality that is smaller or equal to the edits on both
+ # sides of it.
+ if (
+ lastEquality
+ and (
+ len(lastEquality) <= max(length_insertions1, length_deletions1)
+ )
+ and (
+ len(lastEquality) <= max(length_insertions2, length_deletions2)
+ )
+ ):
+ # Duplicate record.
+ diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
+ # Change second copy to insert.
+ diffs[equalities[-1] + 1] = (
+ self.DIFF_INSERT,
+ diffs[equalities[-1] + 1][1],
+ )
+ # Throw away the equality we just deleted.
+ equalities.pop()
+ # Throw away the previous equality (it needs to be reevaluated).
+ if len(equalities):
+ equalities.pop()
+ if len(equalities):
+ pointer = equalities[-1]
+ else:
+ pointer = -1
+ # Reset the counters.
+ length_insertions1, length_deletions1 = 0, 0
+ length_insertions2, length_deletions2 = 0, 0
+ lastEquality = None
+ changes = True
pointer += 1
- else:
- if (overlap_length2 >= len(deletion) / 2.0 or
- overlap_length2 >= len(insertion) / 2.0):
- # Reverse overlap found.
- # Insert an equality and swap and trim the surrounding edits.
- diffs.insert(pointer, (self.DIFF_EQUAL, deletion[:overlap_length2]))
- diffs[pointer - 1] = (self.DIFF_INSERT,
- insertion[:len(insertion) - overlap_length2])
- diffs[pointer + 1] = (self.DIFF_DELETE, deletion[overlap_length2:])
+
+ # Normalize the diff.
+ if changes:
+ self.diff_cleanupMerge(diffs)
+ self.diff_cleanupSemanticLossless(diffs)
+
+ # Find any overlaps between deletions and insertions.
+ # e.g: <del>abcxxx</del><ins>xxxdef</ins>
+ # -> <del>abc</del>xxx<ins>def</ins>
+ # e.g: <del>xxxabc</del><ins>defxxx</ins>
+ # -> <ins>def</ins>xxx<del>abc</del>
+ # Only extract an overlap if it is as big as the edit ahead or behind it.
+ pointer = 1
+ while pointer < len(diffs):
+ if (
+ diffs[pointer - 1][0] == self.DIFF_DELETE
+ and diffs[pointer][0] == self.DIFF_INSERT
+ ):
+ deletion = diffs[pointer - 1][1]
+ insertion = diffs[pointer][1]
+ overlap_length1 = self.diff_commonOverlap(deletion, insertion)
+ overlap_length2 = self.diff_commonOverlap(insertion, deletion)
+ if overlap_length1 >= overlap_length2:
+ if (
+ overlap_length1 >= len(deletion) / 2.0
+ or overlap_length1 >= len(insertion) / 2.0
+ ):
+ # Overlap found. Insert an equality and trim the surrounding edits.
+ diffs.insert(
+ pointer, (self.DIFF_EQUAL, insertion[:overlap_length1])
+ )
+ diffs[pointer - 1] = (
+ self.DIFF_DELETE,
+ deletion[: len(deletion) - overlap_length1],
+ )
+ diffs[pointer + 1] = (
+ self.DIFF_INSERT,
+ insertion[overlap_length1:],
+ )
+ pointer += 1
+ else:
+ if (
+ overlap_length2 >= len(deletion) / 2.0
+ or overlap_length2 >= len(insertion) / 2.0
+ ):
+ # Reverse overlap found.
+ # Insert an equality and swap and trim the surrounding edits.
+ diffs.insert(
+ pointer, (self.DIFF_EQUAL, deletion[:overlap_length2])
+ )
+ diffs[pointer - 1] = (
+ self.DIFF_INSERT,
+ insertion[: len(insertion) - overlap_length2],
+ )
+ diffs[pointer + 1] = (
+ self.DIFF_DELETE,
+ deletion[overlap_length2:],
+ )
+ pointer += 1
+ pointer += 1
pointer += 1
- pointer += 1
- pointer += 1
- def diff_cleanupSemanticLossless(self, diffs):
- """Look for single edits surrounded on both sides by equalities
+ def diff_cleanupSemanticLossless(self, diffs):
+ """Look for single edits surrounded on both sides by equalities
which can be shifted sideways to align the edit to a word boundary.
e.g: The c<ins>at c</ins>ame. -> The <ins>cat </ins>came.
@@ -744,8 +783,8 @@ class diff_match_patch:
diffs: Array of diff tuples.
"""
- def diff_cleanupSemanticScore(one, two):
- """Given two strings, compute a score representing whether the
+ def diff_cleanupSemanticScore(one, two):
+ """Given two strings, compute a score representing whether the
internal boundary falls on logical boundaries.
Scores range from 6 (best) to 0 (worst).
Closure, but does not reference any external variables.
@@ -757,277 +796,306 @@ class diff_match_patch:
Returns:
The score.
"""
- if not one or not two:
- # Edges are the best.
- return 6
-
- # Each port of this function behaves slightly differently due to
- # subtle differences in each language's definition of things like
- # 'whitespace'. Since this function's purpose is largely cosmetic,
- # the choice has been made to use each language's native features
- # rather than force total conformity.
- char1 = one[-1]
- char2 = two[0]
- nonAlphaNumeric1 = not char1.isalnum()
- nonAlphaNumeric2 = not char2.isalnum()
- whitespace1 = nonAlphaNumeric1 and char1.isspace()
- whitespace2 = nonAlphaNumeric2 and char2.isspace()
- lineBreak1 = whitespace1 and (char1 == "\r" or char1 == "\n")
- lineBreak2 = whitespace2 and (char2 == "\r" or char2 == "\n")
- blankLine1 = lineBreak1 and self.BLANKLINEEND.search(one)
- blankLine2 = lineBreak2 and self.BLANKLINESTART.match(two)
-
- if blankLine1 or blankLine2:
- # Five points for blank lines.
- return 5
- elif lineBreak1 or lineBreak2:
- # Four points for line breaks.
- return 4
- elif nonAlphaNumeric1 and not whitespace1 and whitespace2:
- # Three points for end of sentences.
- return 3
- elif whitespace1 or whitespace2:
- # Two points for whitespace.
- return 2
- elif nonAlphaNumeric1 or nonAlphaNumeric2:
- # One point for non-alphanumeric.
- return 1
- return 0
-
- pointer = 1
- # Intentionally ignore the first and last element (don't need checking).
- while pointer < len(diffs) - 1:
- if (diffs[pointer - 1][0] == self.DIFF_EQUAL and
- diffs[pointer + 1][0] == self.DIFF_EQUAL):
- # This is a single edit surrounded by equalities.
- equality1 = diffs[pointer - 1][1]
- edit = diffs[pointer][1]
- equality2 = diffs[pointer + 1][1]
-
- # First, shift the edit as far left as possible.
- commonOffset = self.diff_commonSuffix(equality1, edit)
- if commonOffset:
- commonString = edit[-commonOffset:]
- equality1 = equality1[:-commonOffset]
- edit = commonString + edit[:-commonOffset]
- equality2 = commonString + equality2
-
- # Second, step character by character right, looking for the best fit.
- bestEquality1 = equality1
- bestEdit = edit
- bestEquality2 = equality2
- bestScore = (diff_cleanupSemanticScore(equality1, edit) +
- diff_cleanupSemanticScore(edit, equality2))
- while edit and equality2 and edit[0] == equality2[0]:
- equality1 += edit[0]
- edit = edit[1:] + equality2[0]
- equality2 = equality2[1:]
- score = (diff_cleanupSemanticScore(equality1, edit) +
- diff_cleanupSemanticScore(edit, equality2))
- # The >= encourages trailing rather than leading whitespace on edits.
- if score >= bestScore:
- bestScore = score
- bestEquality1 = equality1
- bestEdit = edit
- bestEquality2 = equality2
-
- if diffs[pointer - 1][1] != bestEquality1:
- # We have an improvement, save it back to the diff.
- if bestEquality1:
- diffs[pointer - 1] = (diffs[pointer - 1][0], bestEquality1)
- else:
- del diffs[pointer - 1]
- pointer -= 1
- diffs[pointer] = (diffs[pointer][0], bestEdit)
- if bestEquality2:
- diffs[pointer + 1] = (diffs[pointer + 1][0], bestEquality2)
- else:
- del diffs[pointer + 1]
- pointer -= 1
- pointer += 1
-
- # Define some regex patterns for matching boundaries.
- BLANKLINEEND = re.compile(r"\n\r?\n$")
- BLANKLINESTART = re.compile(r"^\r?\n\r?\n")
+ if not one or not two:
+ # Edges are the best.
+ return 6
+
+ # Each port of this function behaves slightly differently due to
+ # subtle differences in each language's definition of things like
+ # 'whitespace'. Since this function's purpose is largely cosmetic,
+ # the choice has been made to use each language's native features
+ # rather than force total conformity.
+ char1 = one[-1]
+ char2 = two[0]
+ nonAlphaNumeric1 = not char1.isalnum()
+ nonAlphaNumeric2 = not char2.isalnum()
+ whitespace1 = nonAlphaNumeric1 and char1.isspace()
+ whitespace2 = nonAlphaNumeric2 and char2.isspace()
+ lineBreak1 = whitespace1 and (char1 == "\r" or char1 == "\n")
+ lineBreak2 = whitespace2 and (char2 == "\r" or char2 == "\n")
+ blankLine1 = lineBreak1 and self.BLANKLINEEND.search(one)
+ blankLine2 = lineBreak2 and self.BLANKLINESTART.match(two)
+
+ if blankLine1 or blankLine2:
+ # Five points for blank lines.
+ return 5
+ elif lineBreak1 or lineBreak2:
+ # Four points for line breaks.
+ return 4
+ elif nonAlphaNumeric1 and not whitespace1 and whitespace2:
+ # Three points for end of sentences.
+ return 3
+ elif whitespace1 or whitespace2:
+ # Two points for whitespace.
+ return 2
+ elif nonAlphaNumeric1 or nonAlphaNumeric2:
+ # One point for non-alphanumeric.
+ return 1
+ return 0
+
+ pointer = 1
+ # Intentionally ignore the first and last element (don't need checking).
+ while pointer < len(diffs) - 1:
+ if (
+ diffs[pointer - 1][0] == self.DIFF_EQUAL
+ and diffs[pointer + 1][0] == self.DIFF_EQUAL
+ ):
+ # This is a single edit surrounded by equalities.
+ equality1 = diffs[pointer - 1][1]
+ edit = diffs[pointer][1]
+ equality2 = diffs[pointer + 1][1]
+
+ # First, shift the edit as far left as possible.
+ commonOffset = self.diff_commonSuffix(equality1, edit)
+ if commonOffset:
+ commonString = edit[-commonOffset:]
+ equality1 = equality1[:-commonOffset]
+ edit = commonString + edit[:-commonOffset]
+ equality2 = commonString + equality2
+
+ # Second, step character by character right, looking for the best fit.
+ bestEquality1 = equality1
+ bestEdit = edit
+ bestEquality2 = equality2
+ bestScore = diff_cleanupSemanticScore(
+ equality1, edit
+ ) + diff_cleanupSemanticScore(edit, equality2)
+ while edit and equality2 and edit[0] == equality2[0]:
+ equality1 += edit[0]
+ edit = edit[1:] + equality2[0]
+ equality2 = equality2[1:]
+ score = diff_cleanupSemanticScore(
+ equality1, edit
+ ) + diff_cleanupSemanticScore(edit, equality2)
+ # The >= encourages trailing rather than leading whitespace on edits.
+ if score >= bestScore:
+ bestScore = score
+ bestEquality1 = equality1
+ bestEdit = edit
+ bestEquality2 = equality2
+
+ if diffs[pointer - 1][1] != bestEquality1:
+ # We have an improvement, save it back to the diff.
+ if bestEquality1:
+ diffs[pointer - 1] = (diffs[pointer - 1][0], bestEquality1)
+ else:
+ del diffs[pointer - 1]
+ pointer -= 1
+ diffs[pointer] = (diffs[pointer][0], bestEdit)
+ if bestEquality2:
+ diffs[pointer + 1] = (diffs[pointer + 1][0], bestEquality2)
+ else:
+ del diffs[pointer + 1]
+ pointer -= 1
+ pointer += 1
+
+ # Define some regex patterns for matching boundaries.
+ BLANKLINEEND = re.compile(r"\n\r?\n$")
+ BLANKLINESTART = re.compile(r"^\r?\n\r?\n")
- def diff_cleanupEfficiency(self, diffs):
- """Reduce the number of edits by eliminating operationally trivial
+ def diff_cleanupEfficiency(self, diffs):
+ """Reduce the number of edits by eliminating operationally trivial
equalities.
Args:
diffs: Array of diff tuples.
"""
- changes = False
- equalities = [] # Stack of indices where equalities are found.
- lastEquality = None # Always equal to diffs[equalities[-1]][1]
- pointer = 0 # Index of current position.
- pre_ins = False # Is there an insertion operation before the last equality.
- pre_del = False # Is there a deletion operation before the last equality.
- post_ins = False # Is there an insertion operation after the last equality.
- post_del = False # Is there a deletion operation after the last equality.
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
- if (len(diffs[pointer][1]) < self.Diff_EditCost and
- (post_ins or post_del)):
- # Candidate found.
- equalities.append(pointer)
- pre_ins = post_ins
- pre_del = post_del
- lastEquality = diffs[pointer][1]
- else:
- # Not a candidate, and can never become one.
- equalities = []
- lastEquality = None
-
- post_ins = post_del = False
- else: # An insertion or deletion.
- if diffs[pointer][0] == self.DIFF_DELETE:
- post_del = True
- else:
- post_ins = True
-
- # Five types to be split:
- # <ins>A</ins><del>B</del>XY<ins>C</ins><del>D</del>
- # <ins>A</ins>X<ins>C</ins><del>D</del>
- # <ins>A</ins><del>B</del>X<ins>C</ins>
- # <ins>A</del>X<ins>C</ins><del>D</del>
- # <ins>A</ins><del>B</del>X<del>C</del>
-
- if lastEquality and ((pre_ins and pre_del and post_ins and post_del) or
- ((len(lastEquality) < self.Diff_EditCost / 2) and
- (pre_ins + pre_del + post_ins + post_del) == 3)):
- # Duplicate record.
- diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
- # Change second copy to insert.
- diffs[equalities[-1] + 1] = (self.DIFF_INSERT,
- diffs[equalities[-1] + 1][1])
- equalities.pop() # Throw away the equality we just deleted.
- lastEquality = None
- if pre_ins and pre_del:
- # No changes made which could affect previous entry, keep going.
- post_ins = post_del = True
- equalities = []
- else:
- if len(equalities):
- equalities.pop() # Throw away the previous equality.
- if len(equalities):
- pointer = equalities[-1]
- else:
- pointer = -1
- post_ins = post_del = False
- changes = True
- pointer += 1
+ changes = False
+ equalities = [] # Stack of indices where equalities are found.
+ lastEquality = None # Always equal to diffs[equalities[-1]][1]
+ pointer = 0 # Index of current position.
+ pre_ins = False # Is there an insertion operation before the last equality.
+ pre_del = False # Is there a deletion operation before the last equality.
+ post_ins = False # Is there an insertion operation after the last equality.
+ post_del = False # Is there a deletion operation after the last equality.
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
+ if len(diffs[pointer][1]) < self.Diff_EditCost and (
+ post_ins or post_del
+ ):
+ # Candidate found.
+ equalities.append(pointer)
+ pre_ins = post_ins
+ pre_del = post_del
+ lastEquality = diffs[pointer][1]
+ else:
+ # Not a candidate, and can never become one.
+ equalities = []
+ lastEquality = None
+
+ post_ins = post_del = False
+ else: # An insertion or deletion.
+ if diffs[pointer][0] == self.DIFF_DELETE:
+ post_del = True
+ else:
+ post_ins = True
+
+ # Five types to be split:
+ # <ins>A</ins><del>B</del>XY<ins>C</ins><del>D</del>
+ # <ins>A</ins>X<ins>C</ins><del>D</del>
+ # <ins>A</ins><del>B</del>X<ins>C</ins>
+ # <ins>A</del>X<ins>C</ins><del>D</del>
+ # <ins>A</ins><del>B</del>X<del>C</del>
+
+ if lastEquality and (
+ (pre_ins and pre_del and post_ins and post_del)
+ or (
+ (len(lastEquality) < self.Diff_EditCost / 2)
+ and (pre_ins + pre_del + post_ins + post_del) == 3
+ )
+ ):
+ # Duplicate record.
+ diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
+ # Change second copy to insert.
+ diffs[equalities[-1] + 1] = (
+ self.DIFF_INSERT,
+ diffs[equalities[-1] + 1][1],
+ )
+ equalities.pop() # Throw away the equality we just deleted.
+ lastEquality = None
+ if pre_ins and pre_del:
+ # No changes made which could affect previous entry, keep going.
+ post_ins = post_del = True
+ equalities = []
+ else:
+ if len(equalities):
+ equalities.pop() # Throw away the previous equality.
+ if len(equalities):
+ pointer = equalities[-1]
+ else:
+ pointer = -1
+ post_ins = post_del = False
+ changes = True
+ pointer += 1
- if changes:
- self.diff_cleanupMerge(diffs)
+ if changes:
+ self.diff_cleanupMerge(diffs)
- def diff_cleanupMerge(self, diffs):
- """Reorder and merge like edit sections. Merge equalities.
+ def diff_cleanupMerge(self, diffs):
+ """Reorder and merge like edit sections. Merge equalities.
Any edit section can move as long as it doesn't cross an equality.
Args:
diffs: Array of diff tuples.
"""
- diffs.append((self.DIFF_EQUAL, '')) # Add a dummy entry at the end.
- pointer = 0
- count_delete = 0
- count_insert = 0
- text_delete = ''
- text_insert = ''
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_INSERT:
- count_insert += 1
- text_insert += diffs[pointer][1]
- pointer += 1
- elif diffs[pointer][0] == self.DIFF_DELETE:
- count_delete += 1
- text_delete += diffs[pointer][1]
- pointer += 1
- elif diffs[pointer][0] == self.DIFF_EQUAL:
- # Upon reaching an equality, check for prior redundancies.
- if count_delete + count_insert > 1:
- if count_delete != 0 and count_insert != 0:
- # Factor out any common prefixies.
- commonlength = self.diff_commonPrefix(text_insert, text_delete)
- if commonlength != 0:
- x = pointer - count_delete - count_insert - 1
- if x >= 0 and diffs[x][0] == self.DIFF_EQUAL:
- diffs[x] = (diffs[x][0], diffs[x][1] +
- text_insert[:commonlength])
- else:
- diffs.insert(0, (self.DIFF_EQUAL, text_insert[:commonlength]))
- pointer += 1
- text_insert = text_insert[commonlength:]
- text_delete = text_delete[commonlength:]
- # Factor out any common suffixes.
- commonlength = self.diff_commonSuffix(text_insert, text_delete)
- if commonlength != 0:
- diffs[pointer] = (diffs[pointer][0], text_insert[-commonlength:] +
- diffs[pointer][1])
- text_insert = text_insert[:-commonlength]
- text_delete = text_delete[:-commonlength]
- # Delete the offending records and add the merged ones.
- new_ops = []
- if len(text_delete) != 0:
- new_ops.append((self.DIFF_DELETE, text_delete))
- if len(text_insert) != 0:
- new_ops.append((self.DIFF_INSERT, text_insert))
- pointer -= count_delete + count_insert
- diffs[pointer : pointer + count_delete + count_insert] = new_ops
- pointer += len(new_ops) + 1
- elif pointer != 0 and diffs[pointer - 1][0] == self.DIFF_EQUAL:
- # Merge this equality with the previous one.
- diffs[pointer - 1] = (diffs[pointer - 1][0],
- diffs[pointer - 1][1] + diffs[pointer][1])
- del diffs[pointer]
- else:
- pointer += 1
-
- count_insert = 0
+ diffs.append((self.DIFF_EQUAL, "")) # Add a dummy entry at the end.
+ pointer = 0
count_delete = 0
- text_delete = ''
- text_insert = ''
-
- if diffs[-1][1] == '':
- diffs.pop() # Remove the dummy entry at the end.
+ count_insert = 0
+ text_delete = ""
+ text_insert = ""
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_INSERT:
+ count_insert += 1
+ text_insert += diffs[pointer][1]
+ pointer += 1
+ elif diffs[pointer][0] == self.DIFF_DELETE:
+ count_delete += 1
+ text_delete += diffs[pointer][1]
+ pointer += 1
+ elif diffs[pointer][0] == self.DIFF_EQUAL:
+ # Upon reaching an equality, check for prior redundancies.
+ if count_delete + count_insert > 1:
+ if count_delete != 0 and count_insert != 0:
+ # Factor out any common prefixies.
+ commonlength = self.diff_commonPrefix(text_insert, text_delete)
+ if commonlength != 0:
+ x = pointer - count_delete - count_insert - 1
+ if x >= 0 and diffs[x][0] == self.DIFF_EQUAL:
+ diffs[x] = (
+ diffs[x][0],
+ diffs[x][1] + text_insert[:commonlength],
+ )
+ else:
+ diffs.insert(
+ 0, (self.DIFF_EQUAL, text_insert[:commonlength])
+ )
+ pointer += 1
+ text_insert = text_insert[commonlength:]
+ text_delete = text_delete[commonlength:]
+ # Factor out any common suffixes.
+ commonlength = self.diff_commonSuffix(text_insert, text_delete)
+ if commonlength != 0:
+ diffs[pointer] = (
+ diffs[pointer][0],
+ text_insert[-commonlength:] + diffs[pointer][1],
+ )
+ text_insert = text_insert[:-commonlength]
+ text_delete = text_delete[:-commonlength]
+ # Delete the offending records and add the merged ones.
+ new_ops = []
+ if len(text_delete) != 0:
+ new_ops.append((self.DIFF_DELETE, text_delete))
+ if len(text_insert) != 0:
+ new_ops.append((self.DIFF_INSERT, text_insert))
+ pointer -= count_delete + count_insert
+ diffs[pointer : pointer + count_delete + count_insert] = new_ops
+ pointer += len(new_ops) + 1
+ elif pointer != 0 and diffs[pointer - 1][0] == self.DIFF_EQUAL:
+ # Merge this equality with the previous one.
+ diffs[pointer - 1] = (
+ diffs[pointer - 1][0],
+ diffs[pointer - 1][1] + diffs[pointer][1],
+ )
+ del diffs[pointer]
+ else:
+ pointer += 1
+
+ count_insert = 0
+ count_delete = 0
+ text_delete = ""
+ text_insert = ""
+
+ if diffs[-1][1] == "":
+ diffs.pop() # Remove the dummy entry at the end.
+
+ # Second pass: look for single edits surrounded on both sides by equalities
+ # which can be shifted sideways to eliminate an equality.
+ # e.g: A<ins>BA</ins>C -> <ins>AB</ins>AC
+ changes = False
+ pointer = 1
+ # Intentionally ignore the first and last element (don't need checking).
+ while pointer < len(diffs) - 1:
+ if (
+ diffs[pointer - 1][0] == self.DIFF_EQUAL
+ and diffs[pointer + 1][0] == self.DIFF_EQUAL
+ ):
+ # This is a single edit surrounded by equalities.
+ if diffs[pointer][1].endswith(diffs[pointer - 1][1]):
+ # Shift the edit over the previous equality.
+ if diffs[pointer - 1][1] != "":
+ diffs[pointer] = (
+ diffs[pointer][0],
+ diffs[pointer - 1][1]
+ + diffs[pointer][1][: -len(diffs[pointer - 1][1])],
+ )
+ diffs[pointer + 1] = (
+ diffs[pointer + 1][0],
+ diffs[pointer - 1][1] + diffs[pointer + 1][1],
+ )
+ del diffs[pointer - 1]
+ changes = True
+ elif diffs[pointer][1].startswith(diffs[pointer + 1][1]):
+ # Shift the edit over the next equality.
+ diffs[pointer - 1] = (
+ diffs[pointer - 1][0],
+ diffs[pointer - 1][1] + diffs[pointer + 1][1],
+ )
+ diffs[pointer] = (
+ diffs[pointer][0],
+ diffs[pointer][1][len(diffs[pointer + 1][1]) :]
+ + diffs[pointer + 1][1],
+ )
+ del diffs[pointer + 1]
+ changes = True
+ pointer += 1
- # Second pass: look for single edits surrounded on both sides by equalities
- # which can be shifted sideways to eliminate an equality.
- # e.g: A<ins>BA</ins>C -> <ins>AB</ins>AC
- changes = False
- pointer = 1
- # Intentionally ignore the first and last element (don't need checking).
- while pointer < len(diffs) - 1:
- if (diffs[pointer - 1][0] == self.DIFF_EQUAL and
- diffs[pointer + 1][0] == self.DIFF_EQUAL):
- # This is a single edit surrounded by equalities.
- if diffs[pointer][1].endswith(diffs[pointer - 1][1]):
- # Shift the edit over the previous equality.
- if diffs[pointer - 1][1] != "":
- diffs[pointer] = (diffs[pointer][0],
- diffs[pointer - 1][1] +
- diffs[pointer][1][:-len(diffs[pointer - 1][1])])
- diffs[pointer + 1] = (diffs[pointer + 1][0],
- diffs[pointer - 1][1] + diffs[pointer + 1][1])
- del diffs[pointer - 1]
- changes = True
- elif diffs[pointer][1].startswith(diffs[pointer + 1][1]):
- # Shift the edit over the next equality.
- diffs[pointer - 1] = (diffs[pointer - 1][0],
- diffs[pointer - 1][1] + diffs[pointer + 1][1])
- diffs[pointer] = (diffs[pointer][0],
- diffs[pointer][1][len(diffs[pointer + 1][1]):] +
- diffs[pointer + 1][1])
- del diffs[pointer + 1]
- changes = True
- pointer += 1
-
- # If shifts were made, the diff needs reordering and another shift sweep.
- if changes:
- self.diff_cleanupMerge(diffs)
+ # If shifts were made, the diff needs reordering and another shift sweep.
+ if changes:
+ self.diff_cleanupMerge(diffs)
- def diff_xIndex(self, diffs, loc):
- """loc is a location in text1, compute and return the equivalent location
+ def diff_xIndex(self, diffs, loc):
+ """loc is a location in text1, compute and return the equivalent location
in text2. e.g. "The cat" vs "The big cat", 1->1, 5->8
Args:
@@ -1037,29 +1105,29 @@ class diff_match_patch:
Returns:
Location within text2.
"""
- chars1 = 0
- chars2 = 0
- last_chars1 = 0
- last_chars2 = 0
- for x in xrange(len(diffs)):
- (op, text) = diffs[x]
- if op != self.DIFF_INSERT: # Equality or deletion.
- chars1 += len(text)
- if op != self.DIFF_DELETE: # Equality or insertion.
- chars2 += len(text)
- if chars1 > loc: # Overshot the location.
- break
- last_chars1 = chars1
- last_chars2 = chars2
-
- if len(diffs) != x and diffs[x][0] == self.DIFF_DELETE:
- # The location was deleted.
- return last_chars2
- # Add the remaining len(character).
- return last_chars2 + (loc - last_chars1)
+ chars1 = 0
+ chars2 = 0
+ last_chars1 = 0
+ last_chars2 = 0
+ for x in xrange(len(diffs)):
+ (op, text) = diffs[x]
+ if op != self.DIFF_INSERT: # Equality or deletion.
+ chars1 += len(text)
+ if op != self.DIFF_DELETE: # Equality or insertion.
+ chars2 += len(text)
+ if chars1 > loc: # Overshot the location.
+ break
+ last_chars1 = chars1
+ last_chars2 = chars2
+
+ if len(diffs) != x and diffs[x][0] == self.DIFF_DELETE:
+ # The location was deleted.
+ return last_chars2
+ # Add the remaining len(character).
+ return last_chars2 + (loc - last_chars1)
- def diff_prettyHtml(self, diffs):
- """Convert a diff array into a pretty HTML report.
+ def diff_prettyHtml(self, diffs):
+ """Convert a diff array into a pretty HTML report.
Args:
diffs: Array of diff tuples.
@@ -1067,20 +1135,24 @@ class diff_match_patch:
Returns:
HTML representation.
"""
- html = []
- for (op, data) in diffs:
- text = (data.replace("&", "&amp;").replace("<", "&lt;")
- .replace(">", "&gt;").replace("\n", "&para;<br>"))
- if op == self.DIFF_INSERT:
- html.append("<ins style=\"background:#e6ffe6;\">%s</ins>" % text)
- elif op == self.DIFF_DELETE:
- html.append("<del style=\"background:#ffe6e6;\">%s</del>" % text)
- elif op == self.DIFF_EQUAL:
- html.append("<span>%s</span>" % text)
- return "".join(html)
+ html = []
+ for (op, data) in diffs:
+ text = (
+ data.replace("&", "&amp;")
+ .replace("<", "&lt;")
+ .replace(">", "&gt;")
+ .replace("\n", "&para;<br>")
+ )
+ if op == self.DIFF_INSERT:
+ html.append('<ins style="background:#e6ffe6;">%s</ins>' % text)
+ elif op == self.DIFF_DELETE:
+ html.append('<del style="background:#ffe6e6;">%s</del>' % text)
+ elif op == self.DIFF_EQUAL:
+ html.append("<span>%s</span>" % text)
+ return "".join(html)
- def diff_text1(self, diffs):
- """Compute and return the source text (all equalities and deletions).
+ def diff_text1(self, diffs):
+ """Compute and return the source text (all equalities and deletions).
Args:
diffs: Array of diff tuples.
@@ -1088,14 +1160,14 @@ class diff_match_patch:
Returns:
Source text.
"""
- text = []
- for (op, data) in diffs:
- if op != self.DIFF_INSERT:
- text.append(data)
- return "".join(text)
+ text = []
+ for (op, data) in diffs:
+ if op != self.DIFF_INSERT:
+ text.append(data)
+ return "".join(text)
- def diff_text2(self, diffs):
- """Compute and return the destination text (all equalities and insertions).
+ def diff_text2(self, diffs):
+ """Compute and return the destination text (all equalities and insertions).
Args:
diffs: Array of diff tuples.
@@ -1103,14 +1175,14 @@ class diff_match_patch:
Returns:
Destination text.
"""
- text = []
- for (op, data) in diffs:
- if op != self.DIFF_DELETE:
- text.append(data)
- return "".join(text)
+ text = []
+ for (op, data) in diffs:
+ if op != self.DIFF_DELETE:
+ text.append(data)
+ return "".join(text)
- def diff_levenshtein(self, diffs):
- """Compute the Levenshtein distance; the number of inserted, deleted or
+ def diff_levenshtein(self, diffs):
+ """Compute the Levenshtein distance; the number of inserted, deleted or
substituted characters.
Args:
@@ -1119,24 +1191,24 @@ class diff_match_patch:
Returns:
Number of changes.
"""
- levenshtein = 0
- insertions = 0
- deletions = 0
- for (op, data) in diffs:
- if op == self.DIFF_INSERT:
- insertions += len(data)
- elif op == self.DIFF_DELETE:
- deletions += len(data)
- elif op == self.DIFF_EQUAL:
- # A deletion and an insertion is one substitution.
