Commit Graph

117 Commits

Author SHA1 Message Date
Iain Lane
93f3cb7c6d Revert "glocalfile: Check that parent device is the same when trashing"
This reverts commit 15cdcd2e0b.
2018-08-13 17:23:21 +01:00
Руслан Ижбулатов
075bb0f751 W32: consider NTFS mounpoints to be symlinks
While mountpoints are *not* symlinks, strictly speaking,
they works in a similar enough way, so consider them to be
symlinks for the purpose of querying local file info.
2018-08-08 23:41:07 +00:00
Руслан Ижбулатов
929844fdf7 W32: correctly use st_ctime
On Windows st_ctime field is the file creation time.
POSIX mandates that field to be the file state change time.
Naturally, glib code interpreted st_ctime as POSIX suggested,
and the result was bad.
Fix this by introducing special W32-only logic for setting
attributes from st_ctime field.

Fixes issue #1452.
2018-08-08 23:41:02 +00:00
Xavier Claessens
73b229ab3a Merge branch 'wip/jehan/android-master-v3' into 'master'
Wip/jehan/android master v3

See merge request GNOME/glib!93
2018-06-11 13:37:55 +00:00
Jehan
866275f56b gio: fix various "warning: unused variable". 2018-06-11 15:13:55 +02:00
Ondrej Holy
15cdcd2e0b glocalfile: Check that parent device is the same when trashing
To be honest, I am not sure why, but in some special environments (e.g.
our CI integration) can happen, that file device number is different from
parent device number. Return "Unable to find or create trash directory for
%s" error from g_local_file_trash() in that case and also set
G_FILE_ATTRIBUTE_ACCESS_CAN_TRASH appropriately.
2018-06-08 13:27:13 +02:00
Will Thompson
0c5f084200
fileinfo: make UNIX_IS_MOUNTPOINT be TRUE for /
The previously implementation considered a file to be a mountpoint if
its parent is on a different device. / is its own parent, so by this
definition it is not a mountpoint.

But / is (generally) listed in fstab, and fstab(5) defines the
directories it contains to be mountpoints. This attribute should follow
that definition (and reasonable expectation): the root directory is a
mountpoint.

So, add a special-case for the case where the file's parent has the same
st_dev and st_ino as the file, which is true only at the root.

Test this attribute at / (only on POSIX), /proc (but only on Linux), and
at many files and directories created by the test suite (which cannot be
mountpoints).
2018-06-07 06:38:23 +01:00
Debarshi Ray
25af5ad1da Revert "Return folder as icon for directories"
This workaround is no longer necessary because it was fixed in
shared-mime-info three years ago:
https://bugs.freedesktop.org/show_bug.cgi?id=89150

This reverts commit 90025254fd.

https://gitlab.gnome.org/GNOME/glib/issues/1073
2018-05-28 17:19:59 +02:00
Christoph Reiter
97c28f7fe1 ci: fix warnings and enable --werror for the mingw build
Fix various warnings regarding unused variables, duplicated
branches etc by adjusting the ifdeffery and some missing casts.

gnulib triggers -Wduplicated-branches in one of the copied files,
disable as that just makes updating the code harder.

The warning indicating missing features are made none fatal through
pragmas. They still show but don't abort the build.

https://bugzilla.gnome.org/show_bug.cgi?id=793729
2018-04-25 17:23:50 +02:00
Friedrich Beckmann
018b997dd2 MacOS: fix content type check vs. mime check to show folder icons
In MacOS the file selection dialog does not show folder icons.
With this fix the folder icons are shown. The bug is described
in:

https://bugzilla.gnome.org/show_bug.cgi?id=788936

This bug fix is only partial, because this fix is only the
last resort when no mime information is available.
2017-12-08 11:10:27 +00:00
Philip Withnall
62dece198b gio: Fix querying of thumbnail attributes other than thumbnail::path
The thumbnail attributes would previously only be set if thumbnail::path
was included in the query — so querying for just thumbnail::is-valid
would return no results.

This fixes the behaviour of
    gio info -a thumbnail::is-valid ./some-file.png
vs
    gio info -a thumbnail ./some-file.png

The first command would previously list nothing. The second would
previously list a thumbnail::path and thumbnail::is-valid.

