salt/improve-salt.utils.json.find_json-bsc-1213293.patch

From 4e6b445f2dbe8a79d220c697abff946e00b2e57b Mon Sep 17 00:00:00 2001
From: Victor Zhestkov <vzhestkov@suse.com>
Date: Mon, 2 Oct 2023 13:26:20 +0200
Subject: [PATCH] Improve salt.utils.json.find_json (bsc#1213293)

* Improve salt.utils.json.find_json

* Move tests of find_json to pytest
---
 salt/utils/json.py                    |  39 +++++++-
 tests/pytests/unit/utils/test_json.py | 122 ++++++++++++++++++++++++++
 2 files changed, 158 insertions(+), 3 deletions(-)
 create mode 100644 tests/pytests/unit/utils/test_json.py

diff --git a/salt/utils/json.py b/salt/utils/json.py
index 33cdbf401d..0845b64694 100644
--- a/salt/utils/json.py
+++ b/salt/utils/json.py
@@ -32,18 +32,51 @@ def find_json(raw):
     """
     ret = {}
     lines = __split(raw)
+    lengths = list(map(len, lines))
+    starts = []
+    ends = []
+
+    # Search for possible starts end ends of the json fragments
     for ind, _ in enumerate(lines):
+        line = lines[ind].lstrip()
+        if line == "{" or line == "[":
+            starts.append((ind, line))
+        if line == "}" or line == "]":
+            ends.append((ind, line))
+
+    # List all the possible pairs of starts and ends,
+    # and fill the length of each block to sort by size after
+    starts_ends = []
+    for start, start_br in starts:
+        for end, end_br in reversed(ends):
+            if end > start and (
+                (start_br == "{" and end_br == "}")
+                or (start_br == "[" and end_br == "]")
+            ):
+                starts_ends.append((start, end, sum(lengths[start : end + 1])))
+
+    # Iterate through all the possible pairs starting from the largest
+    starts_ends.sort(key=lambda x: (x[2], x[1] - x[0], x[0]), reverse=True)
+    for start, end, _ in starts_ends:
+        working = "\n".join(lines[start : end + 1])
         try:
-            working = "\n".join(lines[ind:])
-        except UnicodeDecodeError:
-            working = "\n".join(salt.utils.data.decode(lines[ind:]))
+            ret = json.loads(working)
+        except ValueError:
+            continue
+        if ret:
+            return ret
 
+    # Fall back to old implementation for backward compatibility
+    # excpecting json after the text
+    for ind, _ in enumerate(lines):
+        working = "\n".join(lines[ind:])
         try:
             ret = json.loads(working)
         except ValueError:
             continue
         if ret:
             return ret
+
     if not ret:
         # Not json, raise an error
         raise ValueError
diff --git a/tests/pytests/unit/utils/test_json.py b/tests/pytests/unit/utils/test_json.py
new file mode 100644
index 0000000000..72b1023003
--- /dev/null
+++ b/tests/pytests/unit/utils/test_json.py
@@ -0,0 +1,122 @@
+"""
+Tests for salt.utils.json
+"""
+
+import textwrap
+
+import pytest
+
+import salt.utils.json
+
+
+def test_find_json():
+    some_junk_text = textwrap.dedent(
+        """
+        Just some junk text
+        with multiline
+        """
+    )
+    some_warning_message = textwrap.dedent(
+        """
+        [WARNING] Test warning message
+        """
+    )
+    test_small_json = textwrap.dedent(
+        """
+        {
+            "local": true
+        }
+        """
+    )
+    test_sample_json = """
+                       {
+                           "glossary": {
+                               "title": "example glossary",
+                               "GlossDiv": {
+                                   "title": "S",
+                                   "GlossList": {
+                                       "GlossEntry": {
+                                           "ID": "SGML",
+                                           "SortAs": "SGML",
+                                           "GlossTerm": "Standard Generalized Markup Language",
+                                           "Acronym": "SGML",
+                                           "Abbrev": "ISO 8879:1986",
+                                           "GlossDef": {
+                                               "para": "A meta-markup language, used to create markup languages such as DocBook.",
+                                               "GlossSeeAlso": ["GML", "XML"]
+                                           },
+                                           "GlossSee": "markup"
+                                       }
+                                   }
+                               }
+                           }
+                       }
+                       """
+    expected_ret = {
+        "glossary": {
+            "GlossDiv": {
+                "GlossList": {
+                    "GlossEntry": {
+                        "GlossDef": {
+                            "GlossSeeAlso": ["GML", "XML"],
+                            "para": (
+                                "A meta-markup language, used to create markup"
+                                " languages such as DocBook."
+                            ),
+                        },
+                        "GlossSee": "markup",
+                        "Acronym": "SGML",
+                        "GlossTerm": "Standard Generalized Markup Language",
+                        "SortAs": "SGML",
+                        "Abbrev": "ISO 8879:1986",
+                        "ID": "SGML",
+                    }
+                },
+                "title": "S",
+            },
+            "title": "example glossary",
+        }
+    }
+
+    # First test the valid JSON
+    ret = salt.utils.json.find_json(test_sample_json)
+    assert ret == expected_ret
+
+    # Now pre-pend some garbage and re-test
+    garbage_prepend_json = f"{some_junk_text}{test_sample_json}"
+    ret = salt.utils.json.find_json(garbage_prepend_json)
+    assert ret == expected_ret
+
+    # Now post-pend some garbage and re-test
+    garbage_postpend_json = f"{test_sample_json}{some_junk_text}"
+    ret = salt.utils.json.find_json(garbage_postpend_json)
+    assert ret == expected_ret
+
+    # Now pre-pend some warning and re-test
+    warning_prepend_json = f"{some_warning_message}{test_sample_json}"
+    ret = salt.utils.json.find_json(warning_prepend_json)
+    assert ret == expected_ret
+
+    # Now post-pend some warning and re-test
+    warning_postpend_json = f"{test_sample_json}{some_warning_message}"
+    ret = salt.utils.json.find_json(warning_postpend_json)
+    assert ret == expected_ret
+
+    # Now put around some garbage and re-test
+    garbage_around_json = f"{some_junk_text}{test_sample_json}{some_junk_text}"
+    ret = salt.utils.json.find_json(garbage_around_json)
+    assert ret == expected_ret
+
+    # Now pre-pend small json and re-test
+    small_json_pre_json = f"{test_small_json}{test_sample_json}"
+    ret = salt.utils.json.find_json(small_json_pre_json)
+    assert ret == expected_ret
+
+    # Now post-pend small json and re-test
+    small_json_post_json = f"{test_sample_json}{test_small_json}"
+    ret = salt.utils.json.find_json(small_json_post_json)
+    assert ret == expected_ret
+
+    # Test to see if a ValueError is raised if no JSON is passed in
+    with pytest.raises(ValueError):
+        ret = salt.utils.json.find_json(some_junk_text)
-- 
2.42.0