- levenshtein += max(insertions, deletions)
+ levenshtein = 0
insertions = 0
deletions = 0
- levenshtein += max(insertions, deletions)
- return levenshtein
+ for (op, data) in diffs:
+ if op == self.DIFF_INSERT:
+ insertions += len(data)
+ elif op == self.DIFF_DELETE:
+ deletions += len(data)
+ elif op == self.DIFF_EQUAL:
+ # A deletion and an insertion is one substitution.
+ levenshtein += max(insertions, deletions)
+ insertions = 0
+ deletions = 0
+ levenshtein += max(insertions, deletions)
+ return levenshtein
- def diff_toDelta(self, diffs):
- """Crush the diff into an encoded string which describes the operations
+ def diff_toDelta(self, diffs):
+ """Crush the diff into an encoded string which describes the operations
required to transform text1 into text2.
E.g. =3\t-2\t+ing -> Keep 3 chars, delete 2 chars, insert 'ing'.
Operations are tab-separated. Inserted text is escaped using %xx notation.
@@ -1147,20 +1219,20 @@ class diff_match_patch:
Returns:
Delta text.
"""
- text = []
- for (op, data) in diffs:
- if op == self.DIFF_INSERT:
- # High ascii will raise UnicodeDecodeError. Use Unicode instead.
- data = data.encode("utf-8")
- text.append("+" + urllib.quote(data, "!~*'();/?:@&=+$,# "))
- elif op == self.DIFF_DELETE:
- text.append("-%d" % len(data))
- elif op == self.DIFF_EQUAL:
- text.append("=%d" % len(data))
- return "\t".join(text)
+ text = []
+ for (op, data) in diffs:
+ if op == self.DIFF_INSERT:
+ # High ascii will raise UnicodeDecodeError. Use Unicode instead.
+ data = data.encode("utf-8")
+ text.append("+" + urllib.quote(data, "!~*'();/?:@&=+$,# "))
+ elif op == self.DIFF_DELETE:
+ text.append("-%d" % len(data))
+ elif op == self.DIFF_EQUAL:
+ text.append("=%d" % len(data))
+ return "\t".join(text)
- def diff_fromDelta(self, text1, delta):
- """Given the original text1, and an encoded string which describes the
+ def diff_fromDelta(self, text1, delta):
+ """Given the original text1, and an encoded string which describes the
operations required to transform text1 into text2, compute the full diff.
Args:
@@ -1173,50 +1245,52 @@ class diff_match_patch:
Raises:
ValueError: If invalid input.
"""
- if type(delta) == unicode:
- # Deltas should be composed of a subset of ascii chars, Unicode not
- # required. If this encode raises UnicodeEncodeError, delta is invalid.
- delta = delta.encode("ascii")
- diffs = []
- pointer = 0 # Cursor in text1
- tokens = delta.split("\t")
- for token in tokens:
- if token == "":
- # Blank tokens are ok (from a trailing \t).
- continue
- # Each token begins with a one character parameter which specifies the
- # operation of this token (delete, insert, equality).
- param = token[1:]
- if token[0] == "+":
- param = urllib.unquote(param).decode("utf-8")
- diffs.append((self.DIFF_INSERT, param))
- elif token[0] == "-" or token[0] == "=":
- try:
- n = int(param)
- except ValueError:
- raise ValueError("Invalid number in diff_fromDelta: " + param)
- if n < 0:
- raise ValueError("Negative number in diff_fromDelta: " + param)
- text = text1[pointer : pointer + n]
- pointer += n
- if token[0] == "=":
- diffs.append((self.DIFF_EQUAL, text))
- else:
- diffs.append((self.DIFF_DELETE, text))
- else:
- # Anything else is an error.
- raise ValueError("Invalid diff operation in diff_fromDelta: " +
- token[0])
- if pointer != len(text1):
- raise ValueError(
- "Delta length (%d) does not equal source text length (%d)." %
- (pointer, len(text1)))
- return diffs
+ if type(delta) == unicode:
+ # Deltas should be composed of a subset of ascii chars, Unicode not
+ # required. If this encode raises UnicodeEncodeError, delta is invalid.
+ delta = delta.encode("ascii")
+ diffs = []
+ pointer = 0 # Cursor in text1
+ tokens = delta.split("\t")
+ for token in tokens:
+ if token == "":
+ # Blank tokens are ok (from a trailing \t).
+ continue
+ # Each token begins with a one character parameter which specifies the
+ # operation of this token (delete, insert, equality).
+ param = token[1:]
+ if token[0] == "+":
+ param = urllib.unquote(param).decode("utf-8")
+ diffs.append((self.DIFF_INSERT, param))
+ elif token[0] == "-" or token[0] == "=":
+ try:
+ n = int(param)
+ except ValueError:
+ raise ValueError("Invalid number in diff_fromDelta: " + param)
+ if n < 0:
+ raise ValueError("Negative number in diff_fromDelta: " + param)
+ text = text1[pointer : pointer + n]
+ pointer += n
+ if token[0] == "=":
+ diffs.append((self.DIFF_EQUAL, text))
+ else:
+ diffs.append((self.DIFF_DELETE, text))
+ else:
+ # Anything else is an error.
+ raise ValueError(
+ "Invalid diff operation in diff_fromDelta: " + token[0]
+ )
+ if pointer != len(text1):
+ raise ValueError(
+ "Delta length (%d) does not equal source text length (%d)."
+ % (pointer, len(text1))
+ )
+ return diffs
- # MATCH FUNCTIONS
+ # MATCH FUNCTIONS
- def match_main(self, text, pattern, loc):
- """Locate the best instance of 'pattern' in 'text' near 'loc'.
+ def match_main(self, text, pattern, loc):
+ """Locate the best instance of 'pattern' in 'text' near 'loc'.
Args:
text: The text to search.
@@ -1226,27 +1300,27 @@ class diff_match_patch:
Returns:
Best match index or -1.
"""
- # Check for null inputs.
- if text == None or pattern == None:
- raise ValueError("Null inputs. (match_main)")
-
- loc = max(0, min(loc, len(text)))
- if text == pattern:
- # Shortcut (potentially not guaranteed by the algorithm)
- return 0
- elif not text:
- # Nothing to match.
- return -1
- elif text[loc:loc + len(pattern)] == pattern:
- # Perfect match at the perfect spot! (Includes case of null pattern)
- return loc
- else:
- # Do a fuzzy compare.
- match = self.match_bitap(text, pattern, loc)
- return match
+ # Check for null inputs.
+ if text == None or pattern == None:
+ raise ValueError("Null inputs. (match_main)")
+
+ loc = max(0, min(loc, len(text)))
+ if text == pattern:
+ # Shortcut (potentially not guaranteed by the algorithm)
+ return 0
+ elif not text:
+ # Nothing to match.
+ return -1
+ elif text[loc : loc + len(pattern)] == pattern:
+ # Perfect match at the perfect spot! (Includes case of null pattern)
+ return loc
+ else:
+ # Do a fuzzy compare.
+ match = self.match_bitap(text, pattern, loc)
+ return match
- def match_bitap(self, text, pattern, loc):
- """Locate the best instance of 'pattern' in 'text' near 'loc' using the
+ def match_bitap(self, text, pattern, loc):
+ """Locate the best instance of 'pattern' in 'text' near 'loc' using the
Bitap algorithm.
Args:
@@ -1257,15 +1331,15 @@ class diff_match_patch:
Returns:
Best match index or -1.
"""
- # Python doesn't have a maxint limit, so ignore this check.
- #if self.Match_MaxBits != 0 and len(pattern) > self.Match_MaxBits:
- # raise ValueError("Pattern too long for this application.")
+ # Python doesn't have a maxint limit, so ignore this check.
+ # if self.Match_MaxBits != 0 and len(pattern) > self.Match_MaxBits:
+ # raise ValueError("Pattern too long for this application.")
- # Initialise the alphabet.
- s = self.match_alphabet(pattern)
+ # Initialise the alphabet.
+ s = self.match_alphabet(pattern)
- def match_bitapScore(e, x):
- """Compute and return the score for a match with e errors and x location.
+ def match_bitapScore(e, x):
+ """Compute and return the score for a match with e errors and x location.
Accesses loc and pattern through being a closure.
Args:
@@ -1275,84 +1349,87 @@ class diff_match_patch:
Returns:
Overall score for match (0.0 = good, 1.0 = bad).
"""
- accuracy = float(e) / len(pattern)
- proximity = abs(loc - x)
- if not self.Match_Distance:
- # Dodge divide by zero error.
- return proximity and 1.0 or accuracy
- return accuracy + (proximity / float(self.Match_Distance))
-
- # Highest score beyond which we give up.
- score_threshold = self.Match_Threshold
- # Is there a nearby exact match? (speedup)
- best_loc = text.find(pattern, loc)
- if best_loc != -1:
- score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
- # What about in the other direction? (speedup)
- best_loc = text.rfind(pattern, loc + len(pattern))
- if best_loc != -1:
- score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
-
- # Initialise the bit arrays.
- matchmask = 1 << (len(pattern) - 1)
- best_loc = -1
-
- bin_max = len(pattern) + len(text)
- # Empty initialization added to appease pychecker.
- last_rd = None
- for d in xrange(len(pattern)):
- # Scan for the best match each iteration allows for one more error.
- # Run a binary search to determine how far from 'loc' we can stray at
- # this error level.
- bin_min = 0
- bin_mid = bin_max
- while bin_min < bin_mid:
- if match_bitapScore(d, loc + bin_mid) <= score_threshold:
- bin_min = bin_mid
- else:
- bin_max = bin_mid
- bin_mid = (bin_max - bin_min) // 2 + bin_min
-
- # Use the result from this iteration as the maximum for the next.
- bin_max = bin_mid
- start = max(1, loc - bin_mid + 1)
- finish = min(loc + bin_mid, len(text)) + len(pattern)
-
- rd = [0] * (finish + 2)
- rd[finish + 1] = (1 << d) - 1
- for j in xrange(finish, start - 1, -1):
- if len(text) <= j - 1:
- # Out of range.
- charMatch = 0
- else:
- charMatch = s.get(text[j - 1], 0)
- if d == 0: # First pass: exact match.
- rd[j] = ((rd[j + 1] << 1) | 1) & charMatch
- else: # Subsequent passes: fuzzy match.
- rd[j] = (((rd[j + 1] << 1) | 1) & charMatch) | (
- ((last_rd[j + 1] | last_rd[j]) << 1) | 1) | last_rd[j + 1]
- if rd[j] & matchmask:
- score = match_bitapScore(d, j - 1)
- # This match will almost certainly be better than any existing match.
- # But check anyway.
- if score <= score_threshold:
- # Told you so.
- score_threshold = score
- best_loc = j - 1
- if best_loc > loc:
- # When passing loc, don't exceed our current distance from loc.
- start = max(1, 2 * loc - best_loc)
- else:
- # Already passed loc, downhill from here on in.
- break
- # No hope for a (better) match at greater error levels.
- if match_bitapScore(d + 1, loc) > score_threshold:
- break
- last_rd = rd
- return best_loc
+ accuracy = float(e) / len(pattern)
+ proximity = abs(loc - x)
+ if not self.Match_Distance:
+ # Dodge divide by zero error.
+ return proximity and 1.0 or accuracy
+ return accuracy + (proximity / float(self.Match_Distance))
+
+ # Highest score beyond which we give up.
+ score_threshold = self.Match_Threshold
+ # Is there a nearby exact match? (speedup)
+ best_loc = text.find(pattern, loc)
+ if best_loc != -1:
+ score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
+ # What about in the other direction? (speedup)
+ best_loc = text.rfind(pattern, loc + len(pattern))
+ if best_loc != -1:
+ score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
+
+ # Initialise the bit arrays.
+ matchmask = 1 << (len(pattern) - 1)
+ best_loc = -1
+
+ bin_max = len(pattern) + len(text)
+ # Empty initialization added to appease pychecker.
+ last_rd = None
+ for d in xrange(len(pattern)):
+ # Scan for the best match each iteration allows for one more error.
+ # Run a binary search to determine how far from 'loc' we can stray at
+ # this error level.
+ bin_min = 0
+ bin_mid = bin_max
+ while bin_min < bin_mid:
+ if match_bitapScore(d, loc + bin_mid) <= score_threshold:
+ bin_min = bin_mid
+ else:
+ bin_max = bin_mid
+ bin_mid = (bin_max - bin_min) // 2 + bin_min
+
+ # Use the result from this iteration as the maximum for the next.
+ bin_max = bin_mid
+ start = max(1, loc - bin_mid + 1)
+ finish = min(loc + bin_mid, len(text)) + len(pattern)
+
+ rd = [0] * (finish + 2)
+ rd[finish + 1] = (1 << d) - 1
+ for j in xrange(finish, start - 1, -1):
+ if len(text) <= j - 1:
+ # Out of range.
+ charMatch = 0
+ else:
+ charMatch = s.get(text[j - 1], 0)
+ if d == 0: # First pass: exact match.
+ rd[j] = ((rd[j + 1] << 1) | 1) & charMatch
+ else: # Subsequent passes: fuzzy match.
+ rd[j] = (
+ (((rd[j + 1] << 1) | 1) & charMatch)
+ | (((last_rd[j + 1] | last_rd[j]) << 1) | 1)
+ | last_rd[j + 1]
+ )
+ if rd[j] & matchmask:
+ score = match_bitapScore(d, j - 1)
+ # This match will almost certainly be better than any existing match.
+ # But check anyway.
+ if score <= score_threshold:
+ # Told you so.
+ score_threshold = score
+ best_loc = j - 1
+ if best_loc > loc:
+ # When passing loc, don't exceed our current distance from loc.
+ start = max(1, 2 * loc - best_loc)
+ else:
+ # Already passed loc, downhill from here on in.
+ break
+ # No hope for a (better) match at greater error levels.
+ if match_bitapScore(d + 1, loc) > score_threshold:
+ break
+ last_rd = rd
+ return best_loc
- def match_alphabet(self, pattern):
- """Initialise the alphabet for the Bitap algorithm.
+ def match_alphabet(self, pattern):
+ """Initialise the alphabet for the Bitap algorithm.
Args:
pattern: The text to encode.
@@ -1360,58 +1437,61 @@ class diff_match_patch:
Returns:
Hash of character locations.
"""
- s = {}
- for char in pattern:
- s[char] = 0
- for i in xrange(len(pattern)):
- s[pattern[i]] |= 1 << (len(pattern) - i - 1)
- return s
+ s = {}
+ for char in pattern:
+ s[char] = 0
+ for i in xrange(len(pattern)):
+ s[pattern[i]] |= 1 << (len(pattern) - i - 1)
+ return s
- # PATCH FUNCTIONS
+ # PATCH FUNCTIONS
- def patch_addContext(self, patch, text):
- """Increase the context until it is unique,
+ def patch_addContext(self, patch, text):
+ """Increase the context until it is unique,
but don't let the pattern expand beyond Match_MaxBits.
Args:
patch: The patch to grow.
text: Source text.
"""
- if len(text) == 0:
- return
- pattern = text[patch.start2 : patch.start2 + patch.length1]
- padding = 0
-
- # Look for the first and last matches of pattern in text. If two different
- # matches are found, increase the pattern length.
- while (text.find(pattern) != text.rfind(pattern) and (self.Match_MaxBits ==
- 0 or len(pattern) < self.Match_MaxBits - self.Patch_Margin -
- self.Patch_Margin)):
- padding += self.Patch_Margin
- pattern = text[max(0, patch.start2 - padding) :
- patch.start2 + patch.length1 + padding]
- # Add one chunk for good luck.
- padding += self.Patch_Margin
-
- # Add the prefix.
- prefix = text[max(0, patch.start2 - padding) : patch.start2]
- if prefix:
- patch.diffs[:0] = [(self.DIFF_EQUAL, prefix)]
- # Add the suffix.
- suffix = text[patch.start2 + patch.length1 :
- patch.start2 + patch.length1 + padding]
- if suffix:
- patch.diffs.append((self.DIFF_EQUAL, suffix))
-
- # Roll back the start points.
- patch.start1 -= len(prefix)
- patch.start2 -= len(prefix)
- # Extend lengths.
- patch.length1 += len(prefix) + len(suffix)
- patch.length2 += len(prefix) + len(suffix)
+ if len(text) == 0:
+ return
+ pattern = text[patch.start2 : patch.start2 + patch.length1]
+ padding = 0
+
+ # Look for the first and last matches of pattern in text. If two different
+ # matches are found, increase the pattern length.
+ while text.find(pattern) != text.rfind(pattern) and (
+ self.Match_MaxBits == 0
+ or len(pattern) < self.Match_MaxBits - self.Patch_Margin - self.Patch_Margin
+ ):
+ padding += self.Patch_Margin
+ pattern = text[
+ max(0, patch.start2 - padding) : patch.start2 + patch.length1 + padding
+ ]
+ # Add one chunk for good luck.
+ padding += self.Patch_Margin
+
+ # Add the prefix.
+ prefix = text[max(0, patch.start2 - padding) : patch.start2]
+ if prefix:
+ patch.diffs[:0] = [(self.DIFF_EQUAL, prefix)]
+ # Add the suffix.
+ suffix = text[
+ patch.start2 + patch.length1 : patch.start2 + patch.length1 + padding
+ ]
+ if suffix:
+ patch.diffs.append((self.DIFF_EQUAL, suffix))
+
+ # Roll back the start points.
+ patch.start1 -= len(prefix)
+ patch.start2 -= len(prefix)
+ # Extend lengths.
+ patch.length1 += len(prefix) + len(suffix)
+ patch.length2 += len(prefix) + len(suffix)
- def patch_make(self, a, b=None, c=None):
- """Compute a list of patches to turn text1 into text2.
+ def patch_make(self, a, b=None, c=None):
+ """Compute a list of patches to turn text1 into text2.
Use diffs if provided, otherwise compute it ourselves.
There are four ways to call this function, depending on what data is
available to the caller:
@@ -1435,97 +1515,107 @@ class diff_match_patch:
Returns:
Array of Patch objects.
"""
- text1 = None
- diffs = None
- # Note that texts may arrive as 'str' or 'unicode'.
- if isinstance(a, basestring) and isinstance(b, basestring) and c is None:
- # Method 1: text1, text2
- # Compute diffs from text1 and text2.
- text1 = a
- diffs = self.diff_main(text1, b, True)
- if len(diffs) > 2:
- self.diff_cleanupSemantic(diffs)
- self.diff_cleanupEfficiency(diffs)
- elif isinstance(a, list) and b is None and c is None:
- # Method 2: diffs
- # Compute text1 from diffs.
- diffs = a
- text1 = self.diff_text1(diffs)
- elif isinstance(a, basestring) and isinstance(b, list) and c is None:
- # Method 3: text1, diffs
- text1 = a
- diffs = b
- elif (isinstance(a, basestring) and isinstance(b, basestring) and
- isinstance(c, list)):
- # Method 4: text1, text2, diffs
- # text2 is not used.
- text1 = a
- diffs = c
- else:
- raise ValueError("Unknown call format to patch_make.")
-
- if not diffs:
- return [] # Get rid of the None case.
- patches = []
- patch = patch_obj()
- char_count1 = 0 # Number of characters into the text1 string.
- char_count2 = 0 # Number of characters into the text2 string.
- prepatch_text = text1 # Recreate the patches to determine context info.
- postpatch_text = text1
- for x in xrange(len(diffs)):
- (diff_type, diff_text) = diffs[x]
- if len(patch.diffs) == 0 and diff_type != self.DIFF_EQUAL:
- # A new patch starts here.
- patch.start1 = char_count1
- patch.start2 = char_count2
- if diff_type == self.DIFF_INSERT:
- # Insertion
- patch.diffs.append(diffs[x])
- patch.length2 += len(diff_text)
- postpatch_text = (postpatch_text[:char_count2] + diff_text +
- postpatch_text[char_count2:])
- elif diff_type == self.DIFF_DELETE:
- # Deletion.
- patch.length1 += len(diff_text)
- patch.diffs.append(diffs[x])
- postpatch_text = (postpatch_text[:char_count2] +
- postpatch_text[char_count2 + len(diff_text):])
- elif (diff_type == self.DIFF_EQUAL and
- len(diff_text) <= 2 * self.Patch_Margin and
- len(patch.diffs) != 0 and len(diffs) != x + 1):
- # Small equality inside a patch.
- patch.diffs.append(diffs[x])
- patch.length1 += len(diff_text)
- patch.length2 += len(diff_text)
-
- if (diff_type == self.DIFF_EQUAL and
- len(diff_text) >= 2 * self.Patch_Margin):
- # Time for a new patch.
+ text1 = None
+ diffs = None
+ # Note that texts may arrive as 'str' or 'unicode'.
+ if isinstance(a, basestring) and isinstance(b, basestring) and c is None:
+ # Method 1: text1, text2
+ # Compute diffs from text1 and text2.
+ text1 = a
+ diffs = self.diff_main(text1, b, True)
+ if len(diffs) > 2:
+ self.diff_cleanupSemantic(diffs)
+ self.diff_cleanupEfficiency(diffs)
+ elif isinstance(a, list) and b is None and c is None:
+ # Method 2: diffs
+ # Compute text1 from diffs.
+ diffs = a
+ text1 = self.diff_text1(diffs)
+ elif isinstance(a, basestring) and isinstance(b, list) and c is None:
+ # Method 3: text1, diffs
+ text1 = a
+ diffs = b
+ elif (
+ isinstance(a, basestring)
+ and isinstance(b, basestring)
+ and isinstance(c, list)
+ ):
+ # Method 4: text1, text2, diffs
+ # text2 is not used.
+ text1 = a
+ diffs = c
+ else:
+ raise ValueError("Unknown call format to patch_make.")
+
+ if not diffs:
+ return [] # Get rid of the None case.
+ patches = []
+ patch = patch_obj()
+ char_count1 = 0 # Number of characters into the text1 string.
+ char_count2 = 0 # Number of characters into the text2 string.
+ prepatch_text = text1 # Recreate the patches to determine context info.
+ postpatch_text = text1
+ for x in xrange(len(diffs)):
+ (diff_type, diff_text) = diffs[x]
+ if len(patch.diffs) == 0 and diff_type != self.DIFF_EQUAL:
+ # A new patch starts here.
+ patch.start1 = char_count1
+ patch.start2 = char_count2
+ if diff_type == self.DIFF_INSERT:
+ # Insertion
+ patch.diffs.append(diffs[x])
+ patch.length2 += len(diff_text)
+ postpatch_text = (
+ postpatch_text[:char_count2]
+ + diff_text
+ + postpatch_text[char_count2:]
+ )
+ elif diff_type == self.DIFF_DELETE:
+ # Deletion.
+ patch.length1 += len(diff_text)
+ patch.diffs.append(diffs[x])
+ postpatch_text = (
+ postpatch_text[:char_count2]
+ + postpatch_text[char_count2 + len(diff_text) :]
+ )
+ elif (
+ diff_type == self.DIFF_EQUAL
+ and len(diff_text) <= 2 * self.Patch_Margin
+ and len(patch.diffs) != 0
+ and len(diffs) != x + 1
+ ):
+ # Small equality inside a patch.
+ patch.diffs.append(diffs[x])
+ patch.length1 += len(diff_text)
+ patch.length2 += len(diff_text)
+
+ if diff_type == self.DIFF_EQUAL and len(diff_text) >= 2 * self.Patch_Margin:
+ # Time for a new patch.
+ if len(patch.diffs) != 0:
+ self.patch_addContext(patch, prepatch_text)
+ patches.append(patch)
+ patch = patch_obj()
+ # Unlike Unidiff, our patch lists have a rolling context.
+ # https://github.com/google/diff-match-patch/wiki/Unidiff
+ # Update prepatch text & pos to reflect the application of the
+ # just completed patch.
+ prepatch_text = postpatch_text
+ char_count1 = char_count2
+
+ # Update the current character count.
+ if diff_type != self.DIFF_INSERT:
+ char_count1 += len(diff_text)
+ if diff_type != self.DIFF_DELETE:
+ char_count2 += len(diff_text)
+
+ # Pick up the leftover patch if not empty.
if len(patch.diffs) != 0:
- self.patch_addContext(patch, prepatch_text)
- patches.append(patch)
- patch = patch_obj()
- # Unlike Unidiff, our patch lists have a rolling context.
- # https://github.com/google/diff-match-patch/wiki/Unidiff
- # Update prepatch text & pos to reflect the application of the
- # just completed patch.
- prepatch_text = postpatch_text
- char_count1 = char_count2
-
- # Update the current character count.
- if diff_type != self.DIFF_INSERT:
- char_count1 += len(diff_text)
- if diff_type != self.DIFF_DELETE:
- char_count2 += len(diff_text)
-
- # Pick up the leftover patch if not empty.
- if len(patch.diffs) != 0:
- self.patch_addContext(patch, prepatch_text)
- patches.append(patch)
- return patches
+ self.patch_addContext(patch, prepatch_text)
+ patches.append(patch)
+ return patches
- def patch_deepCopy(self, patches):
- """Given an array of patches, return another array that is identical.
+ def patch_deepCopy(self, patches):
+ """Given an array of patches, return another array that is identical.
Args:
patches: Array of Patch objects.
@@ -1533,20 +1623,20 @@ class diff_match_patch:
Returns:
Array of Patch objects.
"""
- patchesCopy = []
- for patch in patches:
- patchCopy = patch_obj()
- # No need to deep copy the tuples since they are immutable.
- patchCopy.diffs = patch.diffs[:]
- patchCopy.start1 = patch.start1
- patchCopy.start2 = patch.start2
- patchCopy.length1 = patch.length1
- patchCopy.length2 = patch.length2
- patchesCopy.append(patchCopy)
- return patchesCopy
+ patchesCopy = []
+ for patch in patches:
+ patchCopy = patch_obj()
+ # No need to deep copy the tuples since they are immutable.
+ patchCopy.diffs = patch.diffs[:]
+ patchCopy.start1 = patch.start1
+ patchCopy.start2 = patch.start2
+ patchCopy.length1 = patch.length1
+ patchCopy.length2 = patch.length2
+ patchesCopy.append(patchCopy)
+ return patchesCopy
- def patch_apply(self, patches, text):
- """Merge a set of patches onto the text. Return a patched text, as well
+ def patch_apply(self, patches, text):
+ """Merge a set of patches onto the text. Return a patched text, as well
as a list of true/false values indicating which patches were applied.
Args:
@@ -1556,85 +1646,102 @@ class diff_match_patch:
Returns:
Two element Array, containing the new text and an array of boolean values.
"""
- if not patches:
- return (text, [])
+ if not patches:
+ return (text, [])
- # Deep copy the patches so that no changes are made to originals.
- patches = self.patch_deepCopy(patches)
+ # Deep copy the patches so that no changes are made to originals.
+ patches = self.patch_deepCopy(patches)
- nullPadding = self.patch_addPadding(patches)
- text = nullPadding + text + nullPadding
- self.patch_splitMax(patches)
-
- # delta keeps track of the offset between the expected and actual location
- # of the previous patch. If there are patches expected at positions 10 and
- # 20, but the first patch was found at 12, delta is 2 and the second patch
- # has an effective expected position of 22.
- delta = 0
- results = []
- for patch in patches:
- expected_loc = patch.start2 + delta
- text1 = self.diff_text1(patch.diffs)
- end_loc = -1
- if len(text1) > self.Match_MaxBits:
- # patch_splitMax will only provide an oversized pattern in the case of
- # a monster delete.
- start_loc = self.match_main(text, text1[:self.Match_MaxBits],
- expected_loc)
- if start_loc != -1:
- end_loc = self.match_main(text, text1[-self.Match_MaxBits:],
- expected_loc + len(text1) - self.Match_MaxBits)
- if end_loc == -1 or start_loc >= end_loc:
- # Can't find valid trailing context. Drop this patch.
- start_loc = -1
- else:
- start_loc = self.match_main(text, text1, expected_loc)
- if start_loc == -1:
- # No match found. :(
- results.append(False)
- # Subtract the delta for this failed patch from subsequent patches.
- delta -= patch.length2 - patch.length1
- else:
- # Found a match. :)
- results.append(True)
- delta = start_loc - expected_loc
- if end_loc == -1:
- text2 = text[start_loc : start_loc + len(text1)]
- else:
- text2 = text[start_loc : end_loc + self.Match_MaxBits]
- if text1 == text2:
- # Perfect match, just shove the replacement text in.
- text = (text[:start_loc] + self.diff_text2(patch.diffs) +
- text[start_loc + len(text1):])
- else:
- # Imperfect match.
- # Run a diff to get a framework of equivalent indices.
- diffs = self.diff_main(text1, text2, False)
- if (len(text1) > self.Match_MaxBits and
- self.diff_levenshtein(diffs) / float(len(text1)) >
- self.Patch_DeleteThreshold):
- # The end points match, but the content is unacceptably bad.
- results[-1] = False
- else:
- self.diff_cleanupSemanticLossless(diffs)
- index1 = 0
- for (op, data) in patch.diffs:
- if op != self.DIFF_EQUAL:
- index2 = self.diff_xIndex(diffs, index1)
- if op == self.DIFF_INSERT: # Insertion
- text = text[:start_loc + index2] + data + text[start_loc +
- index2:]
- elif op == self.DIFF_DELETE: # Deletion
- text = text[:start_loc + index2] + text[start_loc +
- self.diff_xIndex(diffs, index1 + len(data)):]
- if op != self.DIFF_DELETE:
- index1 += len(data)
- # Strip the padding off.
- text = text[len(nullPadding):-len(nullPadding)]
- return (text, results)
+ nullPadding = self.patch_addPadding(patches)
+ text = nullPadding + text + nullPadding
+ self.patch_splitMax(patches)
+
+ # delta keeps track of the offset between the expected and actual location
+ # of the previous patch. If there are patches expected at positions 10 and
+ # 20, but the first patch was found at 12, delta is 2 and the second patch
+ # has an effective expected position of 22.
+ delta = 0
+ results = []
+ for patch in patches:
+ expected_loc = patch.start2 + delta
+ text1 = self.diff_text1(patch.diffs)
+ end_loc = -1
+ if len(text1) > self.Match_MaxBits:
+ # patch_splitMax will only provide an oversized pattern in the case of
+ # a monster delete.
+ start_loc = self.match_main(
+ text, text1[: self.Match_MaxBits], expected_loc
+ )
+ if start_loc != -1:
+ end_loc = self.match_main(
+ text,
+ text1[-self.Match_MaxBits :],
+ expected_loc + len(text1) - self.Match_MaxBits,
+ )
+ if end_loc == -1 or start_loc >= end_loc:
+ # Can't find valid trailing context. Drop this patch.
+ start_loc = -1
+ else:
+ start_loc = self.match_main(text, text1, expected_loc)
+ if start_loc == -1:
+ # No match found. :(
+ results.append(False)
+ # Subtract the delta for this failed patch from subsequent patches.
+ delta -= patch.length2 - patch.length1
+ else:
+ # Found a match. :)
+ results.append(True)
+ delta = start_loc - expected_loc
+ if end_loc == -1:
+ text2 = text[start_loc : start_loc + len(text1)]
+ else:
+ text2 = text[start_loc : end_loc + self.Match_MaxBits]
+ if text1 == text2:
+ # Perfect match, just shove the replacement text in.