Signed-off-by: Philip Withnall <withnall@endlessm.com>

https://bugzilla.gnome.org/show_bug.cgi?id=791325
2017-12-07 13:03:18 +00:00
Philip Withnall
9ab0073321 glocalfileinfo: Fix a leak on an error handling path
Spotted by Clang static analysis, thanks to Leslie Zhai.

Signed-off-by: Philip Withnall <withnall@endlessm.com>

https://bugzilla.gnome.org/show_bug.cgi?id=777075
2017-11-17 14:39:55 +00:00
Руслан Ижбулатов
53bd6a359f W32: Add a stat() implementation for private use
This commit adds new W32-only functions to gstdio.c,
and a new header file, gstdioprivate.h.
These functions are:
g_win32_stat_utf8()
g_win32_lstat_utf8()
g_win32_fstat()
and they fill a private structure, GWin32PrivateStat,
which has all the fields that normal stat has, as well as some
extras.

These functions are then used throughout glib and gio to get better
data about the system. Specifically:
* Full, 64-bit size, guaranteed (g_stat() is forced to use 32-bit st_size)
* Full, 64-bit file identifier (st_ino is 0 when normal stat() is used, and still is)
* W32 File attributes (which stat() doesn't report); in particular, this allows
  symlinks to be correctly identified
* Full, 64-bit time, guaranteed (g_stat() uses 32-bit st_*time on 32-bit Windows)
* Allocated file size (as a W32 replacement for the missing st_blocks)

st_mode remains unchanged (thus, no S_ISLNK), so when these are given back to
glib users (via g_stat(), for example, which is now implemented by calling g_win32_stat_utf8),
this field does not contain anything unexpected.

g_lstat() now calls g_win32_lstat_utf8(), which works on symlinks the way it's supposed to.

Also adds the g_win32_readlink_utf8() function, which behaves like readlink()
(including its inability to return 0-terminated strings and inability to say how large
the output buffer should be; these limitations are purely for compatibility with
existing glib code).

Thus, symlink support should now be much better, although far from being complete.

A new W32-only test in gio/tests/file.c highlights the following features:
* allocated size
* 64-bit time
* unique file IDs

https://bugzilla.gnome.org/show_bug.cgi?id=788180
2017-11-01 12:46:38 +00:00
Руслан Ижбулатов
4a77eb16ce Replace all instances of ssize_t with gssize
ssize_t is supported widely, but not universally, so use gssize instead.
Currently only one piece of code actually *needs* this change to be compilable
with MSVC, the rest are mostly in *nix parts of the code, but these are changed
too, for symmetry.

https://bugzilla.gnome.org/show_bug.cgi?id=788180
2017-10-11 12:56:11 +01:00
Philip Withnall
0a10f7375f glocalfileinfo: Use g_strcmp0() for some comparisons which might be NULL
The return value from g_get_user_special_dir() might be NULL. Safest to
use g_strcmp0() uniformly everywhere.

Signed-off-by: Philip Withnall <withnall@endlessm.com>

https://bugzilla.gnome.org/show_bug.cgi?id=661442
2017-10-06 11:17:40 +01:00
Philip Withnall
5cddde1fb2 Consistently save errno immediately after the operation setting it
Prevent the situation where errno is set by function A, then function B
is called (which is typically _(), but could be anything else) and it
overwrites errno, then errno is checked by the caller.

errno is a horrific API, and we need to be careful to save its value as
soon as a function call (which might set it) returns. i.e. Follow the
pattern:
  int errsv, ret;
  ret = some_call_which_might_set_errno ();
  errsv = errno;

  if (ret < 0)
    puts (strerror (errsv));

This patch implements that pattern throughout GLib. There might be a few
places in the test code which still use errno directly. They should be
ported as necessary. It doesn’t modify all the call sites like this:
  if (some_call_which_might_set_errno () && errno == ESOMETHING)
since the refactoring involved is probably more harmful than beneficial
there. It does, however, refactor other call sites regardless of whether
they were originally buggy.