+ text = (
+ text[:start_loc]
+ + self.diff_text2(patch.diffs)
+ + text[start_loc + len(text1) :]
+ )
+ else:
+ # Imperfect match.
+ # Run a diff to get a framework of equivalent indices.
+ diffs = self.diff_main(text1, text2, False)
+ if (
+ len(text1) > self.Match_MaxBits
+ and self.diff_levenshtein(diffs) / float(len(text1))
+ > self.Patch_DeleteThreshold
+ ):
+ # The end points match, but the content is unacceptably bad.
+ results[-1] = False
+ else:
+ self.diff_cleanupSemanticLossless(diffs)
+ index1 = 0
+ for (op, data) in patch.diffs:
+ if op != self.DIFF_EQUAL:
+ index2 = self.diff_xIndex(diffs, index1)
+ if op == self.DIFF_INSERT: # Insertion
+ text = (
+ text[: start_loc + index2]
+ + data
+ + text[start_loc + index2 :]
+ )
+ elif op == self.DIFF_DELETE: # Deletion
+ text = (
+ text[: start_loc + index2]
+ + text[
+ start_loc
+ + self.diff_xIndex(diffs, index1 + len(data)) :
+ ]
+ )
+ if op != self.DIFF_DELETE:
+ index1 += len(data)
+ # Strip the padding off.
+ text = text[len(nullPadding) : -len(nullPadding)]
+ return (text, results)
- def patch_addPadding(self, patches):
- """Add some padding on text start and end so that edges can match
+ def patch_addPadding(self, patches):
+ """Add some padding on text start and end so that edges can match
something. Intended to be called only from within patch_apply.
Args:
@@ -1643,144 +1750,154 @@ class diff_match_patch:
Returns:
The padding string added to each side.
"""
- paddingLength = self.Patch_Margin
- nullPadding = ""
- for x in xrange(1, paddingLength + 1):
- nullPadding += chr(x)
-
- # Bump all the patches forward.
- for patch in patches:
- patch.start1 += paddingLength
- patch.start2 += paddingLength
-
- # Add some padding on start of first diff.
- patch = patches[0]
- diffs = patch.diffs
- if not diffs or diffs[0][0] != self.DIFF_EQUAL:
- # Add nullPadding equality.
- diffs.insert(0, (self.DIFF_EQUAL, nullPadding))
- patch.start1 -= paddingLength # Should be 0.
- patch.start2 -= paddingLength # Should be 0.
- patch.length1 += paddingLength
- patch.length2 += paddingLength
- elif paddingLength > len(diffs[0][1]):
- # Grow first equality.
- extraLength = paddingLength - len(diffs[0][1])
- newText = nullPadding[len(diffs[0][1]):] + diffs[0][1]
- diffs[0] = (diffs[0][0], newText)
- patch.start1 -= extraLength
- patch.start2 -= extraLength
- patch.length1 += extraLength
- patch.length2 += extraLength
-
- # Add some padding on end of last diff.
- patch = patches[-1]
- diffs = patch.diffs
- if not diffs or diffs[-1][0] != self.DIFF_EQUAL:
- # Add nullPadding equality.
- diffs.append((self.DIFF_EQUAL, nullPadding))
- patch.length1 += paddingLength
- patch.length2 += paddingLength
- elif paddingLength > len(diffs[-1][1]):
- # Grow last equality.
- extraLength = paddingLength - len(diffs[-1][1])
- newText = diffs[-1][1] + nullPadding[:extraLength]
- diffs[-1] = (diffs[-1][0], newText)
- patch.length1 += extraLength
- patch.length2 += extraLength
+ paddingLength = self.Patch_Margin
+ nullPadding = ""
+ for x in xrange(1, paddingLength + 1):
+ nullPadding += chr(x)
+
+ # Bump all the patches forward.
+ for patch in patches:
+ patch.start1 += paddingLength
+ patch.start2 += paddingLength
+
+ # Add some padding on start of first diff.
+ patch = patches[0]
+ diffs = patch.diffs
+ if not diffs or diffs[0][0] != self.DIFF_EQUAL:
+ # Add nullPadding equality.
+ diffs.insert(0, (self.DIFF_EQUAL, nullPadding))
+ patch.start1 -= paddingLength # Should be 0.
+ patch.start2 -= paddingLength # Should be 0.
+ patch.length1 += paddingLength
+ patch.length2 += paddingLength
+ elif paddingLength > len(diffs[0][1]):
+ # Grow first equality.
+ extraLength = paddingLength - len(diffs[0][1])
+ newText = nullPadding[len(diffs[0][1]) :] + diffs[0][1]
+ diffs[0] = (diffs[0][0], newText)
+ patch.start1 -= extraLength
+ patch.start2 -= extraLength
+ patch.length1 += extraLength
+ patch.length2 += extraLength
+
+ # Add some padding on end of last diff.
+ patch = patches[-1]
+ diffs = patch.diffs
+ if not diffs or diffs[-1][0] != self.DIFF_EQUAL:
+ # Add nullPadding equality.
+ diffs.append((self.DIFF_EQUAL, nullPadding))
+ patch.length1 += paddingLength
+ patch.length2 += paddingLength
+ elif paddingLength > len(diffs[-1][1]):
+ # Grow last equality.
+ extraLength = paddingLength - len(diffs[-1][1])
+ newText = diffs[-1][1] + nullPadding[:extraLength]
+ diffs[-1] = (diffs[-1][0], newText)
+ patch.length1 += extraLength
+ patch.length2 += extraLength
- return nullPadding
+ return nullPadding
- def patch_splitMax(self, patches):
- """Look through the patches and break up any which are longer than the
+ def patch_splitMax(self, patches):
+ """Look through the patches and break up any which are longer than the
maximum limit of the match algorithm.
Intended to be called only from within patch_apply.
Args:
patches: Array of Patch objects.
"""
- patch_size = self.Match_MaxBits
- if patch_size == 0:
- # Python has the option of not splitting strings due to its ability
- # to handle integers of arbitrary precision.
- return
- for x in xrange(len(patches)):
- if patches[x].length1 <= patch_size:
- continue
- bigpatch = patches[x]
- # Remove the big old patch.
- del patches[x]
- x -= 1
- start1 = bigpatch.start1
- start2 = bigpatch.start2
- precontext = ''
- while len(bigpatch.diffs) != 0:
- # Create one of several smaller patches.
- patch = patch_obj()
- empty = True
- patch.start1 = start1 - len(precontext)
- patch.start2 = start2 - len(precontext)
- if precontext:
- patch.length1 = patch.length2 = len(precontext)
- patch.diffs.append((self.DIFF_EQUAL, precontext))
-
- while (len(bigpatch.diffs) != 0 and
- patch.length1 < patch_size - self.Patch_Margin):
- (diff_type, diff_text) = bigpatch.diffs[0]
- if diff_type == self.DIFF_INSERT:
- # Insertions are harmless.
- patch.length2 += len(diff_text)
- start2 += len(diff_text)
- patch.diffs.append(bigpatch.diffs.pop(0))
- empty = False
- elif (diff_type == self.DIFF_DELETE and len(patch.diffs) == 1 and
- patch.diffs[0][0] == self.DIFF_EQUAL and
- len(diff_text) > 2 * patch_size):
- # This is a large deletion. Let it pass in one chunk.
- patch.length1 += len(diff_text)
- start1 += len(diff_text)
- empty = False
- patch.diffs.append((diff_type, diff_text))
- del bigpatch.diffs[0]
- else:
- # Deletion or equality. Only take as much as we can stomach.
- diff_text = diff_text[:patch_size - patch.length1 -
- self.Patch_Margin]
- patch.length1 += len(diff_text)
- start1 += len(diff_text)
- if diff_type == self.DIFF_EQUAL:
- patch.length2 += len(diff_text)
- start2 += len(diff_text)
- else:
- empty = False
-
- patch.diffs.append((diff_type, diff_text))
- if diff_text == bigpatch.diffs[0][1]:
- del bigpatch.diffs[0]
- else:
- bigpatch.diffs[0] = (bigpatch.diffs[0][0],
- bigpatch.diffs[0][1][len(diff_text):])
+ patch_size = self.Match_MaxBits
+ if patch_size == 0:
+ # Python has the option of not splitting strings due to its ability
+ # to handle integers of arbitrary precision.
+ return
+ for x in xrange(len(patches)):
+ if patches[x].length1 <= patch_size:
+ continue
+ bigpatch = patches[x]
+ # Remove the big old patch.
+ del patches[x]
+ x -= 1
+ start1 = bigpatch.start1
+ start2 = bigpatch.start2
+ precontext = ""
+ while len(bigpatch.diffs) != 0:
+ # Create one of several smaller patches.
+ patch = patch_obj()
+ empty = True
+ patch.start1 = start1 - len(precontext)
+ patch.start2 = start2 - len(precontext)
+ if precontext:
+ patch.length1 = patch.length2 = len(precontext)
+ patch.diffs.append((self.DIFF_EQUAL, precontext))
+
+ while (
+ len(bigpatch.diffs) != 0
+ and patch.length1 < patch_size - self.Patch_Margin
+ ):
+ (diff_type, diff_text) = bigpatch.diffs[0]
+ if diff_type == self.DIFF_INSERT:
+ # Insertions are harmless.
+ patch.length2 += len(diff_text)
+ start2 += len(diff_text)
+ patch.diffs.append(bigpatch.diffs.pop(0))
+ empty = False
+ elif (
+ diff_type == self.DIFF_DELETE
+ and len(patch.diffs) == 1
+ and patch.diffs[0][0] == self.DIFF_EQUAL
+ and len(diff_text) > 2 * patch_size
+ ):
+ # This is a large deletion. Let it pass in one chunk.
+ patch.length1 += len(diff_text)
+ start1 += len(diff_text)
+ empty = False
+ patch.diffs.append((diff_type, diff_text))
+ del bigpatch.diffs[0]
+ else:
+ # Deletion or equality. Only take as much as we can stomach.
+ diff_text = diff_text[
+ : patch_size - patch.length1 - self.Patch_Margin
+ ]
+ patch.length1 += len(diff_text)
+ start1 += len(diff_text)
+ if diff_type == self.DIFF_EQUAL:
+ patch.length2 += len(diff_text)
+ start2 += len(diff_text)
+ else:
+ empty = False
+
+ patch.diffs.append((diff_type, diff_text))
+ if diff_text == bigpatch.diffs[0][1]:
+ del bigpatch.diffs[0]
+ else:
+ bigpatch.diffs[0] = (
+ bigpatch.diffs[0][0],
+ bigpatch.diffs[0][1][len(diff_text) :],
+ )
+
+ # Compute the head context for the next patch.
+ precontext = self.diff_text2(patch.diffs)
+ precontext = precontext[-self.Patch_Margin :]
+ # Append the end context for this patch.
+ postcontext = self.diff_text1(bigpatch.diffs)[: self.Patch_Margin]
+ if postcontext:
+ patch.length1 += len(postcontext)
+ patch.length2 += len(postcontext)
+ if len(patch.diffs) != 0 and patch.diffs[-1][0] == self.DIFF_EQUAL:
+ patch.diffs[-1] = (
+ self.DIFF_EQUAL,
+ patch.diffs[-1][1] + postcontext,
+ )
+ else:
+ patch.diffs.append((self.DIFF_EQUAL, postcontext))
+
+ if not empty:
+ x += 1
+ patches.insert(x, patch)
- # Compute the head context for the next patch.
- precontext = self.diff_text2(patch.diffs)
- precontext = precontext[-self.Patch_Margin:]
- # Append the end context for this patch.
- postcontext = self.diff_text1(bigpatch.diffs)[:self.Patch_Margin]
- if postcontext:
- patch.length1 += len(postcontext)
- patch.length2 += len(postcontext)
- if len(patch.diffs) != 0 and patch.diffs[-1][0] == self.DIFF_EQUAL:
- patch.diffs[-1] = (self.DIFF_EQUAL, patch.diffs[-1][1] +
- postcontext)
- else:
- patch.diffs.append((self.DIFF_EQUAL, postcontext))
-
- if not empty:
- x += 1
- patches.insert(x, patch)
-
- def patch_toText(self, patches):
- """Take a list of patches and return a textual representation.
+ def patch_toText(self, patches):
+ """Take a list of patches and return a textual representation.
Args:
patches: Array of Patch objects.
@@ -1788,13 +1905,13 @@ class diff_match_patch:
Returns:
Text representation of patches.
"""
- text = []
- for patch in patches:
- text.append(str(patch))
- return "".join(text)
+ text = []
+ for patch in patches:
+ text.append(str(patch))
+ return "".join(text)
- def patch_fromText(self, textline):
- """Parse a textual representation of patches and return a list of patch
+ def patch_fromText(self, textline):
+ """Parse a textual representation of patches and return a list of patch
objects.
Args:
@@ -1806,114 +1923,114 @@ class diff_match_patch:
Raises:
ValueError: If invalid input.
"""
- if type(textline) == unicode:
- # Patches should be composed of a subset of ascii chars, Unicode not
- # required. If this encode raises UnicodeEncodeError, patch is invalid.
- textline = textline.encode("ascii")
- patches = []
- if not textline:
- return patches
- text = textline.split('\n')
- while len(text) != 0:
- m = re.match("^@@ -(\d+),?(\d*) \+(\d+),?(\d*) @@$", text[0])
- if not m:
- raise ValueError("Invalid patch string: " + text[0])
- patch = patch_obj()
- patches.append(patch)
- patch.start1 = int(m.group(1))
- if m.group(2) == '':
- patch.start1 -= 1
- patch.length1 = 1
- elif m.group(2) == '0':
- patch.length1 = 0
- else:
- patch.start1 -= 1
- patch.length1 = int(m.group(2))
-
- patch.start2 = int(m.group(3))
- if m.group(4) == '':
- patch.start2 -= 1
- patch.length2 = 1
- elif m.group(4) == '0':
- patch.length2 = 0
- else:
- patch.start2 -= 1
- patch.length2 = int(m.group(4))
-
- del text[0]
-
- while len(text) != 0:
- if text[0]:
- sign = text[0][0]
- else:
- sign = ''
- line = urllib.unquote(text[0][1:])
- line = line.decode("utf-8")
- if sign == '+':
- # Insertion.
- patch.diffs.append((self.DIFF_INSERT, line))
- elif sign == '-':
- # Deletion.
- patch.diffs.append((self.DIFF_DELETE, line))
- elif sign == ' ':
- # Minor equality.
- patch.diffs.append((self.DIFF_EQUAL, line))
- elif sign == '@':
- # Start of next patch.
- break
- elif sign == '':
- # Blank line? Whatever.
- pass
- else:
- # WTF?
- raise ValueError("Invalid patch mode: '%s'\n%s" % (sign, line))
- del text[0]
- return patches
+ if type(textline) == unicode:
+ # Patches should be composed of a subset of ascii chars, Unicode not
+ # required. If this encode raises UnicodeEncodeError, patch is invalid.
+ textline = textline.encode("ascii")
+ patches = []
+ if not textline:
+ return patches
+ text = textline.split("\n")
+ while len(text) != 0:
+ m = re.match(r"^@@ -(\d+),?(\d*) \+(\d+),?(\d*) @@$", text[0])
+ if not m:
+ raise ValueError("Invalid patch string: " + text[0])
+ patch = patch_obj()
+ patches.append(patch)
+ patch.start1 = int(m.group(1))
+ if m.group(2) == "":
+ patch.start1 -= 1
+ patch.length1 = 1
+ elif m.group(2) == "0":
+ patch.length1 = 0
+ else:
+ patch.start1 -= 1
+ patch.length1 = int(m.group(2))
+
+ patch.start2 = int(m.group(3))
+ if m.group(4) == "":
+ patch.start2 -= 1
+ patch.length2 = 1
+ elif m.group(4) == "0":
+ patch.length2 = 0
+ else:
+ patch.start2 -= 1
+ patch.length2 = int(m.group(4))
+
+ del text[0]
+
+ while len(text) != 0:
+ if text[0]:
+ sign = text[0][0]
+ else:
+ sign = ""
+ line = urllib.unquote(text[0][1:])
+ line = line.decode("utf-8")
+ if sign == "+":
+ # Insertion.
+ patch.diffs.append((self.DIFF_INSERT, line))
+ elif sign == "-":
+ # Deletion.
+ patch.diffs.append((self.DIFF_DELETE, line))
+ elif sign == " ":
+ # Minor equality.
+ patch.diffs.append((self.DIFF_EQUAL, line))
+ elif sign == "@":
+ # Start of next patch.
+ break
+ elif sign == "":
+ # Blank line? Whatever.
+ pass
+ else:
+ # WTF?
+ raise ValueError(f"Invalid patch mode: '{sign}'\n{line}")
+ del text[0]
+ return patches
class patch_obj:
- """Class representing one patch operation.
+ """Class representing one patch operation.
"""
- def __init__(self):
- """Initializes with an empty list of diffs.
+ def __init__(self):
+ """Initializes with an empty list of diffs.
"""
- self.diffs = []
- self.start1 = None
- self.start2 = None
- self.length1 = 0
- self.length2 = 0
+ self.diffs = []
+ self.start1 = None
+ self.start2 = None
+ self.length1 = 0
+ self.length2 = 0
- def __str__(self):
- """Emulate GNU diff's format.
+ def __str__(self):
+ """Emulate GNU diff's format.
Header: @@ -382,8 +481,9 @@
Indices are printed as 1-based, not 0-based.
Returns:
The GNU diff string.
"""
- if self.length1 == 0:
- coords1 = str(self.start1) + ",0"
- elif self.length1 == 1:
- coords1 = str(self.start1 + 1)
- else:
- coords1 = str(self.start1 + 1) + "," + str(self.length1)
- if self.length2 == 0:
- coords2 = str(self.start2) + ",0"
- elif self.length2 == 1:
- coords2 = str(self.start2 + 1)
- else:
- coords2 = str(self.start2 + 1) + "," + str(self.length2)
- text = ["@@ -", coords1, " +", coords2, " @@\n"]
- # Escape the body of the patch with %xx notation.
- for (op, data) in self.diffs:
- if op == diff_match_patch.DIFF_INSERT:
- text.append("+")
- elif op == diff_match_patch.DIFF_DELETE:
- text.append("-")
- elif op == diff_match_patch.DIFF_EQUAL:
- text.append(" ")
- # High ascii will raise UnicodeDecodeError. Use Unicode instead.
- data = data.encode("utf-8")
- text.append(urllib.quote(data, "!~*'();/?:@&=+$,# ") + "\n")
- return "".join(text)
+ if self.length1 == 0:
+ coords1 = str(self.start1) + ",0"
+ elif self.length1 == 1:
+ coords1 = str(self.start1 + 1)
+ else:
+ coords1 = str(self.start1 + 1) + "," + str(self.length1)
+ if self.length2 == 0:
+ coords2 = str(self.start2) + ",0"
+ elif self.length2 == 1:
+ coords2 = str(self.start2 + 1)
+ else:
+ coords2 = str(self.start2 + 1) + "," + str(self.length2)
+ text = ["@@ -", coords1, " +", coords2, " @@\n"]
+ # Escape the body of the patch with %xx notation.
+ for (op, data) in self.diffs:
+ if op == diff_match_patch.DIFF_INSERT:
+ text.append("+")
+ elif op == diff_match_patch.DIFF_DELETE:
+ text.append("-")
+ elif op == diff_match_patch.DIFF_EQUAL:
+ text.append(" ")
+ # High ascii will raise UnicodeDecodeError. Use Unicode instead.
+ data = data.encode("utf-8")
+ text.append(urllib.quote(data, "!~*'();/?:@&=+$,# ") + "\n")
+ return "".join(text)
Index: xmldiff-2.4/xmldiff/_diff_match_patch_py3.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/_diff_match_patch_py3.py
+++ xmldiff-2.4/xmldiff/_diff_match_patch_py3.py
@@ -23,7 +23,7 @@ Computes the difference between two text
Applies the patch onto another text, allowing for errors.
"""
-__author__ = 'fraser@google.com (Neil Fraser)'
+__author__ = "fraser@google.com (Neil Fraser)"
import re
import sys
@@ -32,51 +32,51 @@ import urllib.parse
class diff_match_patch:
- """Class containing the diff, match and patch methods.
+ """Class containing the diff, match and patch methods.
Also contains the behaviour settings.
"""
- def __init__(self):
- """Inits a diff_match_patch object with default settings.
+ def __init__(self):
+ """Inits a diff_match_patch object with default settings.
Redefine these in your program to override the defaults.
"""
- # Number of seconds to map a diff before giving up (0 for infinity).
- self.Diff_Timeout = 1.0
- # Cost of an empty edit operation in terms of edit characters.
- self.Diff_EditCost = 4
- # At what point is no match declared (0.0 = perfection, 1.0 = very loose).
- self.Match_Threshold = 0.5
- # How far to search for a match (0 = exact location, 1000+ = broad match).
- # A match this many characters away from the expected location will add
- # 1.0 to the score (0.0 is a perfect match).
- self.Match_Distance = 1000
- # When deleting a large block of text (over ~64 characters), how close do
- # the contents have to be to match the expected contents. (0.0 = perfection,
- # 1.0 = very loose). Note that Match_Threshold controls how closely the
- # end points of a delete need to match.
- self.Patch_DeleteThreshold = 0.5
- # Chunk size for context length.
- self.Patch_Margin = 4
-
- # The number of bits in an int.
- # Python has no maximum, thus to disable patch splitting set to 0.
- # However to avoid long patches in certain pathological cases, use 32.
- # Multiple short patches (using native ints) are much faster than long ones.
- self.Match_MaxBits = 32
-
- # DIFF FUNCTIONS
-
- # The data structure representing a diff is an array of tuples:
- # [(DIFF_DELETE, "Hello"), (DIFF_INSERT, "Goodbye"), (DIFF_EQUAL, " world.")]
- # which means: delete "Hello", add "Goodbye" and keep " world."
- DIFF_DELETE = -1
- DIFF_INSERT = 1
- DIFF_EQUAL = 0
+ # Number of seconds to map a diff before giving up (0 for infinity).
+ self.Diff_Timeout = 1.0
+ # Cost of an empty edit operation in terms of edit characters.
+ self.Diff_EditCost = 4
+ # At what point is no match declared (0.0 = perfection, 1.0 = very loose).
+ self.Match_Threshold = 0.5
+ # How far to search for a match (0 = exact location, 1000+ = broad match).
+ # A match this many characters away from the expected location will add
+ # 1.0 to the score (0.0 is a perfect match).
+ self.Match_Distance = 1000
+ # When deleting a large block of text (over ~64 characters), how close do
+ # the contents have to be to match the expected contents. (0.0 = perfection,
+ # 1.0 = very loose). Note that Match_Threshold controls how closely the
+ # end points of a delete need to match.
+ self.Patch_DeleteThreshold = 0.5
+ # Chunk size for context length.
+ self.Patch_Margin = 4
+
+ # The number of bits in an int.
+ # Python has no maximum, thus to disable patch splitting set to 0.
+ # However to avoid long patches in certain pathological cases, use 32.
+ # Multiple short patches (using native ints) are much faster than long ones.
+ self.Match_MaxBits = 32
+
+ # DIFF FUNCTIONS
+
+ # The data structure representing a diff is an array of tuples:
+ # [(DIFF_DELETE, "Hello"), (DIFF_INSERT, "Goodbye"), (DIFF_EQUAL, " world.")]
+ # which means: delete "Hello", add "Goodbye" and keep " world."
+ DIFF_DELETE = -1
+ DIFF_INSERT = 1
+ DIFF_EQUAL = 0
- def diff_main(self, text1, text2, checklines=True, deadline=None):
- """Find the differences between two texts. Simplifies the problem by
+ def diff_main(self, text1, text2, checklines=True, deadline=None):
+ """Find the differences between two texts. Simplifies the problem by
stripping any common prefix or suffix off the texts before diffing.
Args:
@@ -91,52 +91,52 @@ class diff_match_patch:
Returns:
Array of changes.
"""
- # Set a deadline by which time the diff must be complete.
- if deadline == None:
- # Unlike in most languages, Python counts time in seconds.
- if self.Diff_Timeout <= 0:
- deadline = sys.maxsize
- else:
- deadline = time.time() + self.Diff_Timeout
-
- # Check for null inputs.
- if text1 == None or text2 == None:
- raise ValueError("Null inputs. (diff_main)")
-
- # Check for equality (speedup).
- if text1 == text2:
- if text1:
- return [(self.DIFF_EQUAL, text1)]
- return []
-
- # Trim off common prefix (speedup).
- commonlength = self.diff_commonPrefix(text1, text2)
- commonprefix = text1[:commonlength]
- text1 = text1[commonlength:]
- text2 = text2[commonlength:]
-
- # Trim off common suffix (speedup).
- commonlength = self.diff_commonSuffix(text1, text2)
- if commonlength == 0:
- commonsuffix = ''
- else:
- commonsuffix = text1[-commonlength:]
- text1 = text1[:-commonlength]
- text2 = text2[:-commonlength]
-
- # Compute the diff on the middle block.
- diffs = self.diff_compute(text1, text2, checklines, deadline)
-
- # Restore the prefix and suffix.
- if commonprefix:
- diffs[:0] = [(self.DIFF_EQUAL, commonprefix)]
- if commonsuffix:
- diffs.append((self.DIFF_EQUAL, commonsuffix))
- self.diff_cleanupMerge(diffs)
- return diffs
+ # Set a deadline by which time the diff must be complete.
+ if deadline == None:
+ # Unlike in most languages, Python counts time in seconds.
+ if self.Diff_Timeout <= 0:
+ deadline = sys.maxsize
+ else:
+ deadline = time.time() + self.Diff_Timeout
+
+ # Check for null inputs.
+ if text1 == None or text2 == None:
+ raise ValueError("Null inputs. (diff_main)")
+
+ # Check for equality (speedup).
+ if text1 == text2:
+ if text1:
+ return [(self.DIFF_EQUAL, text1)]
+ return []
+
+ # Trim off common prefix (speedup).
+ commonlength = self.diff_commonPrefix(text1, text2)
+ commonprefix = text1[:commonlength]
+ text1 = text1[commonlength:]
+ text2 = text2[commonlength:]
+
+ # Trim off common suffix (speedup).
+ commonlength = self.diff_commonSuffix(text1, text2)
+ if commonlength == 0:
+ commonsuffix = ""
+ else:
+ commonsuffix = text1[-commonlength:]
+ text1 = text1[:-commonlength]
+ text2 = text2[:-commonlength]
+
+ # Compute the diff on the middle block.
+ diffs = self.diff_compute(text1, text2, checklines, deadline)
+
+ # Restore the prefix and suffix.
+ if commonprefix:
+ diffs[:0] = [(self.DIFF_EQUAL, commonprefix)]
+ if commonsuffix:
+ diffs.append((self.DIFF_EQUAL, commonsuffix))
+ self.diff_cleanupMerge(diffs)
+ return diffs
- def diff_compute(self, text1, text2, checklines, deadline):
- """Find the differences between two texts. Assumes that the texts do not
+ def diff_compute(self, text1, text2, checklines, deadline):
+ """Find the differences between two texts. Assumes that the texts do not
have any common prefix or suffix.
Args:
@@ -150,52 +150,55 @@ class diff_match_patch:
Returns:
Array of changes.
"""
- if not text1:
- # Just add some text (speedup).
- return [(self.DIFF_INSERT, text2)]
-
- if not text2:
- # Just delete some text (speedup).
- return [(self.DIFF_DELETE, text1)]
-
- if len(text1) > len(text2):
- (longtext, shorttext) = (text1, text2)
- else:
- (shorttext, longtext) = (text1, text2)
- i = longtext.find(shorttext)
- if i != -1:
- # Shorter text is inside the longer text (speedup).
- diffs = [(self.DIFF_INSERT, longtext[:i]), (self.DIFF_EQUAL, shorttext),
- (self.DIFF_INSERT, longtext[i + len(shorttext):])]
- # Swap insertions for deletions if diff is reversed.
- if len(text1) > len(text2):
- diffs[0] = (self.DIFF_DELETE, diffs[0][1])
- diffs[2] = (self.DIFF_DELETE, diffs[2][1])
- return diffs
-
- if len(shorttext) == 1:
- # Single character string.
- # After the previous speedup, the character can't be an equality.
- return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
-
- # Check to see if the problem can be split in two.
- hm = self.diff_halfMatch(text1, text2)
- if hm:
- # A half-match was found, sort out the return data.
- (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
- # Send both pairs off for separate processing.
- diffs_a = self.diff_main(text1_a, text2_a, checklines, deadline)
- diffs_b = self.diff_main(text1_b, text2_b, checklines, deadline)
- # Merge the results.
- return diffs_a + [(self.DIFF_EQUAL, mid_common)] + diffs_b
+ if not text1:
+ # Just add some text (speedup).
+ return [(self.DIFF_INSERT, text2)]
+
+ if not text2:
+ # Just delete some text (speedup).
+ return [(self.DIFF_DELETE, text1)]
- if checklines and len(text1) > 100 and len(text2) > 100:
- return self.diff_lineMode(text1, text2, deadline)
+ if len(text1) > len(text2):
+ (longtext, shorttext) = (text1, text2)
+ else:
+ (shorttext, longtext) = (text1, text2)
+ i = longtext.find(shorttext)
+ if i != -1:
+ # Shorter text is inside the longer text (speedup).
+ diffs = [
+ (self.DIFF_INSERT, longtext[:i]),
+ (self.DIFF_EQUAL, shorttext),
+ (self.DIFF_INSERT, longtext[i + len(shorttext) :]),
+ ]
+ # Swap insertions for deletions if diff is reversed.
+ if len(text1) > len(text2):
+ diffs[0] = (self.DIFF_DELETE, diffs[0][1])
+ diffs[2] = (self.DIFF_DELETE, diffs[2][1])
+ return diffs
+
+ if len(shorttext) == 1:
+ # Single character string.
+ # After the previous speedup, the character can't be an equality.
+ return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
+
+ # Check to see if the problem can be split in two.
+ hm = self.diff_halfMatch(text1, text2)
+ if hm:
+ # A half-match was found, sort out the return data.
+ (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
+ # Send both pairs off for separate processing.
+ diffs_a = self.diff_main(text1_a, text2_a, checklines, deadline)
+ diffs_b = self.diff_main(text1_b, text2_b, checklines, deadline)
+ # Merge the results.
+ return diffs_a + [(self.DIFF_EQUAL, mid_common)] + diffs_b
+
+ if checklines and len(text1) > 100 and len(text2) > 100:
+ return self.diff_lineMode(text1, text2, deadline)
- return self.diff_bisect(text1, text2, deadline)
+ return self.diff_bisect(text1, text2, deadline)
- def diff_lineMode(self, text1, text2, deadline):
- """Do a quick line-level diff on both strings, then rediff the parts for
+ def diff_lineMode(self, text1, text2, deadline):
+ """Do a quick line-level diff on both strings, then rediff the parts for
greater accuracy.
This speedup can produce non-minimal diffs.
@@ -208,51 +211,51 @@ class diff_match_patch:
Array of changes.
"""
- # Scan the text on a line-by-line basis first.
- (text1, text2, linearray) = self.diff_linesToChars(text1, text2)
+ # Scan the text on a line-by-line basis first.
+ (text1, text2, linearray) = self.diff_linesToChars(text1, text2)
- diffs = self.diff_main(text1, text2, False, deadline)
+ diffs = self.diff_main(text1, text2, False, deadline)
- # Convert the diff back to original text.
- self.diff_charsToLines(diffs, linearray)
- # Eliminate freak matches (e.g. blank lines)
- self.diff_cleanupSemantic(diffs)
-
- # Rediff any replacement blocks, this time character-by-character.
- # Add a dummy entry at the end.
- diffs.append((self.DIFF_EQUAL, ''))
- pointer = 0
- count_delete = 0
- count_insert = 0
- text_delete = ''
- text_insert = ''
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_INSERT:
- count_insert += 1
- text_insert += diffs[pointer][1]
- elif diffs[pointer][0] == self.DIFF_DELETE:
- count_delete += 1
- text_delete += diffs[pointer][1]
- elif diffs[pointer][0] == self.DIFF_EQUAL:
- # Upon reaching an equality, check for prior redundancies.
- if count_delete >= 1 and count_insert >= 1:
- # Delete the offending records and add the merged ones.
- subDiff = self.diff_main(text_delete, text_insert, False, deadline)
- diffs[pointer - count_delete - count_insert : pointer] = subDiff
- pointer = pointer - count_delete - count_insert + len(subDiff)
- count_insert = 0
+ # Convert the diff back to original text.
+ self.diff_charsToLines(diffs, linearray)
+ # Eliminate freak matches (e.g. blank lines)
+ self.diff_cleanupSemantic(diffs)
+
+ # Rediff any replacement blocks, this time character-by-character.
+ # Add a dummy entry at the end.
+ diffs.append((self.DIFF_EQUAL, ""))
+ pointer = 0
count_delete = 0
- text_delete = ''
- text_insert = ''
+ count_insert = 0
+ text_delete = ""
+ text_insert = ""
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_INSERT:
+ count_insert += 1
+ text_insert += diffs[pointer][1]
+ elif diffs[pointer][0] == self.DIFF_DELETE:
+ count_delete += 1
+ text_delete += diffs[pointer][1]
+ elif diffs[pointer][0] == self.DIFF_EQUAL:
+ # Upon reaching an equality, check for prior redundancies.
+ if count_delete >= 1 and count_insert >= 1:
+ # Delete the offending records and add the merged ones.
+ subDiff = self.diff_main(text_delete, text_insert, False, deadline)
+ diffs[pointer - count_delete - count_insert : pointer] = subDiff
+ pointer = pointer - count_delete - count_insert + len(subDiff)
+ count_insert = 0
+ count_delete = 0
+ text_delete = ""
+ text_insert = ""
- pointer += 1
+ pointer += 1
- diffs.pop() # Remove the dummy entry at the end.
+ diffs.pop() # Remove the dummy entry at the end.
- return diffs
+ return diffs
- def diff_bisect(self, text1, text2, deadline):
- """Find the 'middle snake' of a diff, split the problem in two
+ def diff_bisect(self, text1, text2, deadline):
+ """Find the 'middle snake' of a diff, split the problem in two
and return the recursively constructed diff.
See Myers 1986 paper: An O(ND) Difference Algorithm and Its Variations.
@@ -265,96 +268,98 @@ class diff_match_patch:
Array of diff tuples.
"""
- # Cache the text lengths to prevent multiple calls.
- text1_length = len(text1)
- text2_length = len(text2)
- max_d = (text1_length + text2_length + 1) // 2
- v_offset = max_d
- v_length = 2 * max_d
- v1 = [-1] * v_length
- v1[v_offset + 1] = 0
- v2 = v1[:]
- delta = text1_length - text2_length
- # If the total number of characters is odd, then the front path will
- # collide with the reverse path.
- front = (delta % 2 != 0)
- # Offsets for start and end of k loop.
- # Prevents mapping of space beyond the grid.
- k1start = 0
- k1end = 0
- k2start = 0
- k2end = 0
- for d in range(max_d):
- # Bail out if deadline is reached.
- if time.time() > deadline:
- break
-
- # Walk the front path one step.
- for k1 in range(-d + k1start, d + 1 - k1end, 2):
- k1_offset = v_offset + k1
- if k1 == -d or (k1 != d and
- v1[k1_offset - 1] < v1[k1_offset + 1]):
- x1 = v1[k1_offset + 1]
- else:
- x1 = v1[k1_offset - 1] + 1
- y1 = x1 - k1
- while (x1 < text1_length and y1 < text2_length and
- text1[x1] == text2[y1]):
- x1 += 1
- y1 += 1
- v1[k1_offset] = x1
- if x1 > text1_length:
- # Ran off the right of the graph.
- k1end += 2
- elif y1 > text2_length:
- # Ran off the bottom of the graph.
- k1start += 2
- elif front:
- k2_offset = v_offset + delta - k1
- if k2_offset >= 0 and k2_offset < v_length and v2[k2_offset] != -1:
- # Mirror x2 onto top-left coordinate system.
- x2 = text1_length - v2[k2_offset]
- if x1 >= x2:
- # Overlap detected.
- return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
-
- # Walk the reverse path one step.
- for k2 in range(-d + k2start, d + 1 - k2end, 2):
- k2_offset = v_offset + k2
- if k2 == -d or (k2 != d and
- v2[k2_offset - 1] < v2[k2_offset + 1]):
- x2 = v2[k2_offset + 1]
- else:
- x2 = v2[k2_offset - 1] + 1
- y2 = x2 - k2
- while (x2 < text1_length and y2 < text2_length and
- text1[-x2 - 1] == text2[-y2 - 1]):
- x2 += 1
- y2 += 1
- v2[k2_offset] = x2
- if x2 > text1_length:
- # Ran off the left of the graph.
- k2end += 2
- elif y2 > text2_length:
- # Ran off the top of the graph.
- k2start += 2
- elif not front:
- k1_offset = v_offset + delta - k2
- if k1_offset >= 0 and k1_offset < v_length and v1[k1_offset] != -1:
- x1 = v1[k1_offset]
- y1 = v_offset + x1 - k1_offset
- # Mirror x2 onto top-left coordinate system.
- x2 = text1_length - x2
- if x1 >= x2:
- # Overlap detected.
- return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
-
- # Diff took too long and hit the deadline or
- # number of diffs equals number of characters, no commonality at all.
- return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
+ # Cache the text lengths to prevent multiple calls.
+ text1_length = len(text1)
+ text2_length = len(text2)
+ max_d = (text1_length + text2_length + 1) // 2
+ v_offset = max_d
+ v_length = 2 * max_d
+ v1 = [-1] * v_length
+ v1[v_offset + 1] = 0
+ v2 = v1[:]
+ delta = text1_length - text2_length
+ # If the total number of characters is odd, then the front path will
+ # collide with the reverse path.
+ front = delta % 2 != 0
+ # Offsets for start and end of k loop.
+ # Prevents mapping of space beyond the grid.
+ k1start = 0
+ k1end = 0
+ k2start = 0
+ k2end = 0
+ for d in range(max_d):
+ # Bail out if deadline is reached.
+ if time.time() > deadline:
+ break
+
+ # Walk the front path one step.
+ for k1 in range(-d + k1start, d + 1 - k1end, 2):
+ k1_offset = v_offset + k1
+ if k1 == -d or (k1 != d and v1[k1_offset - 1] < v1[k1_offset + 1]):
+ x1 = v1[k1_offset + 1]
+ else:
+ x1 = v1[k1_offset - 1] + 1
+ y1 = x1 - k1
+ while (
+ x1 < text1_length and y1 < text2_length and text1[x1] == text2[y1]
+ ):
+ x1 += 1
+ y1 += 1
+ v1[k1_offset] = x1
+ if x1 > text1_length:
+ # Ran off the right of the graph.
+ k1end += 2
+ elif y1 > text2_length:
+ # Ran off the bottom of the graph.
+ k1start += 2
+ elif front:
+ k2_offset = v_offset + delta - k1
+ if k2_offset >= 0 and k2_offset < v_length and v2[k2_offset] != -1:
+ # Mirror x2 onto top-left coordinate system.
+ x2 = text1_length - v2[k2_offset]
+ if x1 >= x2:
+ # Overlap detected.
+ return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
+
+ # Walk the reverse path one step.
+ for k2 in range(-d + k2start, d + 1 - k2end, 2):
+ k2_offset = v_offset + k2
+ if k2 == -d or (k2 != d and v2[k2_offset - 1] < v2[k2_offset + 1]):
+ x2 = v2[k2_offset + 1]
+ else:
+ x2 = v2[k2_offset - 1] + 1
+ y2 = x2 - k2
+ while (
+ x2 < text1_length
+ and y2 < text2_length
+ and text1[-x2 - 1] == text2[-y2 - 1]
+ ):
+ x2 += 1
+ y2 += 1
+ v2[k2_offset] = x2
+ if x2 > text1_length:
+ # Ran off the left of the graph.
+ k2end += 2
+ elif y2 > text2_length:
+ # Ran off the top of the graph.
+ k2start += 2
+ elif not front:
+ k1_offset = v_offset + delta - k2
+ if k1_offset >= 0 and k1_offset < v_length and v1[k1_offset] != -1:
+ x1 = v1[k1_offset]
+ y1 = v_offset + x1 - k1_offset
+ # Mirror x2 onto top-left coordinate system.
+ x2 = text1_length - x2
+ if x1 >= x2:
+ # Overlap detected.
+ return self.diff_bisectSplit(text1, text2, x1, y1, deadline)
+
+ # Diff took too long and hit the deadline or
+ # number of diffs equals number of characters, no commonality at all.
+ return [(self.DIFF_DELETE, text1), (self.DIFF_INSERT, text2)]
- def diff_bisectSplit(self, text1, text2, x, y, deadline):
- """Given the location of the 'middle snake', split the diff in two parts
+ def diff_bisectSplit(self, text1, text2, x, y, deadline):
+ """Given the location of the 'middle snake', split the diff in two parts
and recurse.
Args:
@@ -367,19 +372,19 @@ class diff_match_patch:
Returns:
Array of diff tuples.
"""
- text1a = text1[:x]
- text2a = text2[:y]
- text1b = text1[x:]
- text2b = text2[y:]
-
- # Compute both diffs serially.
- diffs = self.diff_main(text1a, text2a, False, deadline)
- diffsb = self.diff_main(text1b, text2b, False, deadline)
+ text1a = text1[:x]
+ text2a = text2[:y]
+ text1b = text1[x:]
+ text2b = text2[y:]
+
+ # Compute both diffs serially.
+ diffs = self.diff_main(text1a, text2a, False, deadline)
+ diffsb = self.diff_main(text1b, text2b, False, deadline)
- return diffs + diffsb
+ return diffs + diffsb
- def diff_linesToChars(self, text1, text2):
- """Split two texts into an array of strings. Reduce the texts to a string
+ def diff_linesToChars(self, text1, text2):
+ """Split two texts into an array of strings. Reduce the texts to a string
of hashes where each Unicode character represents one line.
Args:
@@ -391,15 +396,15 @@ class diff_match_patch:
the array of unique strings. The zeroth element of the array of unique
strings is intentionally blank.
"""
- lineArray = [] # e.g. lineArray[4] == "Hello\n"
- lineHash = {} # e.g. lineHash["Hello\n"] == 4
+ lineArray = [] # e.g. lineArray[4] == "Hello\n"
+ lineHash = {} # e.g. lineHash["Hello\n"] == 4
- # "\x00" is a valid character, but various debuggers don't like it.
- # So we'll insert a junk entry to avoid generating a null character.
- lineArray.append('')
+ # "\x00" is a valid character, but various debuggers don't like it.
+ # So we'll insert a junk entry to avoid generating a null character.
+ lineArray.append("")
- def diff_linesToCharsMunge(text):
- """Split a text into an array of strings. Reduce the texts to a string
+ def diff_linesToCharsMunge(text):
+ """Split a text into an array of strings. Reduce the texts to a string
of hashes where each Unicode character represents one line.
Modifies linearray and linehash through being a closure.
@@ -409,54 +414,54 @@ class diff_match_patch:
Returns:
Encoded string.
"""
- chars = []
- # Walk the text, pulling out a substring for each line.
- # text.split('\n') would would temporarily double our memory footprint.
- # Modifying text would create many large strings to garbage collect.
- lineStart = 0
- lineEnd = -1
- while lineEnd < len(text) - 1:
- lineEnd = text.find('\n', lineStart)
- if lineEnd == -1:
- lineEnd = len(text) - 1
- line = text[lineStart:lineEnd + 1]
-
- if line in lineHash:
- chars.append(chr(lineHash[line]))
- else:
- if len(lineArray) == maxLines:
- # Bail out at 1114111 because chr(1114112) throws.
- line = text[lineStart:]
- lineEnd = len(text)
- lineArray.append(line)
- lineHash[line] = len(lineArray) - 1
- chars.append(chr(len(lineArray) - 1))
- lineStart = lineEnd + 1
- return "".join(chars)
-
- # Allocate 2/3rds of the space for text1, the rest for text2.
- maxLines = 666666
- chars1 = diff_linesToCharsMunge(text1)
- maxLines = 1114111
- chars2 = diff_linesToCharsMunge(text2)
- return (chars1, chars2, lineArray)
+ chars = []
+ # Walk the text, pulling out a substring for each line.
+ # text.split('\n') would would temporarily double our memory footprint.
+ # Modifying text would create many large strings to garbage collect.
+ lineStart = 0
+ lineEnd = -1
+ while lineEnd < len(text) - 1:
+ lineEnd = text.find("\n", lineStart)
+ if lineEnd == -1:
+ lineEnd = len(text) - 1
+ line = text[lineStart : lineEnd + 1]
+
+ if line in lineHash:
+ chars.append(chr(lineHash[line]))
+ else:
+ if len(lineArray) == maxLines:
+ # Bail out at 1114111 because chr(1114112) throws.
+ line = text[lineStart:]
+ lineEnd = len(text)
+ lineArray.append(line)
+ lineHash[line] = len(lineArray) - 1
+ chars.append(chr(len(lineArray) - 1))
+ lineStart = lineEnd + 1
+ return "".join(chars)
+
+ # Allocate 2/3rds of the space for text1, the rest for text2.
+ maxLines = 666666
+ chars1 = diff_linesToCharsMunge(text1)
+ maxLines = 1114111
+ chars2 = diff_linesToCharsMunge(text2)
+ return (chars1, chars2, lineArray)
- def diff_charsToLines(self, diffs, lineArray):
- """Rehydrate the text in a diff from a string of line hashes to real lines
+ def diff_charsToLines(self, diffs, lineArray):
+ """Rehydrate the text in a diff from a string of line hashes to real lines
of text.
Args:
diffs: Array of diff tuples.
lineArray: Array of unique strings.
"""
- for i in range(len(diffs)):
- text = []
- for char in diffs[i][1]:
- text.append(lineArray[ord(char)])
- diffs[i] = (diffs[i][0], "".join(text))
+ for i in range(len(diffs)):
+ text = []
+ for char in diffs[i][1]:
+ text.append(lineArray[ord(char)])
+ diffs[i] = (diffs[i][0], "".join(text))
- def diff_commonPrefix(self, text1, text2):
- """Determine the common prefix of two strings.
+ def diff_commonPrefix(self, text1, text2):
+ """Determine the common prefix of two strings.
Args:
text1: First string.
@@ -465,26 +470,26 @@ class diff_match_patch:
Returns:
The number of characters common to the start of each string.
"""
- # Quick check for common null cases.
- if not text1 or not text2 or text1[0] != text2[0]:
- return 0
- # Binary search.
- # Performance analysis: https://neil.fraser.name/news/2007/10/09/
- pointermin = 0
- pointermax = min(len(text1), len(text2))
- pointermid = pointermax
- pointerstart = 0
- while pointermin < pointermid:
- if text1[pointerstart:pointermid] == text2[pointerstart:pointermid]:
- pointermin = pointermid
- pointerstart = pointermin
- else:
- pointermax = pointermid
- pointermid = (pointermax - pointermin) // 2 + pointermin
- return pointermid
+ # Quick check for common null cases.
+ if not text1 or not text2 or text1[0] != text2[0]:
+ return 0
+ # Binary search.
+ # Performance analysis: https://neil.fraser.name/news/2007/10/09/
+ pointermin = 0
+ pointermax = min(len(text1), len(text2))
+ pointermid = pointermax
+ pointerstart = 0
+ while pointermin < pointermid:
+ if text1[pointerstart:pointermid] == text2[pointerstart:pointermid]:
+ pointermin = pointermid
+ pointerstart = pointermin
+ else:
+ pointermax = pointermid
+ pointermid = (pointermax - pointermin) // 2 + pointermin
+ return pointermid
- def diff_commonSuffix(self, text1, text2):
- """Determine the common suffix of two strings.
+ def diff_commonSuffix(self, text1, text2):
+ """Determine the common suffix of two strings.
Args:
text1: First string.
@@ -493,27 +498,29 @@ class diff_match_patch:
Returns:
The number of characters common to the end of each string.
"""
- # Quick check for common null cases.
- if not text1 or not text2 or text1[-1] != text2[-1]:
- return 0
- # Binary search.
- # Performance analysis: https://neil.fraser.name/news/2007/10/09/
- pointermin = 0
- pointermax = min(len(text1), len(text2))
- pointermid = pointermax
- pointerend = 0
- while pointermin < pointermid:
- if (text1[-pointermid:len(text1) - pointerend] ==
- text2[-pointermid:len(text2) - pointerend]):
- pointermin = pointermid
- pointerend = pointermin
- else:
- pointermax = pointermid
- pointermid = (pointermax - pointermin) // 2 + pointermin
- return pointermid
+ # Quick check for common null cases.
+ if not text1 or not text2 or text1[-1] != text2[-1]:
+ return 0
+ # Binary search.
+ # Performance analysis: https://neil.fraser.name/news/2007/10/09/
+ pointermin = 0
+ pointermax = min(len(text1), len(text2))
+ pointermid = pointermax
+ pointerend = 0
+ while pointermin < pointermid:
+ if (
+ text1[-pointermid : len(text1) - pointerend]
+ == text2[-pointermid : len(text2) - pointerend]
+ ):
+ pointermin = pointermid
+ pointerend = pointermin
+ else:
+ pointermax = pointermid
+ pointermid = (pointermax - pointermin) // 2 + pointermin
+ return pointermid
- def diff_commonOverlap(self, text1, text2):
- """Determine if the suffix of one string is the prefix of another.
+ def diff_commonOverlap(self, text1, text2):
+ """Determine if the suffix of one string is the prefix of another.
Args:
text1 First string.
@@ -523,39 +530,39 @@ class diff_match_patch:
The number of characters common to the end of the first
string and the start of the second string.
"""
- # Cache the text lengths to prevent multiple calls.
- text1_length = len(text1)
- text2_length = len(text2)
- # Eliminate the null case.
- if text1_length == 0 or text2_length == 0:
- return 0
- # Truncate the longer string.
- if text1_length > text2_length:
- text1 = text1[-text2_length:]
- elif text1_length < text2_length:
- text2 = text2[:text1_length]
- text_length = min(text1_length, text2_length)
- # Quick check for the worst case.
- if text1 == text2:
- return text_length
-
- # Start by looking for a single character match
- # and increase length until no match is found.
- # Performance analysis: https://neil.fraser.name/news/2010/11/04/
- best = 0
- length = 1
- while True:
- pattern = text1[-length:]
- found = text2.find(pattern)
- if found == -1:
- return best
- length += found
- if found == 0 or text1[-length:] == text2[:length]:
- best = length
- length += 1
+ # Cache the text lengths to prevent multiple calls.
+ text1_length = len(text1)
+ text2_length = len(text2)
+ # Eliminate the null case.
+ if text1_length == 0 or text2_length == 0:
+ return 0
+ # Truncate the longer string.
+ if text1_length > text2_length:
+ text1 = text1[-text2_length:]
+ elif text1_length < text2_length:
+ text2 = text2[:text1_length]
+ text_length = min(text1_length, text2_length)
+ # Quick check for the worst case.
+ if text1 == text2:
+ return text_length
+
+ # Start by looking for a single character match
+ # and increase length until no match is found.
+ # Performance analysis: https://neil.fraser.name/news/2010/11/04/
+ best = 0
+ length = 1
+ while True:
+ pattern = text1[-length:]
+ found = text2.find(pattern)
+ if found == -1:
+ return best
+ length += found
+ if found == 0 or text1[-length:] == text2[:length]:
+ best = length
+ length += 1
- def diff_halfMatch(self, text1, text2):
- """Do the two texts share a substring which is at least half the length of
+ def diff_halfMatch(self, text1, text2):
+ """Do the two texts share a substring which is at least half the length of
the longer text?
This speedup can produce non-minimal diffs.
@@ -568,18 +575,18 @@ class diff_match_patch:
the prefix of text2, the suffix of text2 and the common middle. Or None
if there was no match.
"""
- if self.Diff_Timeout <= 0:
- # Don't risk returning a non-optimal diff if we have unlimited time.
- return None
- if len(text1) > len(text2):
- (longtext, shorttext) = (text1, text2)
- else:
- (shorttext, longtext) = (text1, text2)
- if len(longtext) < 4 or len(shorttext) * 2 < len(longtext):
- return None # Pointless.
+ if self.Diff_Timeout <= 0:
+ # Don't risk returning a non-optimal diff if we have unlimited time.
+ return None
+ if len(text1) > len(text2):
+ (longtext, shorttext) = (text1, text2)
+ else:
+ (shorttext, longtext) = (text1, text2)
+ if len(longtext) < 4 or len(shorttext) * 2 < len(longtext):
+ return None # Pointless.
- def diff_halfMatchI(longtext, shorttext, i):
- """Does a substring of shorttext exist within longtext such that the
+ def diff_halfMatchI(longtext, shorttext, i):
+ """Does a substring of shorttext exist within longtext such that the
substring is at least half the length of longtext?
Closure, but does not reference any external variables.
@@ -593,148 +600,181 @@ class diff_match_patch:
longtext, the prefix of shorttext, the suffix of shorttext and the
common middle. Or None if there was no match.
"""
- seed = longtext[i:i + len(longtext) // 4]
- best_common = ''
- j = shorttext.find(seed)
- while j != -1:
- prefixLength = self.diff_commonPrefix(longtext[i:], shorttext[j:])
- suffixLength = self.diff_commonSuffix(longtext[:i], shorttext[:j])
- if len(best_common) < suffixLength + prefixLength:
- best_common = (shorttext[j - suffixLength:j] +
- shorttext[j:j + prefixLength])
- best_longtext_a = longtext[:i - suffixLength]
- best_longtext_b = longtext[i + prefixLength:]
- best_shorttext_a = shorttext[:j - suffixLength]
- best_shorttext_b = shorttext[j + prefixLength:]
- j = shorttext.find(seed, j + 1)
-
- if len(best_common) * 2 >= len(longtext):
- return (best_longtext_a, best_longtext_b,
- best_shorttext_a, best_shorttext_b, best_common)
- else:
- return None
-
- # First check if the second quarter is the seed for a half-match.
- hm1 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 3) // 4)
- # Check again based on the third quarter.
- hm2 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 1) // 2)
- if not hm1 and not hm2:
- return None
- elif not hm2:
- hm = hm1
- elif not hm1:
- hm = hm2
- else:
- # Both matched. Select the longest.
- if len(hm1[4]) > len(hm2[4]):
- hm = hm1
- else:
- hm = hm2
-
- # A half-match was found, sort out the return data.
- if len(text1) > len(text2):
- (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
- else:
- (text2_a, text2_b, text1_a, text1_b, mid_common) = hm
- return (text1_a, text1_b, text2_a, text2_b, mid_common)
+ seed = longtext[i : i + len(longtext) // 4]
+ best_common = ""
+ j = shorttext.find(seed)
+ while j != -1:
+ prefixLength = self.diff_commonPrefix(longtext[i:], shorttext[j:])
+ suffixLength = self.diff_commonSuffix(longtext[:i], shorttext[:j])
+ if len(best_common) < suffixLength + prefixLength:
+ best_common = (
+ shorttext[j - suffixLength : j]
+ + shorttext[j : j + prefixLength]
+ )
+ best_longtext_a = longtext[: i - suffixLength]
+ best_longtext_b = longtext[i + prefixLength :]
+ best_shorttext_a = shorttext[: j - suffixLength]
+ best_shorttext_b = shorttext[j + prefixLength :]
+ j = shorttext.find(seed, j + 1)
+
+ if len(best_common) * 2 >= len(longtext):
+ return (
+ best_longtext_a,
+ best_longtext_b,
+ best_shorttext_a,
+ best_shorttext_b,
+ best_common,
+ )
+ else:
+ return None
+
+ # First check if the second quarter is the seed for a half-match.
+ hm1 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 3) // 4)
+ # Check again based on the third quarter.
+ hm2 = diff_halfMatchI(longtext, shorttext, (len(longtext) + 1) // 2)
+ if not hm1 and not hm2:
+ return None
+ elif not hm2:
+ hm = hm1
+ elif not hm1:
+ hm = hm2
+ else:
+ # Both matched. Select the longest.
+ if len(hm1[4]) > len(hm2[4]):
+ hm = hm1
+ else:
+ hm = hm2
- def diff_cleanupSemantic(self, diffs):
- """Reduce the number of edits by eliminating semantically trivial
+ # A half-match was found, sort out the return data.
+ if len(text1) > len(text2):
+ (text1_a, text1_b, text2_a, text2_b, mid_common) = hm
+ else:
+ (text2_a, text2_b, text1_a, text1_b, mid_common) = hm
+ return (text1_a, text1_b, text2_a, text2_b, mid_common)
+
+ def diff_cleanupSemantic(self, diffs):
+ """Reduce the number of edits by eliminating semantically trivial
equalities.
Args:
diffs: Array of diff tuples.
"""
- changes = False
- equalities = [] # Stack of indices where equalities are found.
- lastEquality = None # Always equal to diffs[equalities[-1]][1]
- pointer = 0 # Index of current position.
- # Number of chars that changed prior to the equality.
- length_insertions1, length_deletions1 = 0, 0
- # Number of chars that changed after the equality.
- length_insertions2, length_deletions2 = 0, 0
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
- equalities.append(pointer)
- length_insertions1, length_insertions2 = length_insertions2, 0
- length_deletions1, length_deletions2 = length_deletions2, 0
- lastEquality = diffs[pointer][1]
- else: # An insertion or deletion.
- if diffs[pointer][0] == self.DIFF_INSERT:
- length_insertions2 += len(diffs[pointer][1])
- else:
- length_deletions2 += len(diffs[pointer][1])
- # Eliminate an equality that is smaller or equal to the edits on both
- # sides of it.
- if (lastEquality and (len(lastEquality) <=
- max(length_insertions1, length_deletions1)) and
- (len(lastEquality) <= max(length_insertions2, length_deletions2))):
- # Duplicate record.
- diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
- # Change second copy to insert.
- diffs[equalities[-1] + 1] = (self.DIFF_INSERT,
- diffs[equalities[-1] + 1][1])
- # Throw away the equality we just deleted.
- equalities.pop()
- # Throw away the previous equality (it needs to be reevaluated).
- if len(equalities):
- equalities.pop()
- if len(equalities):
- pointer = equalities[-1]
- else:
- pointer = -1
- # Reset the counters.
- length_insertions1, length_deletions1 = 0, 0
- length_insertions2, length_deletions2 = 0, 0
- lastEquality = None
- changes = True
- pointer += 1
-
- # Normalize the diff.
- if changes:
- self.diff_cleanupMerge(diffs)
- self.diff_cleanupSemanticLossless(diffs)
-
- # Find any overlaps between deletions and insertions.
- # e.g: <del>abcxxx</del><ins>xxxdef</ins>
- # -> <del>abc</del>xxx<ins>def</ins>
- # e.g: <del>xxxabc</del><ins>defxxx</ins>
- # -> <ins>def</ins>xxx<del>abc</del>
- # Only extract an overlap if it is as big as the edit ahead or behind it.
- pointer = 1
- while pointer < len(diffs):
- if (diffs[pointer - 1][0] == self.DIFF_DELETE and
- diffs[pointer][0] == self.DIFF_INSERT):
- deletion = diffs[pointer - 1][1]
- insertion = diffs[pointer][1]
- overlap_length1 = self.diff_commonOverlap(deletion, insertion)
- overlap_length2 = self.diff_commonOverlap(insertion, deletion)
- if overlap_length1 >= overlap_length2:
- if (overlap_length1 >= len(deletion) / 2.0 or
- overlap_length1 >= len(insertion) / 2.0):
- # Overlap found. Insert an equality and trim the surrounding edits.
- diffs.insert(pointer, (self.DIFF_EQUAL,
- insertion[:overlap_length1]))
- diffs[pointer - 1] = (self.DIFF_DELETE,
- deletion[:len(deletion) - overlap_length1])
- diffs[pointer + 1] = (self.DIFF_INSERT,
- insertion[overlap_length1:])
+ changes = False
+ equalities = [] # Stack of indices where equalities are found.
+ lastEquality = None # Always equal to diffs[equalities[-1]][1]
+ pointer = 0 # Index of current position.
+ # Number of chars that changed prior to the equality.
+ length_insertions1, length_deletions1 = 0, 0
+ # Number of chars that changed after the equality.
+ length_insertions2, length_deletions2 = 0, 0
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
+ equalities.append(pointer)
+ length_insertions1, length_insertions2 = length_insertions2, 0
+ length_deletions1, length_deletions2 = length_deletions2, 0
+ lastEquality = diffs[pointer][1]
+ else: # An insertion or deletion.
+ if diffs[pointer][0] == self.DIFF_INSERT:
+ length_insertions2 += len(diffs[pointer][1])
+ else:
+ length_deletions2 += len(diffs[pointer][1])
+ # Eliminate an equality that is smaller or equal to the edits on both
+ # sides of it.
+ if (
+ lastEquality
+ and (
+ len(lastEquality) <= max(length_insertions1, length_deletions1)
+ )
+ and (
+ len(lastEquality) <= max(length_insertions2, length_deletions2)
+ )
+ ):
+ # Duplicate record.
+ diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
+ # Change second copy to insert.
+ diffs[equalities[-1] + 1] = (
+ self.DIFF_INSERT,
+ diffs[equalities[-1] + 1][1],
+ )
+ # Throw away the equality we just deleted.
+ equalities.pop()
+ # Throw away the previous equality (it needs to be reevaluated).
+ if len(equalities):
+ equalities.pop()
+ if len(equalities):
+ pointer = equalities[-1]
+ else:
+ pointer = -1
+ # Reset the counters.
+ length_insertions1, length_deletions1 = 0, 0
+ length_insertions2, length_deletions2 = 0, 0
+ lastEquality = None
+ changes = True
pointer += 1
- else:
- if (overlap_length2 >= len(deletion) / 2.0 or
- overlap_length2 >= len(insertion) / 2.0):
- # Reverse overlap found.
- # Insert an equality and swap and trim the surrounding edits.
- diffs.insert(pointer, (self.DIFF_EQUAL, deletion[:overlap_length2]))
- diffs[pointer - 1] = (self.DIFF_INSERT,
- insertion[:len(insertion) - overlap_length2])
- diffs[pointer + 1] = (self.DIFF_DELETE, deletion[overlap_length2:])
+
+ # Normalize the diff.
+ if changes:
+ self.diff_cleanupMerge(diffs)
+ self.diff_cleanupSemanticLossless(diffs)
+
+ # Find any overlaps between deletions and insertions.
+ # e.g: <del>abcxxx</del><ins>xxxdef</ins>
+ # -> <del>abc</del>xxx<ins>def</ins>
+ # e.g: <del>xxxabc</del><ins>defxxx</ins>
+ # -> <ins>def</ins>xxx<del>abc</del>
+ # Only extract an overlap if it is as big as the edit ahead or behind it.
+ pointer = 1
+ while pointer < len(diffs):
+ if (
+ diffs[pointer - 1][0] == self.DIFF_DELETE
+ and diffs[pointer][0] == self.DIFF_INSERT
+ ):
+ deletion = diffs[pointer - 1][1]
+ insertion = diffs[pointer][1]
+ overlap_length1 = self.diff_commonOverlap(deletion, insertion)
+ overlap_length2 = self.diff_commonOverlap(insertion, deletion)
+ if overlap_length1 >= overlap_length2:
+ if (
+ overlap_length1 >= len(deletion) / 2.0
+ or overlap_length1 >= len(insertion) / 2.0
+ ):
+ # Overlap found. Insert an equality and trim the surrounding edits.
+ diffs.insert(
+ pointer, (self.DIFF_EQUAL, insertion[:overlap_length1])
+ )
+ diffs[pointer - 1] = (
+ self.DIFF_DELETE,
+ deletion[: len(deletion) - overlap_length1],
+ )
+ diffs[pointer + 1] = (
+ self.DIFF_INSERT,
+ insertion[overlap_length1:],
+ )
+ pointer += 1
+ else:
+ if (
+ overlap_length2 >= len(deletion) / 2.0
+ or overlap_length2 >= len(insertion) / 2.0
+ ):
+ # Reverse overlap found.
+ # Insert an equality and swap and trim the surrounding edits.
+ diffs.insert(
+ pointer, (self.DIFF_EQUAL, deletion[:overlap_length2])
+ )
+ diffs[pointer - 1] = (
+ self.DIFF_INSERT,
+ insertion[: len(insertion) - overlap_length2],
+ )
+ diffs[pointer + 1] = (
+ self.DIFF_DELETE,
+ deletion[overlap_length2:],
+ )
+ pointer += 1
+ pointer += 1
pointer += 1
- pointer += 1
- pointer += 1
- def diff_cleanupSemanticLossless(self, diffs):
- """Look for single edits surrounded on both sides by equalities
+ def diff_cleanupSemanticLossless(self, diffs):
+ """Look for single edits surrounded on both sides by equalities
which can be shifted sideways to align the edit to a word boundary.
e.g: The c<ins>at c</ins>ame. -> The <ins>cat </ins>came.
@@ -742,8 +782,8 @@ class diff_match_patch:
diffs: Array of diff tuples.
"""
- def diff_cleanupSemanticScore(one, two):
- """Given two strings, compute a score representing whether the
+ def diff_cleanupSemanticScore(one, two):
+ """Given two strings, compute a score representing whether the
internal boundary falls on logical boundaries.
Scores range from 6 (best) to 0 (worst).
Closure, but does not reference any external variables.
@@ -755,277 +795,306 @@ class diff_match_patch:
Returns:
The score.
"""
- if not one or not two:
- # Edges are the best.
- return 6
-
- # Each port of this function behaves slightly differently due to
- # subtle differences in each language's definition of things like
- # 'whitespace'. Since this function's purpose is largely cosmetic,
- # the choice has been made to use each language's native features
- # rather than force total conformity.
- char1 = one[-1]
- char2 = two[0]
- nonAlphaNumeric1 = not char1.isalnum()
- nonAlphaNumeric2 = not char2.isalnum()
- whitespace1 = nonAlphaNumeric1 and char1.isspace()
- whitespace2 = nonAlphaNumeric2 and char2.isspace()
- lineBreak1 = whitespace1 and (char1 == "\r" or char1 == "\n")
- lineBreak2 = whitespace2 and (char2 == "\r" or char2 == "\n")
- blankLine1 = lineBreak1 and self.BLANKLINEEND.search(one)
- blankLine2 = lineBreak2 and self.BLANKLINESTART.match(two)
-
- if blankLine1 or blankLine2:
- # Five points for blank lines.
- return 5
- elif lineBreak1 or lineBreak2:
- # Four points for line breaks.
- return 4
- elif nonAlphaNumeric1 and not whitespace1 and whitespace2:
- # Three points for end of sentences.
- return 3
- elif whitespace1 or whitespace2:
- # Two points for whitespace.
- return 2
- elif nonAlphaNumeric1 or nonAlphaNumeric2:
- # One point for non-alphanumeric.
- return 1
- return 0
-
- pointer = 1
- # Intentionally ignore the first and last element (don't need checking).
- while pointer < len(diffs) - 1:
- if (diffs[pointer - 1][0] == self.DIFF_EQUAL and
- diffs[pointer + 1][0] == self.DIFF_EQUAL):
- # This is a single edit surrounded by equalities.
- equality1 = diffs[pointer - 1][1]
- edit = diffs[pointer][1]
- equality2 = diffs[pointer + 1][1]
-
- # First, shift the edit as far left as possible.
- commonOffset = self.diff_commonSuffix(equality1, edit)
- if commonOffset:
- commonString = edit[-commonOffset:]
- equality1 = equality1[:-commonOffset]
- edit = commonString + edit[:-commonOffset]
- equality2 = commonString + equality2
-
- # Second, step character by character right, looking for the best fit.
- bestEquality1 = equality1
- bestEdit = edit
- bestEquality2 = equality2
- bestScore = (diff_cleanupSemanticScore(equality1, edit) +
- diff_cleanupSemanticScore(edit, equality2))
- while edit and equality2 and edit[0] == equality2[0]:
- equality1 += edit[0]
- edit = edit[1:] + equality2[0]
- equality2 = equality2[1:]
- score = (diff_cleanupSemanticScore(equality1, edit) +
- diff_cleanupSemanticScore(edit, equality2))
- # The >= encourages trailing rather than leading whitespace on edits.
- if score >= bestScore:
- bestScore = score
- bestEquality1 = equality1
- bestEdit = edit
- bestEquality2 = equality2
-
- if diffs[pointer - 1][1] != bestEquality1:
- # We have an improvement, save it back to the diff.
- if bestEquality1:
- diffs[pointer - 1] = (diffs[pointer - 1][0], bestEquality1)
- else:
- del diffs[pointer - 1]
- pointer -= 1
- diffs[pointer] = (diffs[pointer][0], bestEdit)
- if bestEquality2:
- diffs[pointer + 1] = (diffs[pointer + 1][0], bestEquality2)
- else:
- del diffs[pointer + 1]
- pointer -= 1
- pointer += 1
-
- # Define some regex patterns for matching boundaries.
- BLANKLINEEND = re.compile(r"\n\r?\n$")
- BLANKLINESTART = re.compile(r"^\r?\n\r?\n")
+ if not one or not two:
+ # Edges are the best.
+ return 6
+
+ # Each port of this function behaves slightly differently due to
+ # subtle differences in each language's definition of things like
+ # 'whitespace'. Since this function's purpose is largely cosmetic,
+ # the choice has been made to use each language's native features
+ # rather than force total conformity.
+ char1 = one[-1]
+ char2 = two[0]
+ nonAlphaNumeric1 = not char1.isalnum()
+ nonAlphaNumeric2 = not char2.isalnum()
+ whitespace1 = nonAlphaNumeric1 and char1.isspace()
+ whitespace2 = nonAlphaNumeric2 and char2.isspace()
+ lineBreak1 = whitespace1 and (char1 == "\r" or char1 == "\n")
+ lineBreak2 = whitespace2 and (char2 == "\r" or char2 == "\n")
+ blankLine1 = lineBreak1 and self.BLANKLINEEND.search(one)
+ blankLine2 = lineBreak2 and self.BLANKLINESTART.match(two)
+
+ if blankLine1 or blankLine2:
+ # Five points for blank lines.
+ return 5
+ elif lineBreak1 or lineBreak2:
+ # Four points for line breaks.
+ return 4
+ elif nonAlphaNumeric1 and not whitespace1 and whitespace2:
+ # Three points for end of sentences.
+ return 3
+ elif whitespace1 or whitespace2:
+ # Two points for whitespace.
+ return 2
+ elif nonAlphaNumeric1 or nonAlphaNumeric2:
+ # One point for non-alphanumeric.
+ return 1
+ return 0
+
+ pointer = 1
+ # Intentionally ignore the first and last element (don't need checking).
+ while pointer < len(diffs) - 1:
+ if (
+ diffs[pointer - 1][0] == self.DIFF_EQUAL
+ and diffs[pointer + 1][0] == self.DIFF_EQUAL
+ ):
+ # This is a single edit surrounded by equalities.
+ equality1 = diffs[pointer - 1][1]
+ edit = diffs[pointer][1]
+ equality2 = diffs[pointer + 1][1]
+
+ # First, shift the edit as far left as possible.
+ commonOffset = self.diff_commonSuffix(equality1, edit)
+ if commonOffset:
+ commonString = edit[-commonOffset:]
+ equality1 = equality1[:-commonOffset]
+ edit = commonString + edit[:-commonOffset]
+ equality2 = commonString + equality2
+
+ # Second, step character by character right, looking for the best fit.
+ bestEquality1 = equality1
+ bestEdit = edit
+ bestEquality2 = equality2
+ bestScore = diff_cleanupSemanticScore(
+ equality1, edit
+ ) + diff_cleanupSemanticScore(edit, equality2)
+ while edit and equality2 and edit[0] == equality2[0]:
+ equality1 += edit[0]
+ edit = edit[1:] + equality2[0]
+ equality2 = equality2[1:]
+ score = diff_cleanupSemanticScore(
+ equality1, edit
+ ) + diff_cleanupSemanticScore(edit, equality2)
+ # The >= encourages trailing rather than leading whitespace on edits.
+ if score >= bestScore:
+ bestScore = score
+ bestEquality1 = equality1
+ bestEdit = edit
+ bestEquality2 = equality2
+
+ if diffs[pointer - 1][1] != bestEquality1:
+ # We have an improvement, save it back to the diff.
+ if bestEquality1:
+ diffs[pointer - 1] = (diffs[pointer - 1][0], bestEquality1)
+ else:
+ del diffs[pointer - 1]
+ pointer -= 1
+ diffs[pointer] = (diffs[pointer][0], bestEdit)
+ if bestEquality2:
+ diffs[pointer + 1] = (diffs[pointer + 1][0], bestEquality2)
+ else:
+ del diffs[pointer + 1]
+ pointer -= 1
+ pointer += 1
+
+ # Define some regex patterns for matching boundaries.
+ BLANKLINEEND = re.compile(r"\n\r?\n$")
+ BLANKLINESTART = re.compile(r"^\r?\n\r?\n")
- def diff_cleanupEfficiency(self, diffs):
- """Reduce the number of edits by eliminating operationally trivial
+ def diff_cleanupEfficiency(self, diffs):
+ """Reduce the number of edits by eliminating operationally trivial
equalities.
Args:
diffs: Array of diff tuples.
"""
- changes = False
- equalities = [] # Stack of indices where equalities are found.
- lastEquality = None # Always equal to diffs[equalities[-1]][1]
- pointer = 0 # Index of current position.
- pre_ins = False # Is there an insertion operation before the last equality.
- pre_del = False # Is there a deletion operation before the last equality.
- post_ins = False # Is there an insertion operation after the last equality.
- post_del = False # Is there a deletion operation after the last equality.
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
- if (len(diffs[pointer][1]) < self.Diff_EditCost and
- (post_ins or post_del)):
- # Candidate found.
- equalities.append(pointer)
- pre_ins = post_ins
- pre_del = post_del
- lastEquality = diffs[pointer][1]
- else:
- # Not a candidate, and can never become one.
- equalities = []
- lastEquality = None
-
- post_ins = post_del = False
- else: # An insertion or deletion.
- if diffs[pointer][0] == self.DIFF_DELETE:
- post_del = True
- else:
- post_ins = True
-
- # Five types to be split:
- # <ins>A</ins><del>B</del>XY<ins>C</ins><del>D</del>
- # <ins>A</ins>X<ins>C</ins><del>D</del>
- # <ins>A</ins><del>B</del>X<ins>C</ins>
- # <ins>A</del>X<ins>C</ins><del>D</del>
- # <ins>A</ins><del>B</del>X<del>C</del>
-
- if lastEquality and ((pre_ins and pre_del and post_ins and post_del) or
- ((len(lastEquality) < self.Diff_EditCost / 2) and
- (pre_ins + pre_del + post_ins + post_del) == 3)):
- # Duplicate record.
- diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
- # Change second copy to insert.
- diffs[equalities[-1] + 1] = (self.DIFF_INSERT,
- diffs[equalities[-1] + 1][1])
- equalities.pop() # Throw away the equality we just deleted.
- lastEquality = None
- if pre_ins and pre_del:
- # No changes made which could affect previous entry, keep going.
- post_ins = post_del = True
- equalities = []
- else:
- if len(equalities):
- equalities.pop() # Throw away the previous equality.
- if len(equalities):
- pointer = equalities[-1]
- else:
- pointer = -1
- post_ins = post_del = False
- changes = True
- pointer += 1
+ changes = False
+ equalities = [] # Stack of indices where equalities are found.
+ lastEquality = None # Always equal to diffs[equalities[-1]][1]
+ pointer = 0 # Index of current position.
+ pre_ins = False # Is there an insertion operation before the last equality.
+ pre_del = False # Is there a deletion operation before the last equality.
+ post_ins = False # Is there an insertion operation after the last equality.
+ post_del = False # Is there a deletion operation after the last equality.
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_EQUAL: # Equality found.
+ if len(diffs[pointer][1]) < self.Diff_EditCost and (
+ post_ins or post_del
+ ):
+ # Candidate found.
+ equalities.append(pointer)
+ pre_ins = post_ins
+ pre_del = post_del
+ lastEquality = diffs[pointer][1]
+ else:
+ # Not a candidate, and can never become one.
+ equalities = []
+ lastEquality = None
+
+ post_ins = post_del = False
+ else: # An insertion or deletion.
+ if diffs[pointer][0] == self.DIFF_DELETE:
+ post_del = True
+ else:
+ post_ins = True
+
+ # Five types to be split:
+ # <ins>A</ins><del>B</del>XY<ins>C</ins><del>D</del>
+ # <ins>A</ins>X<ins>C</ins><del>D</del>
+ # <ins>A</ins><del>B</del>X<ins>C</ins>
+ # <ins>A</del>X<ins>C</ins><del>D</del>
+ # <ins>A</ins><del>B</del>X<del>C</del>
+
+ if lastEquality and (
+ (pre_ins and pre_del and post_ins and post_del)
+ or (
+ (len(lastEquality) < self.Diff_EditCost / 2)
+ and (pre_ins + pre_del + post_ins + post_del) == 3
+ )
+ ):
+ # Duplicate record.
+ diffs.insert(equalities[-1], (self.DIFF_DELETE, lastEquality))
+ # Change second copy to insert.
+ diffs[equalities[-1] + 1] = (
+ self.DIFF_INSERT,
+ diffs[equalities[-1] + 1][1],
+ )
+ equalities.pop() # Throw away the equality we just deleted.
+ lastEquality = None
+ if pre_ins and pre_del:
+ # No changes made which could affect previous entry, keep going.
+ post_ins = post_del = True
+ equalities = []
+ else:
+ if len(equalities):
+ equalities.pop() # Throw away the previous equality.
+ if len(equalities):
+ pointer = equalities[-1]
+ else:
+ pointer = -1
+ post_ins = post_del = False
+ changes = True
+ pointer += 1
- if changes:
- self.diff_cleanupMerge(diffs)
+ if changes:
+ self.diff_cleanupMerge(diffs)
- def diff_cleanupMerge(self, diffs):
- """Reorder and merge like edit sections. Merge equalities.
+ def diff_cleanupMerge(self, diffs):
+ """Reorder and merge like edit sections. Merge equalities.
Any edit section can move as long as it doesn't cross an equality.
Args:
diffs: Array of diff tuples.
"""
- diffs.append((self.DIFF_EQUAL, '')) # Add a dummy entry at the end.
- pointer = 0
- count_delete = 0
- count_insert = 0
- text_delete = ''
- text_insert = ''
- while pointer < len(diffs):
- if diffs[pointer][0] == self.DIFF_INSERT:
- count_insert += 1
- text_insert += diffs[pointer][1]
- pointer += 1
- elif diffs[pointer][0] == self.DIFF_DELETE:
- count_delete += 1
- text_delete += diffs[pointer][1]
- pointer += 1
- elif diffs[pointer][0] == self.DIFF_EQUAL:
- # Upon reaching an equality, check for prior redundancies.
- if count_delete + count_insert > 1:
- if count_delete != 0 and count_insert != 0:
- # Factor out any common prefixies.
- commonlength = self.diff_commonPrefix(text_insert, text_delete)
- if commonlength != 0:
- x = pointer - count_delete - count_insert - 1
- if x >= 0 and diffs[x][0] == self.DIFF_EQUAL:
- diffs[x] = (diffs[x][0], diffs[x][1] +
- text_insert[:commonlength])
- else:
- diffs.insert(0, (self.DIFF_EQUAL, text_insert[:commonlength]))
- pointer += 1
- text_insert = text_insert[commonlength:]
- text_delete = text_delete[commonlength:]
- # Factor out any common suffixes.
- commonlength = self.diff_commonSuffix(text_insert, text_delete)
- if commonlength != 0:
- diffs[pointer] = (diffs[pointer][0], text_insert[-commonlength:] +
- diffs[pointer][1])
- text_insert = text_insert[:-commonlength]
- text_delete = text_delete[:-commonlength]
- # Delete the offending records and add the merged ones.
- new_ops = []
- if len(text_delete) != 0:
- new_ops.append((self.DIFF_DELETE, text_delete))
- if len(text_insert) != 0:
- new_ops.append((self.DIFF_INSERT, text_insert))
- pointer -= count_delete + count_insert
- diffs[pointer : pointer + count_delete + count_insert] = new_ops
- pointer += len(new_ops) + 1
- elif pointer != 0 and diffs[pointer - 1][0] == self.DIFF_EQUAL:
- # Merge this equality with the previous one.
- diffs[pointer - 1] = (diffs[pointer - 1][0],
- diffs[pointer - 1][1] + diffs[pointer][1])
- del diffs[pointer]
- else:
- pointer += 1
-
- count_insert = 0
+ diffs.append((self.DIFF_EQUAL, "")) # Add a dummy entry at the end.
+ pointer = 0
count_delete = 0
- text_delete = ''
- text_insert = ''
-
- if diffs[-1][1] == '':
- diffs.pop() # Remove the dummy entry at the end.
+ count_insert = 0
+ text_delete = ""
+ text_insert = ""
+ while pointer < len(diffs):
+ if diffs[pointer][0] == self.DIFF_INSERT:
+ count_insert += 1
+ text_insert += diffs[pointer][1]
+ pointer += 1
+ elif diffs[pointer][0] == self.DIFF_DELETE:
+ count_delete += 1
+ text_delete += diffs[pointer][1]
+ pointer += 1
+ elif diffs[pointer][0] == self.DIFF_EQUAL:
+ # Upon reaching an equality, check for prior redundancies.
+ if count_delete + count_insert > 1:
+ if count_delete != 0 and count_insert != 0:
+ # Factor out any common prefixies.
+ commonlength = self.diff_commonPrefix(text_insert, text_delete)
+ if commonlength != 0:
+ x = pointer - count_delete - count_insert - 1
+ if x >= 0 and diffs[x][0] == self.DIFF_EQUAL:
+ diffs[x] = (
+ diffs[x][0],
+ diffs[x][1] + text_insert[:commonlength],
+ )
+ else:
+ diffs.insert(
+ 0, (self.DIFF_EQUAL, text_insert[:commonlength])
+ )
+ pointer += 1
+ text_insert = text_insert[commonlength:]
+ text_delete = text_delete[commonlength:]
+ # Factor out any common suffixes.
+ commonlength = self.diff_commonSuffix(text_insert, text_delete)
+ if commonlength != 0:
+ diffs[pointer] = (
+ diffs[pointer][0],
+ text_insert[-commonlength:] + diffs[pointer][1],
+ )
+ text_insert = text_insert[:-commonlength]
+ text_delete = text_delete[:-commonlength]
+ # Delete the offending records and add the merged ones.
+ new_ops = []
+ if len(text_delete) != 0:
+ new_ops.append((self.DIFF_DELETE, text_delete))
+ if len(text_insert) != 0:
+ new_ops.append((self.DIFF_INSERT, text_insert))
+ pointer -= count_delete + count_insert
+ diffs[pointer : pointer + count_delete + count_insert] = new_ops
+ pointer += len(new_ops) + 1
+ elif pointer != 0 and diffs[pointer - 1][0] == self.DIFF_EQUAL:
+ # Merge this equality with the previous one.
+ diffs[pointer - 1] = (
+ diffs[pointer - 1][0],
+ diffs[pointer - 1][1] + diffs[pointer][1],
+ )
+ del diffs[pointer]
+ else:
+ pointer += 1
+
+ count_insert = 0
+ count_delete = 0
+ text_delete = ""
+ text_insert = ""
+
+ if diffs[-1][1] == "":
+ diffs.pop() # Remove the dummy entry at the end.
+
+ # Second pass: look for single edits surrounded on both sides by equalities
+ # which can be shifted sideways to eliminate an equality.
+ # e.g: A<ins>BA</ins>C -> <ins>AB</ins>AC
+ changes = False
+ pointer = 1
+ # Intentionally ignore the first and last element (don't need checking).
+ while pointer < len(diffs) - 1:
+ if (
+ diffs[pointer - 1][0] == self.DIFF_EQUAL
+ and diffs[pointer + 1][0] == self.DIFF_EQUAL
+ ):
+ # This is a single edit surrounded by equalities.
+ if diffs[pointer][1].endswith(diffs[pointer - 1][1]):
+ # Shift the edit over the previous equality.
+ if diffs[pointer - 1][1] != "":
+ diffs[pointer] = (
+ diffs[pointer][0],
+ diffs[pointer - 1][1]
+ + diffs[pointer][1][: -len(diffs[pointer - 1][1])],
+ )
+ diffs[pointer + 1] = (
+ diffs[pointer + 1][0],
+ diffs[pointer - 1][1] + diffs[pointer + 1][1],
+ )
+ del diffs[pointer - 1]
+ changes = True
+ elif diffs[pointer][1].startswith(diffs[pointer + 1][1]):
+ # Shift the edit over the next equality.
+ diffs[pointer - 1] = (
+ diffs[pointer - 1][0],
+ diffs[pointer - 1][1] + diffs[pointer + 1][1],
+ )
+ diffs[pointer] = (
+ diffs[pointer][0],
+ diffs[pointer][1][len(diffs[pointer + 1][1]) :]
+ + diffs[pointer + 1][1],
+ )
+ del diffs[pointer + 1]
+ changes = True
+ pointer += 1
- # Second pass: look for single edits surrounded on both sides by equalities
- # which can be shifted sideways to eliminate an equality.
- # e.g: A<ins>BA</ins>C -> <ins>AB</ins>AC
- changes = False
- pointer = 1
- # Intentionally ignore the first and last element (don't need checking).
- while pointer < len(diffs) - 1:
- if (diffs[pointer - 1][0] == self.DIFF_EQUAL and
- diffs[pointer + 1][0] == self.DIFF_EQUAL):
- # This is a single edit surrounded by equalities.
- if diffs[pointer][1].endswith(diffs[pointer - 1][1]):
- # Shift the edit over the previous equality.
- if diffs[pointer - 1][1] != "":
- diffs[pointer] = (diffs[pointer][0],
- diffs[pointer - 1][1] +
- diffs[pointer][1][:-len(diffs[pointer - 1][1])])
- diffs[pointer + 1] = (diffs[pointer + 1][0],
- diffs[pointer - 1][1] + diffs[pointer + 1][1])
- del diffs[pointer - 1]
- changes = True
- elif diffs[pointer][1].startswith(diffs[pointer + 1][1]):
- # Shift the edit over the next equality.
- diffs[pointer - 1] = (diffs[pointer - 1][0],
- diffs[pointer - 1][1] + diffs[pointer + 1][1])
- diffs[pointer] = (diffs[pointer][0],
- diffs[pointer][1][len(diffs[pointer + 1][1]):] +
- diffs[pointer + 1][1])
- del diffs[pointer + 1]
- changes = True
- pointer += 1
-
- # If shifts were made, the diff needs reordering and another shift sweep.
- if changes:
- self.diff_cleanupMerge(diffs)
+ # If shifts were made, the diff needs reordering and another shift sweep.
+ if changes:
+ self.diff_cleanupMerge(diffs)
- def diff_xIndex(self, diffs, loc):
- """loc is a location in text1, compute and return the equivalent location
+ def diff_xIndex(self, diffs, loc):
+ """loc is a location in text1, compute and return the equivalent location
in text2. e.g. "The cat" vs "The big cat", 1->1, 5->8
Args:
@@ -1035,29 +1104,29 @@ class diff_match_patch:
Returns:
Location within text2.
"""
- chars1 = 0
- chars2 = 0
- last_chars1 = 0
- last_chars2 = 0
- for x in range(len(diffs)):
- (op, text) = diffs[x]
- if op != self.DIFF_INSERT: # Equality or deletion.
- chars1 += len(text)
- if op != self.DIFF_DELETE: # Equality or insertion.
- chars2 += len(text)
- if chars1 > loc: # Overshot the location.
- break
- last_chars1 = chars1
- last_chars2 = chars2
-
- if len(diffs) != x and diffs[x][0] == self.DIFF_DELETE:
- # The location was deleted.
- return last_chars2
- # Add the remaining len(character).
- return last_chars2 + (loc - last_chars1)
+ chars1 = 0
+ chars2 = 0
+ last_chars1 = 0
+ last_chars2 = 0
+ for x in range(len(diffs)):
+ (op, text) = diffs[x]
+ if op != self.DIFF_INSERT: # Equality or deletion.
+ chars1 += len(text)
+ if op != self.DIFF_DELETE: # Equality or insertion.
+ chars2 += len(text)
+ if chars1 > loc: # Overshot the location.
+ break
+ last_chars1 = chars1
+ last_chars2 = chars2
+
+ if len(diffs) != x and diffs[x][0] == self.DIFF_DELETE:
+ # The location was deleted.
+ return last_chars2
+ # Add the remaining len(character).
+ return last_chars2 + (loc - last_chars1)
- def diff_prettyHtml(self, diffs):
- """Convert a diff array into a pretty HTML report.
+ def diff_prettyHtml(self, diffs):
+ """Convert a diff array into a pretty HTML report.
Args:
diffs: Array of diff tuples.
@@ -1065,20 +1134,24 @@ class diff_match_patch:
Returns:
HTML representation.
"""
- html = []
- for (op, data) in diffs:
- text = (data.replace("&", "&amp;").replace("<", "&lt;")
- .replace(">", "&gt;").replace("\n", "&para;<br>"))
- if op == self.DIFF_INSERT:
- html.append("<ins style=\"background:#e6ffe6;\">%s</ins>" % text)
- elif op == self.DIFF_DELETE:
- html.append("<del style=\"background:#ffe6e6;\">%s</del>" % text)
- elif op == self.DIFF_EQUAL:
- html.append("<span>%s</span>" % text)
- return "".join(html)
+ html = []
+ for (op, data) in diffs:
+ text = (
+ data.replace("&", "&amp;")
+ .replace("<", "&lt;")
+ .replace(">", "&gt;")
+ .replace("\n", "&para;<br>")
+ )
+ if op == self.DIFF_INSERT:
+ html.append('<ins style="background:#e6ffe6;">%s</ins>' % text)
+ elif op == self.DIFF_DELETE:
+ html.append('<del style="background:#ffe6e6;">%s</del>' % text)
+ elif op == self.DIFF_EQUAL:
+ html.append("<span>%s</span>" % text)
+ return "".join(html)
- def diff_text1(self, diffs):
- """Compute and return the source text (all equalities and deletions).
+ def diff_text1(self, diffs):
+ """Compute and return the source text (all equalities and deletions).
Args:
diffs: Array of diff tuples.
@@ -1086,14 +1159,14 @@ class diff_match_patch:
Returns:
Source text.
"""
- text = []
- for (op, data) in diffs:
- if op != self.DIFF_INSERT:
- text.append(data)
- return "".join(text)
+ text = []
+ for (op, data) in diffs:
+ if op != self.DIFF_INSERT:
+ text.append(data)
+ return "".join(text)
- def diff_text2(self, diffs):
- """Compute and return the destination text (all equalities and insertions).
+ def diff_text2(self, diffs):
+ """Compute and return the destination text (all equalities and insertions).
Args:
diffs: Array of diff tuples.
@@ -1101,14 +1174,14 @@ class diff_match_patch:
Returns:
Destination text.
"""
- text = []
- for (op, data) in diffs:
- if op != self.DIFF_DELETE:
- text.append(data)
- return "".join(text)
+ text = []
+ for (op, data) in diffs:
+ if op != self.DIFF_DELETE:
+ text.append(data)
+ return "".join(text)
- def diff_levenshtein(self, diffs):
- """Compute the Levenshtein distance; the number of inserted, deleted or
+ def diff_levenshtein(self, diffs):
+ """Compute the Levenshtein distance; the number of inserted, deleted or
substituted characters.
Args:
@@ -1117,24 +1190,24 @@ class diff_match_patch:
Returns:
Number of changes.
"""
- levenshtein = 0
- insertions = 0
- deletions = 0
- for (op, data) in diffs:
- if op == self.DIFF_INSERT:
- insertions += len(data)
- elif op == self.DIFF_DELETE:
- deletions += len(data)
- elif op == self.DIFF_EQUAL:
- # A deletion and an insertion is one substitution.
- levenshtein += max(insertions, deletions)
+ levenshtein = 0
insertions = 0
deletions = 0
- levenshtein += max(insertions, deletions)
- return levenshtein
+ for (op, data) in diffs:
+ if op == self.DIFF_INSERT:
+ insertions += len(data)
+ elif op == self.DIFF_DELETE:
+ deletions += len(data)
+ elif op == self.DIFF_EQUAL:
+ # A deletion and an insertion is one substitution.
+ levenshtein += max(insertions, deletions)
+ insertions = 0
+ deletions = 0
+ levenshtein += max(insertions, deletions)
+ return levenshtein
- def diff_toDelta(self, diffs):
- """Crush the diff into an encoded string which describes the operations
+ def diff_toDelta(self, diffs):
+ """Crush the diff into an encoded string which describes the operations
required to transform text1 into text2.
E.g. =3\t-2\t+ing -> Keep 3 chars, delete 2 chars, insert 'ing'.
Operations are tab-separated. Inserted text is escaped using %xx notation.
@@ -1145,20 +1218,20 @@ class diff_match_patch:
Returns:
Delta text.
"""
- text = []
- for (op, data) in diffs:
- if op == self.DIFF_INSERT:
- # High ascii will raise UnicodeDecodeError. Use Unicode instead.
- data = data.encode("utf-8")
- text.append("+" + urllib.parse.quote(data, "!~*'();/?:@&=+$,# "))
- elif op == self.DIFF_DELETE:
- text.append("-%d" % len(data))
- elif op == self.DIFF_EQUAL:
- text.append("=%d" % len(data))
- return "\t".join(text)
+ text = []
+ for (op, data) in diffs:
+ if op == self.DIFF_INSERT:
+ # High ascii will raise UnicodeDecodeError. Use Unicode instead.
+ data = data.encode("utf-8")
+ text.append("+" + urllib.parse.quote(data, "!~*'();/?:@&=+$,# "))
+ elif op == self.DIFF_DELETE:
+ text.append("-%d" % len(data))
+ elif op == self.DIFF_EQUAL:
+ text.append("=%d" % len(data))
+ return "\t".join(text)
- def diff_fromDelta(self, text1, delta):
- """Given the original text1, and an encoded string which describes the
+ def diff_fromDelta(self, text1, delta):
+ """Given the original text1, and an encoded string which describes the
operations required to transform text1 into text2, compute the full diff.
Args:
@@ -1171,46 +1244,48 @@ class diff_match_patch:
Raises:
ValueError: If invalid input.
"""
- diffs = []
- pointer = 0 # Cursor in text1
- tokens = delta.split("\t")
- for token in tokens:
- if token == "":
- # Blank tokens are ok (from a trailing \t).
- continue
- # Each token begins with a one character parameter which specifies the
- # operation of this token (delete, insert, equality).
- param = token[1:]
- if token[0] == "+":
- param = urllib.parse.unquote(param)
- diffs.append((self.DIFF_INSERT, param))
- elif token[0] == "-" or token[0] == "=":
- try:
- n = int(param)
- except ValueError:
- raise ValueError("Invalid number in diff_fromDelta: " + param)
- if n < 0:
- raise ValueError("Negative number in diff_fromDelta: " + param)
- text = text1[pointer : pointer + n]
- pointer += n
- if token[0] == "=":
- diffs.append((self.DIFF_EQUAL, text))
- else:
- diffs.append((self.DIFF_DELETE, text))
- else:
- # Anything else is an error.
- raise ValueError("Invalid diff operation in diff_fromDelta: " +
- token[0])
- if pointer != len(text1):
- raise ValueError(
- "Delta length (%d) does not equal source text length (%d)." %
- (pointer, len(text1)))
- return diffs
+ diffs = []
+ pointer = 0 # Cursor in text1
+ tokens = delta.split("\t")
+ for token in tokens:
+ if token == "":
+ # Blank tokens are ok (from a trailing \t).
+ continue
+ # Each token begins with a one character parameter which specifies the
+ # operation of this token (delete, insert, equality).
+ param = token[1:]
+ if token[0] == "+":
+ param = urllib.parse.unquote(param)
+ diffs.append((self.DIFF_INSERT, param))
+ elif token[0] == "-" or token[0] == "=":
+ try:
+ n = int(param)
+ except ValueError:
+ raise ValueError("Invalid number in diff_fromDelta: " + param)
+ if n < 0:
+ raise ValueError("Negative number in diff_fromDelta: " + param)
+ text = text1[pointer : pointer + n]
+ pointer += n
+ if token[0] == "=":
+ diffs.append((self.DIFF_EQUAL, text))
+ else:
+ diffs.append((self.DIFF_DELETE, text))
+ else:
+ # Anything else is an error.
+ raise ValueError(
+ "Invalid diff operation in diff_fromDelta: " + token[0]
+ )
+ if pointer != len(text1):
+ raise ValueError(
+ "Delta length (%d) does not equal source text length (%d)."
+ % (pointer, len(text1))
+ )
+ return diffs
- # MATCH FUNCTIONS
+ # MATCH FUNCTIONS
- def match_main(self, text, pattern, loc):
- """Locate the best instance of 'pattern' in 'text' near 'loc'.
+ def match_main(self, text, pattern, loc):
+ """Locate the best instance of 'pattern' in 'text' near 'loc'.
Args:
text: The text to search.
@@ -1220,27 +1295,27 @@ class diff_match_patch:
Returns:
Best match index or -1.
"""
- # Check for null inputs.
- if text == None or pattern == None:
- raise ValueError("Null inputs. (match_main)")
-
- loc = max(0, min(loc, len(text)))
- if text == pattern:
- # Shortcut (potentially not guaranteed by the algorithm)
- return 0
- elif not text:
- # Nothing to match.
- return -1
- elif text[loc:loc + len(pattern)] == pattern:
- # Perfect match at the perfect spot! (Includes case of null pattern)
- return loc
- else:
- # Do a fuzzy compare.
- match = self.match_bitap(text, pattern, loc)
- return match
+ # Check for null inputs.
+ if text == None or pattern == None:
+ raise ValueError("Null inputs. (match_main)")
+
+ loc = max(0, min(loc, len(text)))
+ if text == pattern:
+ # Shortcut (potentially not guaranteed by the algorithm)
+ return 0
+ elif not text:
+ # Nothing to match.
+ return -1
+ elif text[loc : loc + len(pattern)] == pattern:
+ # Perfect match at the perfect spot! (Includes case of null pattern)
+ return loc
+ else:
+ # Do a fuzzy compare.
+ match = self.match_bitap(text, pattern, loc)
+ return match
- def match_bitap(self, text, pattern, loc):
- """Locate the best instance of 'pattern' in 'text' near 'loc' using the
+ def match_bitap(self, text, pattern, loc):
+ """Locate the best instance of 'pattern' in 'text' near 'loc' using the
Bitap algorithm.
Args:
@@ -1251,15 +1326,15 @@ class diff_match_patch:
Returns:
Best match index or -1.
"""
- # Python doesn't have a maxint limit, so ignore this check.
- #if self.Match_MaxBits != 0 and len(pattern) > self.Match_MaxBits:
- # raise ValueError("Pattern too long for this application.")
+ # Python doesn't have a maxint limit, so ignore this check.
+ # if self.Match_MaxBits != 0 and len(pattern) > self.Match_MaxBits:
+ # raise ValueError("Pattern too long for this application.")
- # Initialise the alphabet.
- s = self.match_alphabet(pattern)
+ # Initialise the alphabet.
+ s = self.match_alphabet(pattern)
- def match_bitapScore(e, x):
- """Compute and return the score for a match with e errors and x location.
+ def match_bitapScore(e, x):
+ """Compute and return the score for a match with e errors and x location.
Accesses loc and pattern through being a closure.
Args:
@@ -1269,84 +1344,87 @@ class diff_match_patch:
Returns:
Overall score for match (0.0 = good, 1.0 = bad).
"""
- accuracy = float(e) / len(pattern)
- proximity = abs(loc - x)
- if not self.Match_Distance:
- # Dodge divide by zero error.
- return proximity and 1.0 or accuracy
- return accuracy + (proximity / float(self.Match_Distance))
-
- # Highest score beyond which we give up.
- score_threshold = self.Match_Threshold
- # Is there a nearby exact match? (speedup)
- best_loc = text.find(pattern, loc)
- if best_loc != -1:
- score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
- # What about in the other direction? (speedup)
- best_loc = text.rfind(pattern, loc + len(pattern))
- if best_loc != -1:
- score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
-
- # Initialise the bit arrays.
- matchmask = 1 << (len(pattern) - 1)
- best_loc = -1
-
- bin_max = len(pattern) + len(text)
- # Empty initialization added to appease pychecker.
- last_rd = None
- for d in range(len(pattern)):
- # Scan for the best match each iteration allows for one more error.
- # Run a binary search to determine how far from 'loc' we can stray at
- # this error level.
- bin_min = 0
- bin_mid = bin_max
- while bin_min < bin_mid:
- if match_bitapScore(d, loc + bin_mid) <= score_threshold:
- bin_min = bin_mid
- else:
- bin_max = bin_mid
- bin_mid = (bin_max - bin_min) // 2 + bin_min
-
- # Use the result from this iteration as the maximum for the next.
- bin_max = bin_mid
- start = max(1, loc - bin_mid + 1)
- finish = min(loc + bin_mid, len(text)) + len(pattern)
-
- rd = [0] * (finish + 2)
- rd[finish + 1] = (1 << d) - 1
- for j in range(finish, start - 1, -1):
- if len(text) <= j - 1:
- # Out of range.
- charMatch = 0
- else:
- charMatch = s.get(text[j - 1], 0)
- if d == 0: # First pass: exact match.
- rd[j] = ((rd[j + 1] << 1) | 1) & charMatch
- else: # Subsequent passes: fuzzy match.
- rd[j] = (((rd[j + 1] << 1) | 1) & charMatch) | (
- ((last_rd[j + 1] | last_rd[j]) << 1) | 1) | last_rd[j + 1]
- if rd[j] & matchmask:
- score = match_bitapScore(d, j - 1)
- # This match will almost certainly be better than any existing match.
- # But check anyway.
- if score <= score_threshold:
- # Told you so.
- score_threshold = score
- best_loc = j - 1
- if best_loc > loc:
- # When passing loc, don't exceed our current distance from loc.
- start = max(1, 2 * loc - best_loc)
- else:
- # Already passed loc, downhill from here on in.
- break
- # No hope for a (better) match at greater error levels.
- if match_bitapScore(d + 1, loc) > score_threshold:
- break
- last_rd = rd
- return best_loc
+ accuracy = float(e) / len(pattern)
+ proximity = abs(loc - x)
+ if not self.Match_Distance:
+ # Dodge divide by zero error.
+ return proximity and 1.0 or accuracy
+ return accuracy + (proximity / float(self.Match_Distance))
+
+ # Highest score beyond which we give up.
+ score_threshold = self.Match_Threshold
+ # Is there a nearby exact match? (speedup)
+ best_loc = text.find(pattern, loc)
+ if best_loc != -1:
+ score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
+ # What about in the other direction? (speedup)
+ best_loc = text.rfind(pattern, loc + len(pattern))
+ if best_loc != -1:
+ score_threshold = min(match_bitapScore(0, best_loc), score_threshold)
+
+ # Initialise the bit arrays.
+ matchmask = 1 << (len(pattern) - 1)
+ best_loc = -1
+
+ bin_max = len(pattern) + len(text)
+ # Empty initialization added to appease pychecker.
+ last_rd = None
+ for d in range(len(pattern)):
+ # Scan for the best match each iteration allows for one more error.
+ # Run a binary search to determine how far from 'loc' we can stray at
+ # this error level.
+ bin_min = 0
+ bin_mid = bin_max
+ while bin_min < bin_mid:
+ if match_bitapScore(d, loc + bin_mid) <= score_threshold:
+ bin_min = bin_mid
+ else:
+ bin_max = bin_mid
+ bin_mid = (bin_max - bin_min) // 2 + bin_min
+
+ # Use the result from this iteration as the maximum for the next.
+ bin_max = bin_mid
+ start = max(1, loc - bin_mid + 1)
+ finish = min(loc + bin_mid, len(text)) + len(pattern)
+
+ rd = [0] * (finish + 2)
+ rd[finish + 1] = (1 << d) - 1
+ for j in range(finish, start - 1, -1):
+ if len(text) <= j - 1:
+ # Out of range.
+ charMatch = 0
+ else:
+ charMatch = s.get(text[j - 1], 0)
+ if d == 0: # First pass: exact match.
+ rd[j] = ((rd[j + 1] << 1) | 1) & charMatch
+ else: # Subsequent passes: fuzzy match.
+ rd[j] = (
+ (((rd[j + 1] << 1) | 1) & charMatch)
+ | (((last_rd[j + 1] | last_rd[j]) << 1) | 1)
+ | last_rd[j + 1]
+ )
+ if rd[j] & matchmask:
+ score = match_bitapScore(d, j - 1)
+ # This match will almost certainly be better than any existing match.
+ # But check anyway.
+ if score <= score_threshold:
+ # Told you so.
+ score_threshold = score
+ best_loc = j - 1
+ if best_loc > loc:
+ # When passing loc, don't exceed our current distance from loc.
+ start = max(1, 2 * loc - best_loc)
+ else:
+ # Already passed loc, downhill from here on in.
+ break
+ # No hope for a (better) match at greater error levels.
+ if match_bitapScore(d + 1, loc) > score_threshold:
+ break
+ last_rd = rd
+ return best_loc
- def match_alphabet(self, pattern):
- """Initialise the alphabet for the Bitap algorithm.
+ def match_alphabet(self, pattern):
+ """Initialise the alphabet for the Bitap algorithm.
Args:
pattern: The text to encode.
@@ -1354,58 +1432,61 @@ class diff_match_patch:
Returns:
Hash of character locations.
"""
- s = {}
- for char in pattern:
- s[char] = 0
- for i in range(len(pattern)):
- s[pattern[i]] |= 1 << (len(pattern) - i - 1)
- return s
+ s = {}
+ for char in pattern:
+ s[char] = 0
+ for i in range(len(pattern)):
+ s[pattern[i]] |= 1 << (len(pattern) - i - 1)
+ return s
- # PATCH FUNCTIONS
+ # PATCH FUNCTIONS
- def patch_addContext(self, patch, text):
- """Increase the context until it is unique,
+ def patch_addContext(self, patch, text):
+ """Increase the context until it is unique,
but don't let the pattern expand beyond Match_MaxBits.
Args:
patch: The patch to grow.
text: Source text.
"""
- if len(text) == 0:
- return
- pattern = text[patch.start2 : patch.start2 + patch.length1]
- padding = 0
-
- # Look for the first and last matches of pattern in text. If two different
- # matches are found, increase the pattern length.
- while (text.find(pattern) != text.rfind(pattern) and (self.Match_MaxBits ==
- 0 or len(pattern) < self.Match_MaxBits - self.Patch_Margin -
- self.Patch_Margin)):
- padding += self.Patch_Margin
- pattern = text[max(0, patch.start2 - padding) :
- patch.start2 + patch.length1 + padding]
- # Add one chunk for good luck.
- padding += self.Patch_Margin
-
- # Add the prefix.
- prefix = text[max(0, patch.start2 - padding) : patch.start2]
- if prefix:
- patch.diffs[:0] = [(self.DIFF_EQUAL, prefix)]
- # Add the suffix.
- suffix = text[patch.start2 + patch.length1 :
- patch.start2 + patch.length1 + padding]
- if suffix:
- patch.diffs.append((self.DIFF_EQUAL, suffix))
-
- # Roll back the start points.
- patch.start1 -= len(prefix)
- patch.start2 -= len(prefix)
- # Extend lengths.
- patch.length1 += len(prefix) + len(suffix)
- patch.length2 += len(prefix) + len(suffix)
+ if len(text) == 0:
+ return
+ pattern = text[patch.start2 : patch.start2 + patch.length1]
+ padding = 0
+
+ # Look for the first and last matches of pattern in text. If two different
+ # matches are found, increase the pattern length.
+ while text.find(pattern) != text.rfind(pattern) and (
+ self.Match_MaxBits == 0
+ or len(pattern) < self.Match_MaxBits - self.Patch_Margin - self.Patch_Margin
+ ):
+ padding += self.Patch_Margin
+ pattern = text[
+ max(0, patch.start2 - padding) : patch.start2 + patch.length1 + padding
+ ]
+ # Add one chunk for good luck.
+ padding += self.Patch_Margin
+
+ # Add the prefix.
+ prefix = text[max(0, patch.start2 - padding) : patch.start2]
+ if prefix:
+ patch.diffs[:0] = [(self.DIFF_EQUAL, prefix)]
+ # Add the suffix.
+ suffix = text[
+ patch.start2 + patch.length1 : patch.start2 + patch.length1 + padding
+ ]
+ if suffix:
+ patch.diffs.append((self.DIFF_EQUAL, suffix))
+
+ # Roll back the start points.
+ patch.start1 -= len(prefix)
+ patch.start2 -= len(prefix)
+ # Extend lengths.
+ patch.length1 += len(prefix) + len(suffix)
+ patch.length2 += len(prefix) + len(suffix)
- def patch_make(self, a, b=None, c=None):
- """Compute a list of patches to turn text1 into text2.
+ def patch_make(self, a, b=None, c=None):
+ """Compute a list of patches to turn text1 into text2.
Use diffs if provided, otherwise compute it ourselves.
There are four ways to call this function, depending on what data is
available to the caller:
@@ -1429,96 +1510,102 @@ class diff_match_patch:
Returns:
Array of Patch objects.
"""
- text1 = None
- diffs = None
- if isinstance(a, str) and isinstance(b, str) and c is None:
- # Method 1: text1, text2
- # Compute diffs from text1 and text2.
- text1 = a
- diffs = self.diff_main(text1, b, True)
- if len(diffs) > 2:
- self.diff_cleanupSemantic(diffs)
- self.diff_cleanupEfficiency(diffs)
- elif isinstance(a, list) and b is None and c is None:
- # Method 2: diffs
- # Compute text1 from diffs.
- diffs = a
- text1 = self.diff_text1(diffs)
- elif isinstance(a, str) and isinstance(b, list) and c is None:
- # Method 3: text1, diffs
- text1 = a
- diffs = b
- elif (isinstance(a, str) and isinstance(b, str) and
- isinstance(c, list)):
- # Method 4: text1, text2, diffs
- # text2 is not used.
- text1 = a
- diffs = c
- else:
- raise ValueError("Unknown call format to patch_make.")
-
- if not diffs:
- return [] # Get rid of the None case.
- patches = []
- patch = patch_obj()
- char_count1 = 0 # Number of characters into the text1 string.
- char_count2 = 0 # Number of characters into the text2 string.
- prepatch_text = text1 # Recreate the patches to determine context info.
- postpatch_text = text1
- for x in range(len(diffs)):
- (diff_type, diff_text) = diffs[x]
- if len(patch.diffs) == 0 and diff_type != self.DIFF_EQUAL:
- # A new patch starts here.
- patch.start1 = char_count1
- patch.start2 = char_count2
- if diff_type == self.DIFF_INSERT:
- # Insertion
- patch.diffs.append(diffs[x])
- patch.length2 += len(diff_text)
- postpatch_text = (postpatch_text[:char_count2] + diff_text +
- postpatch_text[char_count2:])
- elif diff_type == self.DIFF_DELETE:
- # Deletion.
- patch.length1 += len(diff_text)
- patch.diffs.append(diffs[x])
- postpatch_text = (postpatch_text[:char_count2] +
- postpatch_text[char_count2 + len(diff_text):])
- elif (diff_type == self.DIFF_EQUAL and
- len(diff_text) <= 2 * self.Patch_Margin and
- len(patch.diffs) != 0 and len(diffs) != x + 1):
- # Small equality inside a patch.
- patch.diffs.append(diffs[x])
- patch.length1 += len(diff_text)
- patch.length2 += len(diff_text)
-
- if (diff_type == self.DIFF_EQUAL and
- len(diff_text) >= 2 * self.Patch_Margin):
- # Time for a new patch.
+ text1 = None
+ diffs = None
+ if isinstance(a, str) and isinstance(b, str) and c is None:
+ # Method 1: text1, text2
+ # Compute diffs from text1 and text2.
+ text1 = a
+ diffs = self.diff_main(text1, b, True)
+ if len(diffs) > 2:
+ self.diff_cleanupSemantic(diffs)
+ self.diff_cleanupEfficiency(diffs)
+ elif isinstance(a, list) and b is None and c is None:
+ # Method 2: diffs
+ # Compute text1 from diffs.
+ diffs = a
+ text1 = self.diff_text1(diffs)
+ elif isinstance(a, str) and isinstance(b, list) and c is None:
+ # Method 3: text1, diffs
+ text1 = a
+ diffs = b
+ elif isinstance(a, str) and isinstance(b, str) and isinstance(c, list):
+ # Method 4: text1, text2, diffs
+ # text2 is not used.
+ text1 = a
+ diffs = c
+ else:
+ raise ValueError("Unknown call format to patch_make.")
+
+ if not diffs:
+ return [] # Get rid of the None case.
+ patches = []
+ patch = patch_obj()
+ char_count1 = 0 # Number of characters into the text1 string.
+ char_count2 = 0 # Number of characters into the text2 string.
+ prepatch_text = text1 # Recreate the patches to determine context info.
+ postpatch_text = text1
+ for x in range(len(diffs)):
+ (diff_type, diff_text) = diffs[x]
+ if len(patch.diffs) == 0 and diff_type != self.DIFF_EQUAL:
+ # A new patch starts here.
+ patch.start1 = char_count1
+ patch.start2 = char_count2
+ if diff_type == self.DIFF_INSERT:
+ # Insertion
+ patch.diffs.append(diffs[x])
+ patch.length2 += len(diff_text)
+ postpatch_text = (
+ postpatch_text[:char_count2]
+ + diff_text
+ + postpatch_text[char_count2:]
+ )
+ elif diff_type == self.DIFF_DELETE:
+ # Deletion.
+ patch.length1 += len(diff_text)
+ patch.diffs.append(diffs[x])
+ postpatch_text = (
+ postpatch_text[:char_count2]
+ + postpatch_text[char_count2 + len(diff_text) :]
+ )
+ elif (
+ diff_type == self.DIFF_EQUAL
+ and len(diff_text) <= 2 * self.Patch_Margin
+ and len(patch.diffs) != 0
+ and len(diffs) != x + 1
+ ):
+ # Small equality inside a patch.
+ patch.diffs.append(diffs[x])
+ patch.length1 += len(diff_text)
+ patch.length2 += len(diff_text)
+
+ if diff_type == self.DIFF_EQUAL and len(diff_text) >= 2 * self.Patch_Margin:
+ # Time for a new patch.
+ if len(patch.diffs) != 0:
+ self.patch_addContext(patch, prepatch_text)
+ patches.append(patch)
+ patch = patch_obj()
+ # Unlike Unidiff, our patch lists have a rolling context.
+ # https://github.com/google/diff-match-patch/wiki/Unidiff
+ # Update prepatch text & pos to reflect the application of the
+ # just completed patch.
+ prepatch_text = postpatch_text
+ char_count1 = char_count2
+
+ # Update the current character count.
+ if diff_type != self.DIFF_INSERT:
+ char_count1 += len(diff_text)
+ if diff_type != self.DIFF_DELETE:
+ char_count2 += len(diff_text)
+
+ # Pick up the leftover patch if not empty.
if len(patch.diffs) != 0:
- self.patch_addContext(patch, prepatch_text)
- patches.append(patch)
- patch = patch_obj()
- # Unlike Unidiff, our patch lists have a rolling context.
- # https://github.com/google/diff-match-patch/wiki/Unidiff
- # Update prepatch text & pos to reflect the application of the
- # just completed patch.
- prepatch_text = postpatch_text
- char_count1 = char_count2
-
- # Update the current character count.
- if diff_type != self.DIFF_INSERT:
- char_count1 += len(diff_text)
- if diff_type != self.DIFF_DELETE:
- char_count2 += len(diff_text)
-
- # Pick up the leftover patch if not empty.
- if len(patch.diffs) != 0:
- self.patch_addContext(patch, prepatch_text)
- patches.append(patch)
- return patches
+ self.patch_addContext(patch, prepatch_text)
+ patches.append(patch)
+ return patches
- def patch_deepCopy(self, patches):
- """Given an array of patches, return another array that is identical.
+ def patch_deepCopy(self, patches):
+ """Given an array of patches, return another array that is identical.
Args:
patches: Array of Patch objects.
@@ -1526,20 +1613,20 @@ class diff_match_patch:
Returns:
Array of Patch objects.
"""
- patchesCopy = []
- for patch in patches:
- patchCopy = patch_obj()
- # No need to deep copy the tuples since they are immutable.
- patchCopy.diffs = patch.diffs[:]
- patchCopy.start1 = patch.start1
- patchCopy.start2 = patch.start2
- patchCopy.length1 = patch.length1
- patchCopy.length2 = patch.length2
- patchesCopy.append(patchCopy)
- return patchesCopy
+ patchesCopy = []
+ for patch in patches:
+ patchCopy = patch_obj()
+ # No need to deep copy the tuples since they are immutable.
+ patchCopy.diffs = patch.diffs[:]
+ patchCopy.start1 = patch.start1
+ patchCopy.start2 = patch.start2
+ patchCopy.length1 = patch.length1
+ patchCopy.length2 = patch.length2
+ patchesCopy.append(patchCopy)
+ return patchesCopy
- def patch_apply(self, patches, text):
- """Merge a set of patches onto the text. Return a patched text, as well
+ def patch_apply(self, patches, text):
+ """Merge a set of patches onto the text. Return a patched text, as well
as a list of true/false values indicating which patches were applied.
Args:
@@ -1549,85 +1636,102 @@ class diff_match_patch:
Returns:
Two element Array, containing the new text and an array of boolean values.
"""
- if not patches:
- return (text, [])
+ if not patches:
+ return (text, [])
- # Deep copy the patches so that no changes are made to originals.
- patches = self.patch_deepCopy(patches)
+ # Deep copy the patches so that no changes are made to originals.
+ patches = self.patch_deepCopy(patches)
- nullPadding = self.patch_addPadding(patches)
- text = nullPadding + text + nullPadding
- self.patch_splitMax(patches)
-
- # delta keeps track of the offset between the expected and actual location
- # of the previous patch. If there are patches expected at positions 10 and
- # 20, but the first patch was found at 12, delta is 2 and the second patch
- # has an effective expected position of 22.
- delta = 0
- results = []
- for patch in patches:
- expected_loc = patch.start2 + delta
- text1 = self.diff_text1(patch.diffs)
- end_loc = -1
- if len(text1) > self.Match_MaxBits:
- # patch_splitMax will only provide an oversized pattern in the case of
- # a monster delete.
- start_loc = self.match_main(text, text1[:self.Match_MaxBits],
- expected_loc)
- if start_loc != -1:
- end_loc = self.match_main(text, text1[-self.Match_MaxBits:],
- expected_loc + len(text1) - self.Match_MaxBits)
- if end_loc == -1 or start_loc >= end_loc:
- # Can't find valid trailing context. Drop this patch.
- start_loc = -1
- else:
- start_loc = self.match_main(text, text1, expected_loc)
- if start_loc == -1:
- # No match found. :(
- results.append(False)
- # Subtract the delta for this failed patch from subsequent patches.
- delta -= patch.length2 - patch.length1
- else:
- # Found a match. :)
- results.append(True)
- delta = start_loc - expected_loc
- if end_loc == -1:
- text2 = text[start_loc : start_loc + len(text1)]
- else:
- text2 = text[start_loc : end_loc + self.Match_MaxBits]
- if text1 == text2:
- # Perfect match, just shove the replacement text in.
- text = (text[:start_loc] + self.diff_text2(patch.diffs) +
- text[start_loc + len(text1):])
- else:
- # Imperfect match.
- # Run a diff to get a framework of equivalent indices.
- diffs = self.diff_main(text1, text2, False)
- if (len(text1) > self.Match_MaxBits and
- self.diff_levenshtein(diffs) / float(len(text1)) >
- self.Patch_DeleteThreshold):
- # The end points match, but the content is unacceptably bad.
- results[-1] = False
- else:
- self.diff_cleanupSemanticLossless(diffs)
- index1 = 0
- for (op, data) in patch.diffs:
- if op != self.DIFF_EQUAL:
- index2 = self.diff_xIndex(diffs, index1)
- if op == self.DIFF_INSERT: # Insertion
- text = text[:start_loc + index2] + data + text[start_loc +
- index2:]
- elif op == self.DIFF_DELETE: # Deletion
- text = text[:start_loc + index2] + text[start_loc +
- self.diff_xIndex(diffs, index1 + len(data)):]
- if op != self.DIFF_DELETE:
- index1 += len(data)
- # Strip the padding off.
- text = text[len(nullPadding):-len(nullPadding)]
- return (text, results)
+ nullPadding = self.patch_addPadding(patches)
+ text = nullPadding + text + nullPadding
+ self.patch_splitMax(patches)
+
+ # delta keeps track of the offset between the expected and actual location
+ # of the previous patch. If there are patches expected at positions 10 and
+ # 20, but the first patch was found at 12, delta is 2 and the second patch
+ # has an effective expected position of 22.
+ delta = 0
+ results = []
+ for patch in patches:
+ expected_loc = patch.start2 + delta
+ text1 = self.diff_text1(patch.diffs)
+ end_loc = -1
+ if len(text1) > self.Match_MaxBits:
+ # patch_splitMax will only provide an oversized pattern in the case of
+ # a monster delete.
+ start_loc = self.match_main(
+ text, text1[: self.Match_MaxBits], expected_loc
+ )
+ if start_loc != -1:
+ end_loc = self.match_main(
+ text,
+ text1[-self.Match_MaxBits :],
+ expected_loc + len(text1) - self.Match_MaxBits,
+ )
+ if end_loc == -1 or start_loc >= end_loc:
+ # Can't find valid trailing context. Drop this patch.
+ start_loc = -1
+ else:
+ start_loc = self.match_main(text, text1, expected_loc)
+ if start_loc == -1:
+ # No match found. :(
+ results.append(False)
+ # Subtract the delta for this failed patch from subsequent patches.
+ delta -= patch.length2 - patch.length1
+ else:
+ # Found a match. :)
+ results.append(True)
+ delta = start_loc - expected_loc
+ if end_loc == -1:
+ text2 = text[start_loc : start_loc + len(text1)]
+ else:
+ text2 = text[start_loc : end_loc + self.Match_MaxBits]
+ if text1 == text2:
+ # Perfect match, just shove the replacement text in.
+ text = (
+ text[:start_loc]
+ + self.diff_text2(patch.diffs)
+ + text[start_loc + len(text1) :]
+ )
+ else:
+ # Imperfect match.
+ # Run a diff to get a framework of equivalent indices.
+ diffs = self.diff_main(text1, text2, False)
+ if (
+ len(text1) > self.Match_MaxBits
+ and self.diff_levenshtein(diffs) / float(len(text1))
+ > self.Patch_DeleteThreshold
+ ):
+ # The end points match, but the content is unacceptably bad.
+ results[-1] = False
+ else:
+ self.diff_cleanupSemanticLossless(diffs)
+ index1 = 0
+ for (op, data) in patch.diffs:
+ if op != self.DIFF_EQUAL:
+ index2 = self.diff_xIndex(diffs, index1)
+ if op == self.DIFF_INSERT: # Insertion
+ text = (
+ text[: start_loc + index2]
+ + data
+ + text[start_loc + index2 :]
+ )
+ elif op == self.DIFF_DELETE: # Deletion
+ text = (
+ text[: start_loc + index2]
+ + text[
+ start_loc
+ + self.diff_xIndex(diffs, index1 + len(data)) :
+ ]
+ )
+ if op != self.DIFF_DELETE:
+ index1 += len(data)
+ # Strip the padding off.
+ text = text[len(nullPadding) : -len(nullPadding)]
+ return (text, results)
- def patch_addPadding(self, patches):
- """Add some padding on text start and end so that edges can match
+ def patch_addPadding(self, patches):
+ """Add some padding on text start and end so that edges can match
something. Intended to be called only from within patch_apply.
Args:
@@ -1636,144 +1740,154 @@ class diff_match_patch:
Returns:
The padding string added to each side.
"""
- paddingLength = self.Patch_Margin
- nullPadding = ""
- for x in range(1, paddingLength + 1):
- nullPadding += chr(x)
-
- # Bump all the patches forward.
- for patch in patches:
- patch.start1 += paddingLength
- patch.start2 += paddingLength
-
- # Add some padding on start of first diff.
- patch = patches[0]
- diffs = patch.diffs
- if not diffs or diffs[0][0] != self.DIFF_EQUAL:
- # Add nullPadding equality.
- diffs.insert(0, (self.DIFF_EQUAL, nullPadding))
- patch.start1 -= paddingLength # Should be 0.
- patch.start2 -= paddingLength # Should be 0.
- patch.length1 += paddingLength
- patch.length2 += paddingLength
- elif paddingLength > len(diffs[0][1]):
- # Grow first equality.
- extraLength = paddingLength - len(diffs[0][1])
- newText = nullPadding[len(diffs[0][1]):] + diffs[0][1]
- diffs[0] = (diffs[0][0], newText)
- patch.start1 -= extraLength
- patch.start2 -= extraLength
- patch.length1 += extraLength
- patch.length2 += extraLength
-
- # Add some padding on end of last diff.
- patch = patches[-1]
- diffs = patch.diffs
- if not diffs or diffs[-1][0] != self.DIFF_EQUAL:
- # Add nullPadding equality.
- diffs.append((self.DIFF_EQUAL, nullPadding))
- patch.length1 += paddingLength
- patch.length2 += paddingLength
- elif paddingLength > len(diffs[-1][1]):
- # Grow last equality.
- extraLength = paddingLength - len(diffs[-1][1])
- newText = diffs[-1][1] + nullPadding[:extraLength]
- diffs[-1] = (diffs[-1][0], newText)
- patch.length1 += extraLength
- patch.length2 += extraLength
+ paddingLength = self.Patch_Margin
+ nullPadding = ""
+ for x in range(1, paddingLength + 1):
+ nullPadding += chr(x)
+
+ # Bump all the patches forward.
+ for patch in patches:
+ patch.start1 += paddingLength
+ patch.start2 += paddingLength
+
+ # Add some padding on start of first diff.
+ patch = patches[0]
+ diffs = patch.diffs
+ if not diffs or diffs[0][0] != self.DIFF_EQUAL:
+ # Add nullPadding equality.
+ diffs.insert(0, (self.DIFF_EQUAL, nullPadding))
+ patch.start1 -= paddingLength # Should be 0.
+ patch.start2 -= paddingLength # Should be 0.
+ patch.length1 += paddingLength
+ patch.length2 += paddingLength
+ elif paddingLength > len(diffs[0][1]):
+ # Grow first equality.
+ extraLength = paddingLength - len(diffs[0][1])
+ newText = nullPadding[len(diffs[0][1]) :] + diffs[0][1]
+ diffs[0] = (diffs[0][0], newText)
+ patch.start1 -= extraLength
+ patch.start2 -= extraLength
+ patch.length1 += extraLength
+ patch.length2 += extraLength
+
+ # Add some padding on end of last diff.
+ patch = patches[-1]
+ diffs = patch.diffs
+ if not diffs or diffs[-1][0] != self.DIFF_EQUAL:
+ # Add nullPadding equality.
+ diffs.append((self.DIFF_EQUAL, nullPadding))
+ patch.length1 += paddingLength
+ patch.length2 += paddingLength
+ elif paddingLength > len(diffs[-1][1]):
+ # Grow last equality.
+ extraLength = paddingLength - len(diffs[-1][1])
+ newText = diffs[-1][1] + nullPadding[:extraLength]
+ diffs[-1] = (diffs[-1][0], newText)
+ patch.length1 += extraLength
+ patch.length2 += extraLength
- return nullPadding
+ return nullPadding
- def patch_splitMax(self, patches):
- """Look through the patches and break up any which are longer than the
+ def patch_splitMax(self, patches):
+ """Look through the patches and break up any which are longer than the
maximum limit of the match algorithm.
Intended to be called only from within patch_apply.
Args:
patches: Array of Patch objects.
"""
- patch_size = self.Match_MaxBits
- if patch_size == 0:
- # Python has the option of not splitting strings due to its ability
- # to handle integers of arbitrary precision.
- return
- for x in range(len(patches)):
- if patches[x].length1 <= patch_size:
- continue
- bigpatch = patches[x]
- # Remove the big old patch.
- del patches[x]
- x -= 1
- start1 = bigpatch.start1
- start2 = bigpatch.start2
- precontext = ''
- while len(bigpatch.diffs) != 0:
- # Create one of several smaller patches.
- patch = patch_obj()
- empty = True
- patch.start1 = start1 - len(precontext)
- patch.start2 = start2 - len(precontext)
- if precontext:
- patch.length1 = patch.length2 = len(precontext)
- patch.diffs.append((self.DIFF_EQUAL, precontext))
-
- while (len(bigpatch.diffs) != 0 and
- patch.length1 < patch_size - self.Patch_Margin):
- (diff_type, diff_text) = bigpatch.diffs[0]
- if diff_type == self.DIFF_INSERT:
- # Insertions are harmless.
- patch.length2 += len(diff_text)
- start2 += len(diff_text)
- patch.diffs.append(bigpatch.diffs.pop(0))
- empty = False
- elif (diff_type == self.DIFF_DELETE and len(patch.diffs) == 1 and
- patch.diffs[0][0] == self.DIFF_EQUAL and
- len(diff_text) > 2 * patch_size):
- # This is a large deletion. Let it pass in one chunk.
- patch.length1 += len(diff_text)
- start1 += len(diff_text)
- empty = False
- patch.diffs.append((diff_type, diff_text))
- del bigpatch.diffs[0]
- else:
- # Deletion or equality. Only take as much as we can stomach.
- diff_text = diff_text[:patch_size - patch.length1 -
- self.Patch_Margin]
- patch.length1 += len(diff_text)
- start1 += len(diff_text)
- if diff_type == self.DIFF_EQUAL:
- patch.length2 += len(diff_text)
- start2 += len(diff_text)
- else:
- empty = False
-
- patch.diffs.append((diff_type, diff_text))
- if diff_text == bigpatch.diffs[0][1]:
- del bigpatch.diffs[0]
- else:
- bigpatch.diffs[0] = (bigpatch.diffs[0][0],
- bigpatch.diffs[0][1][len(diff_text):])
+ patch_size = self.Match_MaxBits
+ if patch_size == 0:
+ # Python has the option of not splitting strings due to its ability
+ # to handle integers of arbitrary precision.
+ return
+ for x in range(len(patches)):
+ if patches[x].length1 <= patch_size:
+ continue
+ bigpatch = patches[x]
+ # Remove the big old patch.
+ del patches[x]
+ x -= 1
+ start1 = bigpatch.start1
+ start2 = bigpatch.start2
+ precontext = ""
+ while len(bigpatch.diffs) != 0:
+ # Create one of several smaller patches.
+ patch = patch_obj()
+ empty = True
+ patch.start1 = start1 - len(precontext)
+ patch.start2 = start2 - len(precontext)
+ if precontext:
+ patch.length1 = patch.length2 = len(precontext)
+ patch.diffs.append((self.DIFF_EQUAL, precontext))
+
+ while (
+ len(bigpatch.diffs) != 0
+ and patch.length1 < patch_size - self.Patch_Margin
+ ):
+ (diff_type, diff_text) = bigpatch.diffs[0]
+ if diff_type == self.DIFF_INSERT:
+ # Insertions are harmless.
+ patch.length2 += len(diff_text)
+ start2 += len(diff_text)
+ patch.diffs.append(bigpatch.diffs.pop(0))
+ empty = False
+ elif (
+ diff_type == self.DIFF_DELETE
+ and len(patch.diffs) == 1
+ and patch.diffs[0][0] == self.DIFF_EQUAL
+ and len(diff_text) > 2 * patch_size
+ ):
+ # This is a large deletion. Let it pass in one chunk.
+ patch.length1 += len(diff_text)
+ start1 += len(diff_text)
+ empty = False
+ patch.diffs.append((diff_type, diff_text))
+ del bigpatch.diffs[0]
+ else:
+ # Deletion or equality. Only take as much as we can stomach.
+ diff_text = diff_text[
+ : patch_size - patch.length1 - self.Patch_Margin
+ ]
+ patch.length1 += len(diff_text)
+ start1 += len(diff_text)
+ if diff_type == self.DIFF_EQUAL:
+ patch.length2 += len(diff_text)
+ start2 += len(diff_text)
+ else:
+ empty = False
+
+ patch.diffs.append((diff_type, diff_text))
+ if diff_text == bigpatch.diffs[0][1]:
+ del bigpatch.diffs[0]
+ else:
+ bigpatch.diffs[0] = (
+ bigpatch.diffs[0][0],
+ bigpatch.diffs[0][1][len(diff_text) :],
+ )
+
+ # Compute the head context for the next patch.
+ precontext = self.diff_text2(patch.diffs)
+ precontext = precontext[-self.Patch_Margin :]
+ # Append the end context for this patch.
+ postcontext = self.diff_text1(bigpatch.diffs)[: self.Patch_Margin]
+ if postcontext:
+ patch.length1 += len(postcontext)
+ patch.length2 += len(postcontext)
+ if len(patch.diffs) != 0 and patch.diffs[-1][0] == self.DIFF_EQUAL:
+ patch.diffs[-1] = (
+ self.DIFF_EQUAL,
+ patch.diffs[-1][1] + postcontext,
+ )
+ else:
+ patch.diffs.append((self.DIFF_EQUAL, postcontext))
+
+ if not empty:
+ x += 1
+ patches.insert(x, patch)
- # Compute the head context for the next patch.
- precontext = self.diff_text2(patch.diffs)
- precontext = precontext[-self.Patch_Margin:]
- # Append the end context for this patch.
- postcontext = self.diff_text1(bigpatch.diffs)[:self.Patch_Margin]
- if postcontext:
- patch.length1 += len(postcontext)
- patch.length2 += len(postcontext)
- if len(patch.diffs) != 0 and patch.diffs[-1][0] == self.DIFF_EQUAL:
- patch.diffs[-1] = (self.DIFF_EQUAL, patch.diffs[-1][1] +
- postcontext)
- else:
- patch.diffs.append((self.DIFF_EQUAL, postcontext))
-
- if not empty:
- x += 1
- patches.insert(x, patch)
-
- def patch_toText(self, patches):
- """Take a list of patches and return a textual representation.
+ def patch_toText(self, patches):
+ """Take a list of patches and return a textual representation.
Args:
patches: Array of Patch objects.
@@ -1781,13 +1895,13 @@ class diff_match_patch:
Returns:
Text representation of patches.
"""
- text = []
- for patch in patches:
- text.append(str(patch))
- return "".join(text)
+ text = []
+ for patch in patches:
+ text.append(str(patch))
+ return "".join(text)
- def patch_fromText(self, textline):
- """Parse a textual representation of patches and return a list of patch
+ def patch_fromText(self, textline):
+ """Parse a textual representation of patches and return a list of patch
objects.
Args:
@@ -1799,109 +1913,109 @@ class diff_match_patch:
Raises:
ValueError: If invalid input.
"""
- patches = []
- if not textline:
- return patches
- text = textline.split('\n')
- while len(text) != 0:
- m = re.match("^@@ -(\d+),?(\d*) \+(\d+),?(\d*) @@$", text[0])
- if not m:
- raise ValueError("Invalid patch string: " + text[0])
- patch = patch_obj()
- patches.append(patch)
- patch.start1 = int(m.group(1))
- if m.group(2) == '':
- patch.start1 -= 1
- patch.length1 = 1
- elif m.group(2) == '0':
- patch.length1 = 0
- else:
- patch.start1 -= 1
- patch.length1 = int(m.group(2))
-
- patch.start2 = int(m.group(3))
- if m.group(4) == '':
- patch.start2 -= 1
- patch.length2 = 1
- elif m.group(4) == '0':
- patch.length2 = 0
- else:
- patch.start2 -= 1
- patch.length2 = int(m.group(4))
-
- del text[0]
-
- while len(text) != 0:
- if text[0]:
- sign = text[0][0]
- else:
- sign = ''
- line = urllib.parse.unquote(text[0][1:])
- if sign == '+':
- # Insertion.
- patch.diffs.append((self.DIFF_INSERT, line))
- elif sign == '-':
- # Deletion.
- patch.diffs.append((self.DIFF_DELETE, line))
- elif sign == ' ':
- # Minor equality.
- patch.diffs.append((self.DIFF_EQUAL, line))
- elif sign == '@':
- # Start of next patch.
- break
- elif sign == '':
- # Blank line? Whatever.
- pass
- else:
- # WTF?
- raise ValueError("Invalid patch mode: '%s'\n%s" % (sign, line))
- del text[0]
- return patches
+ patches = []
+ if not textline:
+ return patches
+ text = textline.split("\n")
+ while len(text) != 0:
+ m = re.match(r"^@@ -(\d+),?(\d*) \+(\d+),?(\d*) @@$", text[0])
+ if not m:
+ raise ValueError("Invalid patch string: " + text[0])
+ patch = patch_obj()
+ patches.append(patch)
+ patch.start1 = int(m.group(1))
+ if m.group(2) == "":
+ patch.start1 -= 1
+ patch.length1 = 1
+ elif m.group(2) == "0":
+ patch.length1 = 0
+ else:
+ patch.start1 -= 1
+ patch.length1 = int(m.group(2))
+
+ patch.start2 = int(m.group(3))
+ if m.group(4) == "":
+ patch.start2 -= 1
+ patch.length2 = 1
+ elif m.group(4) == "0":
+ patch.length2 = 0
+ else:
+ patch.start2 -= 1
+ patch.length2 = int(m.group(4))
+
+ del text[0]
+
+ while len(text) != 0:
+ if text[0]:
+ sign = text[0][0]
+ else:
+ sign = ""
+ line = urllib.parse.unquote(text[0][1:])
+ if sign == "+":
+ # Insertion.
+ patch.diffs.append((self.DIFF_INSERT, line))
+ elif sign == "-":
+ # Deletion.
+ patch.diffs.append((self.DIFF_DELETE, line))
+ elif sign == " ":
+ # Minor equality.
+ patch.diffs.append((self.DIFF_EQUAL, line))
+ elif sign == "@":
+ # Start of next patch.
+ break
+ elif sign == "":
+ # Blank line? Whatever.
+ pass
+ else:
+ # WTF?
+ raise ValueError(f"Invalid patch mode: '{sign}'\n{line}")
+ del text[0]
+ return patches
class patch_obj:
- """Class representing one patch operation.
+ """Class representing one patch operation.
"""
- def __init__(self):
- """Initializes with an empty list of diffs.
+ def __init__(self):
+ """Initializes with an empty list of diffs.
"""
- self.diffs = []
- self.start1 = None
- self.start2 = None
- self.length1 = 0
- self.length2 = 0
+ self.diffs = []
+ self.start1 = None
+ self.start2 = None
+ self.length1 = 0
+ self.length2 = 0
- def __str__(self):
- """Emulate GNU diff's format.
+ def __str__(self):
+ """Emulate GNU diff's format.
Header: @@ -382,8 +481,9 @@
Indices are printed as 1-based, not 0-based.
Returns:
The GNU diff string.
"""
- if self.length1 == 0:
- coords1 = str(self.start1) + ",0"
- elif self.length1 == 1:
- coords1 = str(self.start1 + 1)
- else:
- coords1 = str(self.start1 + 1) + "," + str(self.length1)
- if self.length2 == 0:
- coords2 = str(self.start2) + ",0"
- elif self.length2 == 1:
- coords2 = str(self.start2 + 1)
- else:
- coords2 = str(self.start2 + 1) + "," + str(self.length2)
- text = ["@@ -", coords1, " +", coords2, " @@\n"]
- # Escape the body of the patch with %xx notation.
- for (op, data) in self.diffs:
- if op == diff_match_patch.DIFF_INSERT:
- text.append("+")
- elif op == diff_match_patch.DIFF_DELETE:
- text.append("-")
- elif op == diff_match_patch.DIFF_EQUAL:
- text.append(" ")
- # High ascii will raise UnicodeDecodeError. Use Unicode instead.
- data = data.encode("utf-8")
- text.append(urllib.parse.quote(data, "!~*'();/?:@&=+$,# ") + "\n")
- return "".join(text)
+ if self.length1 == 0:
+ coords1 = str(self.start1) + ",0"
+ elif self.length1 == 1:
+ coords1 = str(self.start1 + 1)
+ else:
+ coords1 = str(self.start1 + 1) + "," + str(self.length1)
+ if self.length2 == 0:
+ coords2 = str(self.start2) + ",0"
+ elif self.length2 == 1:
+ coords2 = str(self.start2 + 1)
+ else:
+ coords2 = str(self.start2 + 1) + "," + str(self.length2)
+ text = ["@@ -", coords1, " +", coords2, " @@\n"]
+ # Escape the body of the patch with %xx notation.
+ for (op, data) in self.diffs:
+ if op == diff_match_patch.DIFF_INSERT:
+ text.append("+")
+ elif op == diff_match_patch.DIFF_DELETE:
+ text.append("-")
+ elif op == diff_match_patch.DIFF_EQUAL:
+ text.append(" ")
+ # High ascii will raise UnicodeDecodeError. Use Unicode instead.
+ data = data.encode("utf-8")
+ text.append(urllib.parse.quote(data, "!~*'();/?:@&=+$,# ") + "\n")
+ return "".join(text)
Index: xmldiff-2.4/xmldiff/actions.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/actions.py
+++ xmldiff-2.4/xmldiff/actions.py
@@ -1,17 +1,17 @@
from collections import namedtuple
# The edit script actions used in xmldiff
-DeleteNode = namedtuple('DeleteNode', 'node')
-InsertNode = namedtuple('InsertNode', 'target tag position')
-RenameNode = namedtuple('RenameNode', 'node tag')
-MoveNode = namedtuple('MoveNode', 'node target position')
+DeleteNode = namedtuple("DeleteNode", "node")
+InsertNode = namedtuple("InsertNode", "target tag position")
+RenameNode = namedtuple("RenameNode", "node tag")
+MoveNode = namedtuple("MoveNode", "node target position")
-UpdateTextIn = namedtuple('UpdateTextIn', 'node text')
-UpdateTextAfter = namedtuple('UpdateTextAfter', 'node text')
+UpdateTextIn = namedtuple("UpdateTextIn", "node text")
+UpdateTextAfter = namedtuple("UpdateTextAfter", "node text")
-UpdateAttrib = namedtuple('UpdateAttrib', 'node name value')
-DeleteAttrib = namedtuple('DeleteAttrib', 'node name')
-InsertAttrib = namedtuple('InsertAttrib', 'node name value')
-RenameAttrib = namedtuple('RenameAttrib', 'node oldname newname')
+UpdateAttrib = namedtuple("UpdateAttrib", "node name value")
+DeleteAttrib = namedtuple("DeleteAttrib", "node name")
+InsertAttrib = namedtuple("InsertAttrib", "node name value")
+RenameAttrib = namedtuple("RenameAttrib", "node oldname newname")
-InsertComment = namedtuple('InsertComment', 'target position text')
+InsertComment = namedtuple("InsertComment", "target position text")
Index: xmldiff-2.4/xmldiff/diff.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/diff.py
+++ xmldiff-2.4/xmldiff/diff.py
@@ -1,15 +1,11 @@
-from __future__ import division
-
from copy import deepcopy
from difflib import SequenceMatcher
from lxml import etree
from xmldiff import utils, actions
-class Differ(object):
-
- def __init__(self, F=None, uniqueattrs=None, ratio_mode='fast',
- fast_match=False):
+class Differ:
+ def __init__(self, F=None, uniqueattrs=None, ratio_mode="fast", fast_match=False):
# The minimum similarity between two nodes to consider them equal
if F is None:
F = 0.5
@@ -18,17 +14,17 @@ class Differ(object):
# that uniquely identifies a node inside a document. Defaults
# to 'xml:id'.
if uniqueattrs is None:
- uniqueattrs = ['{http://www.w3.org/XML/1998/namespace}id']
+ uniqueattrs = ["{http://www.w3.org/XML/1998/namespace}id"]
self.uniqueattrs = uniqueattrs
self.fast_match = fast_match
# Avoid recreating this for every node
self._sequencematcher = SequenceMatcher()
- if ratio_mode == 'fast':
+ if ratio_mode == "fast":
self._sequence_ratio = self._sequencematcher.quick_ratio
- elif ratio_mode == 'accurate':
+ elif ratio_mode == "accurate":
self._sequence_ratio = self._sequencematcher.ratio
- elif ratio_mode == 'faster':
+ elif ratio_mode == "faster":
self._sequence_ratio = self._sequencematcher.real_quick_ratio
else:
raise ValueError("Unknown ratio_mode '%s'" % ratio_mode)
@@ -57,8 +53,9 @@ class Differ(object):
right = right.getroot()
if not (etree.iselement(left) and etree.iselement(right)):
- raise TypeError("The 'left' and 'right' parameters must be "
- "lxml Elements.")
+ raise TypeError(
+ "The 'left' and 'right' parameters must be " "lxml Elements."
+ )
# Left gets modified as a part of the diff, deepcopy it first.
self.left = deepcopy(left)
@@ -111,14 +108,15 @@ class Differ(object):
if self.fast_match:
# First find matches with longest_common_subsequence:
- matches = list(utils.longest_common_subsequence(
- lnodes, rnodes, lambda x, y: self.node_ratio(x, y) >= 0.5))
+ matches = list(
+ utils.longest_common_subsequence(
+ lnodes, rnodes, lambda x, y: self.node_ratio(x, y) >= 0.5
+ )
+ )
# Add the matches (I prefer this from start to finish):
for left_match, right_match in matches:
- self.append_match(lnodes[left_match],
- rnodes[right_match],
- None)
+ self.append_match(lnodes[left_match], rnodes[right_match], None)
# Then remove the nodes (needs to be done backwards):
for left_match, right_match in reversed(matches):
@@ -185,16 +183,16 @@ class Differ(object):
if node in self._text_cache:
return self._text_cache[node]
# Get the texts and the tag as a start
- texts = node.xpath('text()')
+ texts = node.xpath("text()")
# Then add attributes and values
for tag, value in sorted(node.attrib.items()):
- if tag[0] == '{':
- tag = tag.split('}',)[-1]
- texts.append('%s:%s' % (tag, value))
+ if tag[0] == "{":
+ tag = tag.split("}",)[-1]
+ texts.append(f"{tag}:{value}")
# Finally make one string, useful to see how similar two nodes are
- text = u' '.join(texts).strip()
+ text = " ".join(texts).strip()
result = utils.cleanup_whitespace(text)
self._text_cache[node] = result
return result
@@ -256,8 +254,7 @@ class Differ(object):
# Move: Check if any of the new attributes have the same value
# as the removed attributes. If they do, it's actually
# a renaming, and a move is one action instead of remove + insert
- newattrmap = {v: k for (k, v) in right.attrib.items()
- if k in new_keys}
+ newattrmap = {v: k for (k, v) in right.attrib.items() if k in new_keys}
for lk in sorted(removed_keys):
value = left.attrib[lk]
if value in newattrmap:
@@ -332,19 +329,23 @@ class Differ(object):
return i
def align_children(self, left, right):
- lchildren = [c for c in left.getchildren()
- if (id(c) in self._l2rmap and
- self._l2rmap[id(c)].getparent() is right)]
- rchildren = [c for c in right.getchildren()
- if (id(c) in self._r2lmap and
- self._r2lmap[id(c)].getparent() is left)]
+ lchildren = [
+ c
+ for c in left.getchildren()
+ if (id(c) in self._l2rmap and self._l2rmap[id(c)].getparent() is right)
+ ]
+ rchildren = [
+ c
+ for c in right.getchildren()
+ if (id(c) in self._r2lmap and self._r2lmap[id(c)].getparent() is left)
+ ]
if not lchildren or not rchildren:
# Nothing to align
return
lcs = utils.longest_common_subsequence(
- lchildren, rchildren,
- lambda x, y: self._l2rmap[id(x)] is y)
+ lchildren, rchildren, lambda x, y: self._l2rmap[id(x)] is y
+ )
for x, y in lcs:
# Mark these as in order
@@ -362,9 +363,8 @@ class Differ(object):
rtarget = rchild.getparent()
ltarget = self._r2lmap[id(rtarget)]
yield actions.MoveNode(
- utils.getpath(lchild),
- utils.getpath(ltarget),
- right_pos)
+ utils.getpath(lchild), utils.getpath(ltarget), right_pos
+ )
# Do the actual move:
left.remove(lchild)
ltarget.insert(right_pos, lchild)
@@ -394,11 +394,13 @@ class Differ(object):
# (ii)
if rnode.tag is etree.Comment:
yield actions.InsertComment(
- utils.getpath(ltarget, ltree), pos, rnode.text)
+ utils.getpath(ltarget, ltree), pos, rnode.text
+ )
lnode = etree.Comment(rnode.text)
else:
- yield actions.InsertNode(utils.getpath(ltarget, ltree),
- rnode.tag, pos)
+ yield actions.InsertNode(
+ utils.getpath(ltarget, ltree), rnode.tag, pos
+ )
lnode = ltarget.makeelement(rnode.tag)
# (iii)
@@ -409,8 +411,7 @@ class Differ(object):
# And then we update attributes. This is different from the
# paper, because the paper assumes nodes only has labels and
# values. Nodes also has texts, we do them later.
- for action in self.update_node_attr(lnode, rnode):
- yield action
+ yield from self.update_node_attr(lnode, rnode)
# (c)
else:
@@ -426,9 +427,8 @@ class Differ(object):
if ltarget is not lparent:
pos = self.find_pos(rnode)
yield actions.MoveNode(
- utils.getpath(lnode, ltree),
- utils.getpath(ltarget, ltree),
- pos)
+ utils.getpath(lnode, ltree), utils.getpath(ltarget, ltree), pos
+ )
# Move the node from current parent to target
lparent.remove(lnode)
ltarget.insert(pos, lnode)
@@ -436,26 +436,22 @@ class Differ(object):
self._inorder.add(rnode)
# Rename
- for action in self.update_node_tag(lnode, rnode):
- yield action
+ yield from self.update_node_tag(lnode, rnode)
# (ii) Update
# XXX If they are exactly equal, we can skip this,
# maybe store match results in a cache?
- for action in self.update_node_attr(lnode, rnode):
- yield action
+ yield from self.update_node_attr(lnode, rnode)
# (d) Align
- for action in self.align_children(lnode, rnode):
- yield action
+ yield from self.align_children(lnode, rnode)
# And lastly, we update all node texts. We do this after
# aligning children, because when you generate an XML diff
# from this, that XML diff update generates more children,
# confusing later inserts or deletes.
lnode = self._r2lmap[id(rnode)]
- for action in self.update_node_text(lnode, rnode):
- yield action
+ yield from self.update_node_text(lnode, rnode)
for lnode in utils.reverse_post_order_traverse(self.left):
if id(lnode) not in self._l2rmap:
Index: xmldiff-2.4/xmldiff/diff_match_patch.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/diff_match_patch.py
+++ xmldiff-2.4/xmldiff/diff_match_patch.py
@@ -1,4 +1,5 @@
import sys
+
if sys.version_info[0] == 3:
from xmldiff._diff_match_patch_py3 import *
else:
Index: xmldiff-2.4/xmldiff/formatting.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/formatting.py
+++ xmldiff-2.4/xmldiff/formatting.py
@@ -1,6 +1,5 @@
import json
import re
-import six
from collections import namedtuple
from copy import deepcopy
@@ -9,12 +8,12 @@ from xmldiff.diff_match_patch import dif
from xmldiff import utils
-DIFF_NS = 'http://namespaces.shoobx.com/diff'
-DIFF_PREFIX = 'diff'
+DIFF_NS = "http://namespaces.shoobx.com/diff"
+DIFF_PREFIX = "diff"
-INSERT_NAME = '{%s}insert' % DIFF_NS
-DELETE_NAME = '{%s}delete' % DIFF_NS
-RENAME_NAME = '{%s}rename' % DIFF_NS
+INSERT_NAME = "{%s}insert" % DIFF_NS
+DELETE_NAME = "{%s}delete" % DIFF_NS
+RENAME_NAME = "{%s}rename" % DIFF_NS
# Flags for whitespace handling in the text aware formatters:
WS_BOTH = 3 # Normalize ignorable whitespace and text whitespace
@@ -34,15 +33,15 @@ T_SINGLE = 2
# that have narrow builds, we can change this to 0xf00000, which is
# the start of two 64,000 private use blocks.
# PY3: Once Python 2.7 support is dropped we should change this to 0xf00000
-PLACEHOLDER_START = 0xe000
+PLACEHOLDER_START = 0xE000
# These Bases can be abstract baseclasses, but it's a pain to support
# Python 2.7 in that case, because there is no abc.ABC. Right now this
# is just a description of the API.
-class BaseFormatter(object):
+class BaseFormatter:
def __init__(self, normalize=WS_TAGS, pretty_print=False):
"""Formatters must as a minimum have a normalize parameter
@@ -75,10 +74,10 @@ class BaseFormatter(object):
"""
-PlaceholderEntry = namedtuple('PlaceholderEntry', 'element ttype close_ph')
+PlaceholderEntry = namedtuple("PlaceholderEntry", "element ttype close_ph")
-class PlaceholderMaker(object):
+class PlaceholderMaker:
"""Replace tags with unicode placeholders
This class searches for certain tags in an XML tree and replaces them
@@ -98,20 +97,17 @@ class PlaceholderMaker(object):
self.placeholder = PLACEHOLDER_START
insert_elem = etree.Element(INSERT_NAME)
- insert_close = self.get_placeholder(
- insert_elem, T_CLOSE, None)
- insert_open = self.get_placeholder(
- insert_elem, T_OPEN, insert_close)
+ insert_close = self.get_placeholder(insert_elem, T_CLOSE, None)
+ insert_open = self.get_placeholder(insert_elem, T_OPEN, insert_close)
delete_elem = etree.Element(DELETE_NAME)
- delete_close = self.get_placeholder(
- delete_elem, T_CLOSE, None)
- delete_open = self.get_placeholder(
- delete_elem, T_OPEN, delete_close)
+ delete_close = self.get_placeholder(delete_elem, T_CLOSE, None)
+ delete_open = self.get_placeholder(delete_elem, T_OPEN, delete_close)
self.diff_tags = {
- 'insert': (insert_open, insert_close),
- 'delete': (delete_open, delete_close)}
+ "insert": (insert_open, insert_close),
+ "delete": (delete_open, delete_close),
+ }
def get_placeholder(self, element, ttype, close_ph):
tag = etree.tounicode(element)
@@ -120,7 +116,7 @@ class PlaceholderMaker(object):
return ph
self.placeholder += 1
- ph = six.unichr(self.placeholder)
+ ph = chr(self.placeholder)
self.placeholder2tag[ph] = PlaceholderEntry(element, ttype, close_ph)
self.tag2placeholder[tag, ttype, close_ph] = ph
return ph
@@ -135,17 +131,17 @@ class PlaceholderMaker(object):
for child in element:
# Resolve all formatting text by allowing the inside text to
# participate in the text diffing.
- tail = child.tail or u''
- child.tail = u''
- new_text = element.text or u''
+ tail = child.tail or ""
+ child.tail = ""
+ new_text = element.text or ""
if self.is_formatting(child):
ph_close = self.get_placeholder(child, T_CLOSE, None)
ph_open = self.get_placeholder(child, T_OPEN, ph_close)
# If it's known text formatting tags, do this hierarchically
self.do_element(child)
- text = child.text or u''
- child.text = u''
+ text = child.text or ""
+ child.text = ""
# Stick the placeholder in instead of the start and end tags:
element.text = new_text + ph_open + text + ph_close + tail
else:
@@ -159,15 +155,15 @@ class PlaceholderMaker(object):
def do_tree(self, tree):
if self.text_tags:
- for elem in tree.xpath('//'+'|//'.join(self.text_tags)):
+ for elem in tree.xpath("//" + "|//".join(self.text_tags)):
self.do_element(elem)
def split_string(self, text):
- regexp = u'([%s])' % u''.join(self.placeholder2tag)
+ regexp = "([%s])" % "".join(self.placeholder2tag)
return re.split(regexp, text, flags=re.MULTILINE)
def undo_string(self, text):
- result = etree.Element('wrap')
+ result = etree.Element("wrap")
element = None
segments = self.split_string(text)
@@ -184,7 +180,7 @@ class PlaceholderMaker(object):
if entry.ttype == T_OPEN:
# Yup
next_seg = segments.pop(0)
- new_text = u''
+ new_text = ""
while next_seg != entry.close_ph:
new_text += next_seg
next_seg = segments.pop(0)
@@ -195,9 +191,9 @@ class PlaceholderMaker(object):
result.append(element)
else:
if element is not None:
- element.tail = element.tail or u'' + seg
+ element.tail = element.tail or "" + seg
else:
- result.text = result.text or u'' + seg
+ result.text = result.text or "" + seg
return result
@@ -244,8 +240,8 @@ class PlaceholderMaker(object):
elem = deepcopy(elem)
if self.is_formatting(elem):
# Formatting element, add a diff attribute
- action += '-formatting'
- elem.attrib['{%s}%s' % (DIFF_NS, action)] = ''
+ action += "-formatting"
+ elem.attrib[f"{{{DIFF_NS}}}{action}"] = ""
# And make a new placeholder for this new entry:
return self.get_placeholder(elem, entry.ttype, entry.close_ph)
@@ -301,15 +297,17 @@ class XMLFormatter(BaseFormatter):
all whitespace.
"""
- def __init__(self, normalize=WS_NONE, pretty_print=True,
- text_tags=(), formatting_tags=()):
+ def __init__(
+ self, normalize=WS_NONE, pretty_print=True, text_tags=(), formatting_tags=()
+ ):
# Mapping from placeholders -> structural content and vice versa.
self.normalize = normalize
self.pretty_print = pretty_print
self.text_tags = text_tags
self.formatting_tags = formatting_tags
self.placeholderer = PlaceholderMaker(
- text_tags=text_tags, formatting_tags=formatting_tags)
+ text_tags=text_tags, formatting_tags=formatting_tags
+ )
def prepare(self, left_tree, right_tree):
"""prepare() is run on the trees before diffing
@@ -352,11 +350,11 @@ class XMLFormatter(BaseFormatter):
def handle_action(self, action, result):
action_type = type(action)
- method = getattr(self, '_handle_' + action_type.__name__)
+ method = getattr(self, "_handle_" + action_type.__name__)
method(action, result)
def _remove_comments(self, tree):
- comments = tree.xpath('//comment()')
+ comments = tree.xpath("//comment()")
for element in comments:
parent = element.getparent()
@@ -371,20 +369,20 @@ class XMLFormatter(BaseFormatter):
# one and exactly one element is found. This is to protect against
# formatting a diff on the wrong tree, or against using ambiguous
# edit script xpaths.
- if xpath[0] == '/':
+ if xpath[0] == "/":
root = True
xpath = xpath[1:]
else:
root = False
- if '/' in xpath:
- path, rest = xpath.split('/', 1)
+ if "/" in xpath:
+ path, rest = xpath.split("/", 1)
else:
path = xpath
- rest = ''
+ rest = ""
- if '[' in path:
- path, index = path[:-1].split('[')
+ if "[" in path:
+ path, index = path[:-1].split("[")
index = int(index) - 1
multiple = False
else:
@@ -392,7 +390,7 @@ class XMLFormatter(BaseFormatter):
multiple = True
if root:
- path = '/' + path
+ path = "/" + path
matches = []
for match in node.xpath(path, namespaces=node.nsmap):
@@ -401,33 +399,39 @@ class XMLFormatter(BaseFormatter):
matches.append(match)
if index >= len(matches):
- raise ValueError('xpath %s[%s] not found at %s.' % (
- path, index + 1, utils.getpath(node)))
+ raise ValueError(
+ "xpath {}[{}] not found at {}.".format(
+ path, index + 1, utils.getpath(node)
+ )
+ )
if len(matches) > 1 and multiple:
- raise ValueError('Multiple nodes found for xpath %s at %s.' % (
- path, utils.getpath(node)))
+ raise ValueError(
+ "Multiple nodes found for xpath {} at {}.".format(
+ path, utils.getpath(node)
+ )
+ )
match = matches[index]
if rest:
return self._xpath(match, rest)
return match
def _extend_diff_attr(self, node, action, value):
- diffattr = '{%s}%s-attr' % (DIFF_NS, action)
- oldvalue = node.attrib.get(diffattr, '')
+ diffattr = f"{{{DIFF_NS}}}{action}-attr"
+ oldvalue = node.attrib.get(diffattr, "")
if oldvalue:
- value = oldvalue + ';' + value
+ value = oldvalue + ";" + value
node.attrib[diffattr] = value
def _delete_attrib(self, node, name):
del node.attrib[name]
- self._extend_diff_attr(node, 'delete', name)
+ self._extend_diff_attr(node, "delete", name)
def _handle_DeleteAttrib(self, action, tree):
node = self._xpath(tree, action.node)
self._delete_attrib(node, action.name)
def _delete_node(self, node):
- node.attrib[DELETE_NAME] = ''
+ node.attrib[DELETE_NAME] = ""
def _handle_DeleteNode(self, action, tree):
node = self._xpath(tree, action.node)
@@ -435,14 +439,14 @@ class XMLFormatter(BaseFormatter):
def _insert_attrib(self, node, name, value):
node.attrib[name] = value
- self._extend_diff_attr(node, 'add', name)
+ self._extend_diff_attr(node, "add", name)
def _handle_InsertAttrib(self, action, tree):
node = self._xpath(tree, action.node)
self._insert_attrib(node, action.name, action.value)
def _insert_node(self, target, node, position):
- node.attrib[INSERT_NAME] = ''
+ node.attrib[INSERT_NAME] = ""
target.insert(position, node)
def _get_real_insert_position(self, target, position):
@@ -472,7 +476,7 @@ class XMLFormatter(BaseFormatter):
def _rename_attrib(self, node, oldname, newname):
node.attrib[newname] = node.attrib[oldname]
del node.attrib[oldname]
- self._extend_diff_attr(node, 'rename', '%s:%s' % (oldname, newname))
+ self._extend_diff_attr(node, "rename", f"{oldname}:{newname}")
def _handle_RenameAttrib(self, action, tree):
node = self._xpath(tree, action.node)
@@ -494,7 +498,7 @@ class XMLFormatter(BaseFormatter):
def _update_attrib(self, node, name, value):
oldval = node.attrib[name]
node.attrib[name] = value
- self._extend_diff_attr(node, 'update', '%s:%s' % (name, oldval))
+ self._extend_diff_attr(node, "update", f"{name}:{oldval}")
def _handle_UpdateAttrib(self, action, tree):
node = self._xpath(tree, action.node)
@@ -540,10 +544,7 @@ class XMLFormatter(BaseFormatter):
# needs to be closed before the requested node closure can
# happen.
stack_op, stack_entry = _stack_pop()
- while (
- stack_entry is not None and
- stack_entry.close_ph != seg
- ):
+ while stack_entry is not None and stack_entry.close_ph != seg:
new_diff.append((stack_op, stack_entry.close_ph))
stack_op, stack_entry = _stack_pop()
@@ -564,11 +565,11 @@ class XMLFormatter(BaseFormatter):
def _make_diff_tags(self, left_value, right_value, node, target=None):
if bool(self.normalize & WS_TEXT):
- left_value = utils.cleanup_whitespace(left_value or u'').strip()
- right_value = utils.cleanup_whitespace(right_value or u'').strip()
+ left_value = utils.cleanup_whitespace(left_value or "").strip()
+ right_value = utils.cleanup_whitespace(right_value or "").strip()
text_diff = diff_match_patch()
- diff = text_diff.diff_main(left_value or '', right_value or '')
+ diff = text_diff.diff_main(left_value or "", right_value or "")
text_diff.diff_cleanupSemantic(diff)
diff = self._realign_placeholders(diff)
@@ -582,29 +583,29 @@ class XMLFormatter(BaseFormatter):
for op, text in diff:
if op == 0:
if cur_child is None:
- node.text = (node.text or u'') + text
+ node.text = (node.text or "") + text
else:
- cur_child.tail = (cur_child.tail or u'') + text
+ cur_child.tail = (cur_child.tail or "") + text
continue
if op == -1:
- action = 'delete'
+ action = "delete"
elif op == 1:
- action = 'insert'
+ action = "insert"
if self.placeholderer.is_placeholder(text):
ph = self.placeholderer.mark_diff(text, action)
if cur_child is None:
- node.text = (node.text or u'') + ph
+ node.text = (node.text or "") + ph
else:
new_text = self.placeholderer.wrap_diff(text, action)
if cur_child is None:
- node.text = (node.text or u'') + new_text
+ node.text = (node.text or "") + new_text
else:
- cur_child.tail = (cur_child.tail or u'') + new_text
+ cur_child.tail = (cur_child.tail or "") + new_text
def _handle_UpdateTextIn(self, action, tree):
node = self._xpath(tree, action.node)
@@ -635,66 +636,70 @@ class XMLFormatter(BaseFormatter):
class DiffFormatter(BaseFormatter):
-
def __init__(self, normalize=WS_TAGS, pretty_print=False):
self.normalize = normalize
# No pretty print support, nothing to be pretty about
# Nothing to prepare or finalize (one-liners for code coverage)
- def prepare(self, left, right): return
+ def prepare(self, left, right):
+ return
- def finalize(self, left, right): return
+ def finalize(self, left, right):
+ return
def format(self, diff, orig_tree):
# This Formatter don't need the left tree, but the XMLFormatter
# does, so the parameter is required.
- res = u'\n'.join(self._format_action(action) for action in diff)
+ res = "\n".join(self._format_action(action) for action in diff)
return res
- def _format_action(self, action, ):
- return u'[%s]' % self.handle_action(action)
+ def _format_action(
+ self, action,
+ ):
+ return "[%s]" % self.handle_action(action)
def handle_action(self, action):
action_type = type(action)
- method = getattr(self, '_handle_' + action_type.__name__)
- return u', '.join(method(action))
+ method = getattr(self, "_handle_" + action_type.__name__)
+ return ", ".join(method(action))
def _handle_DeleteAttrib(self, action):
- return u"delete-attribute", action.node, action.name
+ return "delete-attribute", action.node, action.name
def _handle_DeleteNode(self, action):
- return u"delete", action.node
+ return "delete", action.node
def _handle_InsertAttrib(self, action):
- return (u"insert-attribute", action.node, action.name,
- json.dumps(action.value))
+ return ("insert-attribute", action.node, action.name, json.dumps(action.value))
def _handle_InsertNode(self, action):
- return u"insert", action.target, action.tag, str(action.position)
+ return "insert", action.target, action.tag, str(action.position)
def _handle_RenameAttrib(self, action):
- return (u"rename-attribute", action.node, action.oldname,
- action.newname)
+ return ("rename-attribute", action.node, action.oldname, action.newname)
def _handle_MoveNode(self, action):
- return u"move", action.node, action.target, str(action.position)
+ return "move", action.node, action.target, str(action.position)
def _handle_UpdateAttrib(self, action):
- return (u"update-attribute", action.node, action.name,
- json.dumps(action.value))
+ return ("update-attribute", action.node, action.name, json.dumps(action.value))
def _handle_UpdateTextIn(self, action):
- return u"update-text", action.node, json.dumps(action.text)
+ return "update-text", action.node, json.dumps(action.text)
def _handle_UpdateTextAfter(self, action):
- return u"update-text-after", action.node, json.dumps(action.text)
+ return "update-text-after", action.node, json.dumps(action.text)
def _handle_RenameNode(self, action):
- return u"rename", action.node, action.tag
+ return "rename", action.node, action.tag
def _handle_InsertComment(self, action):
- return (u"insert-comment", action.target, str(action.position),
- json.dumps(action.text))
+ return (
+ "insert-comment",
+ action.target,
+ str(action.position),
+ json.dumps(action.text),
+ )
class XmlDiffFormatter(BaseFormatter):
@@ -705,9 +710,11 @@ class XmlDiffFormatter(BaseFormatter):
# No pretty print support, nothing to be pretty about
# Nothing to prepare or finalize (one-liners for code coverage)
- def prepare(self, left, right): return
+ def prepare(self, left, right):
+ return
- def finalize(self, left, right): return
+ def finalize(self, left, right):
+ return
def format(self, diff, orig_tree):
# This Formatter don't need the left tree, but the XMLFormatter
@@ -715,45 +722,44 @@ class XmlDiffFormatter(BaseFormatter):
actions = []
for action in diff:
actions.extend(self.handle_action(action, orig_tree))
- res = u'\n'.join(self._format_action(action) for action in actions)
+ res = "\n".join(self._format_action(action) for action in actions)
return res
def _format_action(self, action):
- return u'[%s]' % ', '.join(action)
+ return "[%s]" % ", ".join(action)
def handle_action(self, action, orig_tree):
action_type = type(action)
- method = getattr(self, '_handle_' + action_type.__name__)
- for item in method(action, orig_tree):
- yield item
+ method = getattr(self, "_handle_" + action_type.__name__)
+ yield from method(action, orig_tree)
def _handle_DeleteAttrib(self, action, orig_tree):
- yield u"remove", '%s/@%s' % (action.node, action.name)
+ yield "remove", f"{action.node}/@{action.name}"
def _handle_DeleteNode(self, action, orig_tree):
- yield u"remove", action.node
+ yield "remove", action.node
def _handle_InsertAttrib(self, action, orig_tree):
value_text = "\n<@{0}>\n{1}\n</@{0}>".format(action.name, action.value)
- yield u"insert", action.node, value_text
+ yield "insert", action.node, value_text
def _handle_InsertNode(self, action, orig_tree):
if action.position == 0:
- yield u"insert-first", action.target, '\n<%s/>' % action.tag
+ yield "insert-first", action.target, "\n<%s/>" % action.tag
return
sibling = orig_tree.xpath(action.target)[0][action.position - 1]
- yield u"insert-after", utils.getpath(sibling), '\n<%s/>' % action.tag
+ yield "insert-after", utils.getpath(sibling), "\n<%s/>" % action.tag
def _handle_RenameAttrib(self, action, orig_tree):
node = orig_tree.xpath(action.node)[0]
value = node.attrib[action.oldname]
value_text = "\n<@{0}>\n{1}\n</@{0}>".format(action.newname, value)
- yield u"remove", '%s/@%s' % (action.node, action.oldname)
- yield u"insert", action.node, value_text
+ yield "remove", f"{action.node}/@{action.oldname}"
+ yield "insert", action.node, value_text
def _handle_MoveNode(self, action, orig_tree):
if action.position == 0:
- yield u"move-first", action.node, action.target
+ yield "move-first", action.node, action.target
return
node = orig_tree.xpath(action.node)[0]
target = orig_tree.xpath(action.target)[0]
@@ -766,21 +772,23 @@ class XmlDiffFormatter(BaseFormatter):
position += 1
sibling = target[position]
- yield u"move-after", action.node, utils.getpath(sibling)
+ yield "move-after", action.node, utils.getpath(sibling)
def _handle_UpdateAttrib(self, action, orig_tree):
- yield (u"update", '%s/@%s' % (action.node, action.name),
- json.dumps(action.value))
+ yield (
+ "update",
+ f"{action.node}/@{action.name}",
+ json.dumps(action.value),
+ )
def _handle_UpdateTextIn(self, action, orig_tree):
- yield u"update", action.node + '/text()[1]', json.dumps(action.text)
+ yield "update", action.node + "/text()[1]", json.dumps(action.text)
def _handle_UpdateTextAfter(self, action, orig_tree):
- yield u"update", action.node + '/text()[2]', json.dumps(action.text)
+ yield "update", action.node + "/text()[2]", json.dumps(action.text)
def _handle_RenameNode(self, action, orig_tree):
- yield u"rename", action.node, action.tag
+ yield "rename", action.node, action.tag
def _handle_InsertComment(self, action, orig_tree):
- yield (u"insert-comment", action.target, str(action.position),
- action.text)
+ yield ("insert-comment", action.target, str(action.position), action.text)
Index: xmldiff-2.4/xmldiff/main.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/main.py
+++ xmldiff-2.4/xmldiff/main.py
@@ -1,6 +1,5 @@
"""All major API points and command-line tools"""
import pkg_resources
-import six
from argparse import ArgumentParser, FileType
from lxml import etree
@@ -9,9 +8,9 @@ from xmldiff import diff, formatting, pa
__version__ = pkg_resources.require("xmldiff")[0].version
FORMATTERS = {
- 'diff': formatting.DiffFormatter,
- 'xml': formatting.XMLFormatter,
- 'old': formatting.XmlDiffFormatter,
+ "diff": formatting.DiffFormatter,
+ "xml": formatting.XMLFormatter,
+ "old": formatting.XmlDiffFormatter,
}
@@ -31,59 +30,89 @@ def diff_trees(left, right, diff_options
def _diff(parse_method, left, right, diff_options=None, formatter=None):
- normalize = bool(getattr(formatter, 'normalize', 1) & formatting.WS_TAGS)
+ normalize = bool(getattr(formatter, "normalize", 1) & formatting.WS_TAGS)
parser = etree.XMLParser(remove_blank_text=normalize)
left_tree = parse_method(left, parser)
right_tree = parse_method(right, parser)
- return diff_trees(left_tree, right_tree, diff_options=diff_options,
- formatter=formatter)
+ return diff_trees(
+ left_tree, right_tree, diff_options=diff_options, formatter=formatter
+ )
def diff_texts(left, right, diff_options=None, formatter=None):
"""Takes two Unicode strings containing XML"""
- return _diff(etree.fromstring, left, right,
- diff_options=diff_options, formatter=formatter)
+ return _diff(
+ etree.fromstring, left, right, diff_options=diff_options, formatter=formatter
+ )
def diff_files(left, right, diff_options=None, formatter=None):
"""Takes two filenames or streams, and diffs the XML in those files"""
- return _diff(etree.parse, left, right,
- diff_options=diff_options, formatter=formatter)
+ return _diff(
+ etree.parse, left, right, diff_options=diff_options, formatter=formatter
+ )
def make_diff_parser():
- parser = ArgumentParser(description='Create a diff for two XML files.',
- add_help=False)
- parser.add_argument('file1', type=FileType('r'),
- help='The first input file.')
- parser.add_argument('file2', type=FileType('r'),
- help='The second input file.')
- parser.add_argument('-h', '--help', action='help',
- help='Show this help message and exit.')
- parser.add_argument('-v', '--version', action='version',
- help='Display version and exit.',
- version='xmldiff %s' % __version__)
- parser.add_argument('-f', '--formatter', default='diff',
- choices=list(FORMATTERS.keys()),
- help='Formatter selection.')
- parser.add_argument('-w', '--keep-whitespace', action='store_true',
- help='Do not strip ignorable whitespace.')
- parser.add_argument('-p', '--pretty-print', action='store_true',
- help='Try to make XML output more readable.')
- parser.add_argument('-F', type=float,
- help='A value between 0 and 1 that determines how '
- 'similar nodes must be to match.')
- parser.add_argument('--unique-attributes', type=str, nargs='?',
- default='{http://www.w3.org/XML/1998/namespace}id',
- help='A comma separated list of attributes '
- 'that uniquely identify a node. Can be empty. '
- 'Unique attributes for certain elements can '
- 'be specified in the format {NS}element@attr.')
- parser.add_argument('--ratio-mode', default='fast',
- choices={'accurate', 'fast', 'faster'},
- help='Choose the node comparison optimization.')
- parser.add_argument('--fast-match', action='store_true',
- help='A faster, less optimal match run.')
+ parser = ArgumentParser(
+ description="Create a diff for two XML files.", add_help=False
+ )
+ parser.add_argument("file1", type=FileType("r"), help="The first input file.")
+ parser.add_argument("file2", type=FileType("r"), help="The second input file.")
+ parser.add_argument(
+ "-h", "--help", action="help", help="Show this help message and exit."
+ )
+ parser.add_argument(
+ "-v",
+ "--version",
+ action="version",
+ help="Display version and exit.",
+ version="xmldiff %s" % __version__,
+ )
+ parser.add_argument(
+ "-f",
+ "--formatter",
+ default="diff",
+ choices=list(FORMATTERS.keys()),
+ help="Formatter selection.",
+ )
+ parser.add_argument(
+ "-w",
+ "--keep-whitespace",
+ action="store_true",
+ help="Do not strip ignorable whitespace.",
+ )
+ parser.add_argument(
+ "-p",
+ "--pretty-print",
+ action="store_true",
+ help="Try to make XML output more readable.",
+ )
+ parser.add_argument(
+ "-F",
+ type=float,
+ help="A value between 0 and 1 that determines how "
+ "similar nodes must be to match.",
+ )
+ parser.add_argument(
+ "--unique-attributes",
+ type=str,
+ nargs="?",
+ default="{http://www.w3.org/XML/1998/namespace}id",
+ help="A comma separated list of attributes "
+ "that uniquely identify a node. Can be empty. "
+ "Unique attributes for certain elements can "
+ "be specified in the format {NS}element@attr.",
+ )
+ parser.add_argument(
+ "--ratio-mode",
+ default="fast",
+ choices={"accurate", "fast", "faster"},
+ help="Choose the node comparison optimization.",
+ )
+ parser.add_argument(
+ "--fast-match", action="store_true", help="A faster, less optimal match run."
+ )
return parser
@@ -91,8 +120,8 @@ def _parse_uniqueattrs(uniqueattrs):
if uniqueattrs is None:
return []
return [
- attr if '@' not in attr else attr.split('@', 1)
- for attr in uniqueattrs.split(',')
+ attr if "@" not in attr else attr.split("@", 1)
+ for attr in uniqueattrs.split(",")
]
@@ -105,16 +134,19 @@ def diff_command(args=None):
else:
normalize = formatting.WS_BOTH
- formatter = FORMATTERS[args.formatter](normalize=normalize,
- pretty_print=args.pretty_print)
-
- diff_options = {'ratio_mode': args.ratio_mode,
- 'F': args.F,
- 'fast_match': args.fast_match,
- 'uniqueattrs': _parse_uniqueattrs(args.unique_attributes),
- }
- result = diff_files(args.file1, args.file2, diff_options=diff_options,
- formatter=formatter)
+ formatter = FORMATTERS[args.formatter](
+ normalize=normalize, pretty_print=args.pretty_print
+ )
+
+ diff_options = {
+ "ratio_mode": args.ratio_mode,
+ "F": args.F,
+ "fast_match": args.fast_match,
+ "uniqueattrs": _parse_uniqueattrs(args.unique_attributes),
+ }
+ result = diff_files(
+ args.file1, args.file2, diff_options=diff_options, formatter=formatter
+ )
print(result)
@@ -136,7 +168,7 @@ def patch_file(actions, tree):
"""Takes two filenames or streams, one with XML the other a diff"""
tree = etree.parse(tree)
- if isinstance(actions, six.string_types):
+ if isinstance(actions, str):
# It's a string, so it's a filename
with open(actions) as f:
actions = f.read()
@@ -150,17 +182,21 @@ def patch_file(actions, tree):
def make_patch_parser():
- parser = ArgumentParser(description='Patch an XML file with an xmldiff',
- add_help=False)
- parser.add_argument('patchfile', type=FileType('r'),
- help='An xmldiff diff file.')
- parser.add_argument('xmlfile', type=FileType('r'),
- help='An unpatched XML file.')
- parser.add_argument('-h', '--help', action='help',
- help='Show this help message and exit.')
- parser.add_argument('-v', '--version', action='version',
- help='Display version and exit.',
- version='xmldiff %s' % __version__)
+ parser = ArgumentParser(
+ description="Patch an XML file with an xmldiff", add_help=False
+ )
+ parser.add_argument("patchfile", type=FileType("r"), help="An xmldiff diff file.")
+ parser.add_argument("xmlfile", type=FileType("r"), help="An unpatched XML file.")
+ parser.add_argument(
+ "-h", "--help", action="help", help="Show this help message and exit."
+ )
+ parser.add_argument(
+ "-v",
+ "--version",
+ action="version",
+ help="Display version and exit.",
+ version="xmldiff %s" % __version__,
+ )
return parser
Index: xmldiff-2.4/xmldiff/patch.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/patch.py
+++ xmldiff-2.4/xmldiff/patch.py
@@ -5,8 +5,7 @@ from lxml import etree
from xmldiff import actions
-class Patcher(object):
-
+class Patcher:
def patch(self, actions, tree):
# Copy the tree so we don't modify the original
result = deepcopy(tree)
@@ -18,7 +17,7 @@ class Patcher(object):
def handle_action(self, action, tree):
action_type = type(action)
- method = getattr(self, '_handle_' + action_type.__name__)
+ method = getattr(self, "_handle_" + action_type.__name__)
method(action, tree)
def _handle_DeleteNode(self, action, tree):
@@ -72,25 +71,25 @@ class Patcher(object):
target.insert(action.position, etree.Comment(action.text))
-class DiffParser(object):
+class DiffParser:
"""Makes a text diff into a list of actions"""
def parse(self, diff):
- incomplete = ''
+ incomplete = ""
for line in diff.splitlines():
line = incomplete + line
- if line[0] != '[':
+ if line[0] != "[":
# All actions should start with "["
raise ValueError("Unknown diff format")
- if line[-1] != ']':
+ if line[-1] != "]":
# This line has been broken into several lines
incomplete = line
continue
# OK, we found an action
- incomplete = ''
+ incomplete = ""
yield self.make_action(line)
if incomplete:
@@ -105,7 +104,7 @@ class DiffParser(object):
action = parts[0]
params = parts[1:]
# Get the method, and return the result of calling it
- method = getattr(self, '_handle_' + action.replace('-', '_'))
+ method = getattr(self, "_handle_" + action.replace("-", "_"))
return method(*params)
def _handle_delete(self, node):
Index: xmldiff-2.4/xmldiff/utils.py
===================================================================
--- xmldiff-2.4.orig/xmldiff/utils.py
+++ xmldiff-2.4/xmldiff/utils.py
@@ -1,5 +1,3 @@
-from __future__ import division
-
import re
from operator import eq
@@ -8,16 +6,14 @@ from operator import eq
def post_order_traverse(node):
for child in node.getchildren():
# PY3: Man, I want yield from!
- for item in post_order_traverse(child):
- yield item
+ yield from post_order_traverse(child)
yield node
def reverse_post_order_traverse(node):
for child in reversed(node.getchildren()):
# PY3: Man, I want yield from!
- for item in reverse_post_order_traverse(child):
- yield item
+ yield from reverse_post_order_traverse(child)
yield node
@@ -43,13 +39,19 @@ def longest_common_subsequence(left_sequ
rend = rslen = len(right_sequence)
# Trim off the matching items at the beginning
- while (start < lend and start < rend and
- eqfn(left_sequence[start], right_sequence[start])):
+ while (
+ start < lend
+ and start < rend
+ and eqfn(left_sequence[start], right_sequence[start])
+ ):
start += 1
# trim off the matching items at the end
- while (start < lend and start < rend and
- eqfn(left_sequence[lend - 1], right_sequence[rend - 1])):
+ while (
+ start < lend
+ and start < rend
+ and eqfn(left_sequence[lend - 1], right_sequence[rend - 1])
+ ):
lend -= 1
rend -= 1
@@ -67,8 +69,7 @@ def longest_common_subsequence(left_sequ
for d in range(0, lmax + rmax + 1):
for k in range(-d, d + 1, 2):
- if (k == -d or
- (k != d and furthest[k - 1][0] < furthest[k + 1][0])):
+ if k == -d or (k != d and furthest[k - 1][0] < furthest[k + 1][0]):
# Go down
old_x, history = furthest[k + 1]
x = old_x
@@ -89,47 +90,51 @@ def longest_common_subsequence(left_sequ
if x >= lmax and y >= rmax:
# This is the best match
- return [(e, e) for e in range(start)] + history + \
- list(zip(range(lend, lslen), range(rend, rslen)))
+ return (
+ [(e, e) for e in range(start)]
+ + history
+ + list(zip(range(lend, lslen), range(rend, rslen)))
+ )
else:
furthest[k] = (x, history)
-WHITESPACE = re.compile(u'\\s+', flags=re.MULTILINE)
+WHITESPACE = re.compile("\\s+", flags=re.MULTILINE)
def cleanup_whitespace(text):
- return WHITESPACE.sub(' ', text)
+ return WHITESPACE.sub(" ", text)
def getpath(element, tree=None):
if tree is None:
tree = element.getroottree()
xpath = tree.getpath(element)
- if xpath[-1] != ']':
+ if xpath[-1] != "]":
# The path is unique without specifying a count. However, we always
# want that count, so we add [1].
- xpath = xpath + '[1]'
+ xpath = xpath + "[1]"
return xpath
# The remainder of the functions here are helpful when debugging.
# They aren't documented, nor very well tested.
-def _make_ascii_tree(element, indent=''):
+def _make_ascii_tree(element, indent=""):
from xmldiff.formatting import DIFF_NS # Avoid circular imports
- diffns = '{%s}' % DIFF_NS
+
+ diffns = "{%s}" % DIFF_NS
if element.prefix:
- name = '%s:%s' % (element.prefix, element.tag.split('}')[1])
+ name = "{}:{}".format(element.prefix, element.tag.split("}")[1])
else:
name = element.tag
diff_attrs = [attr for attr in element.attrib if attr.startswith(diffns)]
if diff_attrs:
- diff = '(%s)' % ', '.join(attr.split('}')[1] for attr in diff_attrs)
+ diff = "(%s)" % ", ".join(attr.split("}")[1] for attr in diff_attrs)
else:
- diff = ''
+ diff = ""
- result = [' '.join((indent, name, diff))]
- indent = ' ' + indent
+ result = [" ".join((indent, name, diff))]
+ indent = " " + indent
for child in element.getchildren():
result.extend(_make_ascii_tree(child, indent))
@@ -138,4 +143,4 @@ def _make_ascii_tree(element, indent='')
def make_ascii_tree(element):
result = _make_ascii_tree(element)
- return '\n'.join(result)
+ return "\n".join(result)
View Git Blame Copy Permalink