https://bugzilla.gnome.org/show_bug.cgi?id=785577
2017-08-03 10:21:13 +01:00
Sébastien Wilmet
3bf4a720c3 gio/: LGPLv2+ -> LGPLv2.1+
Sub-directories inside gio/ already processed in a previous commit:
- fam/
- gdbus-2.0/ (which contains only codegen/)
- gvdb/
- inotify/
- tests/
- win32/
- xdgmime/

Other sub-directories inside gio/:
- completion/: no license headers
- kqueue/: not LGPL, BSD-style license

https://bugzilla.gnome.org/show_bug.cgi?id=776504
2017-05-29 19:53:34 +02:00
TingPing
3953d85a92 Implement GContentType on OSX
This is an implementation of most of GContentType using the OS X
UTType APIs.

Missing at this point is an implementation of
g_content_types_get_registered() and g_content_type_guess_for_tree().

https://bugzilla.gnome.org/show_bug.cgi?id=734946
2017-03-13 10:32:02 -04:00
Piotr Drąg
10c490cdfe Use Unicode in translatable strings
See https://developer.gnome.org/hig/stable/typography.html

https://bugzilla.gnome.org/show_bug.cgi?id=772221
2016-10-12 21:30:42 +02:00
Ryan Lortie
aa16359986 Stop supporting non-POSIX getpwuid_r, getgrgid_r
Bug 13403 introduced support for the non-POSIX variants of these APIs
found on a system called "DG/UX".  Meanwhile, the complicated checks
here are breaking cross-builds on systems that we actually care about.

Remove the complicated checks and replace them with AC_CHECK_FUNCS.
Remove the resulting dead code from a couple of .c files.

https://bugzilla.gnome.org/show_bug.cgi?id=756475
2015-12-16 07:47:53 -05:00
Ryan Lortie
202a9c3497 GLocalFile: return text/plain for empty files
Previously, GLib returned text/plain for empty files.

This is important because people may want to open empty (eg:
just-created) text files with the text editor.

An unintended side-effect of b6fc1df022
caused GLib to start returning application/octet-stream instead of
text/plain for these files.

This commit is essentially a revert of that commit, with a different
solution: we move the special-case up a bit in the function and
hard-code it to text/plain.

This change does not exactly maintain the old behaviour: previously, a
"fast" lookup would have returned application/octet-stream on an empty
file and now it will return text/plain.  I consider this to be an
improvement (since we're returning better data) and don't expect it to
cause problems.

https://bugzilla.gnome.org/show_bug.cgi?id=755795
2015-09-29 12:29:10 -04:00
Ryan Lortie
b6fc1df022 GLocalFileInfo: don't content-sniff zero-length files
This will prevent attempting to read from some files that appear normal but are
really device-like, such as those in /proc and /sys.

If we can't stat() the file then don't bother attempting to sniff, either.

https://bugzilla.gnome.org/show_bug.cgi?id=708525
2015-08-21 01:00:49 -04:00
Ting-Wei Lan
8f662e7259 glocalfileinfo: Support file creation time on FreeBSD and NetBSD
FreeBSD and NetBSD have field st_birthtim and st_birthtime in struct stat,
respectively, which can be used to get file creation time on supported file
systems such as UFS2 and tmpfs.

https://bugzilla.gnome.org/show_bug.cgi?id=749492
2015-08-21 00:52:50 -04:00
TingPing
aa4e2d4dc3 Fix GContentType usage
https://bugzilla.gnome.org/show_bug.cgi?id=734946
2015-01-04 22:09:37 -05:00
Colin Walters
7ea4bf3f30 glocalfileinfo: Suppress static analysis return value warning
Just ignore the return value, since we're checking contents != NULL.

https://bugzilla.gnome.org/show_bug.cgi?id=733576
2014-07-23 07:43:41 -04:00
Matthias Clasen
90025254fd Return folder as icon for directories
'folder' is the name of the folder icon in the incon naming spec,
and the Adwaita icon theme doesn't include an inode-directory icon.
This fixes folders appearing as generic file in the file chooser.

https://bugzilla.gnome.org/show_bug.cgi?id=731996
2014-06-22 14:34:27 -04:00
Lars Uebernickel
8c78fb827b glocalfileinfo: don't special-case directory icons
As of e6af432, g_content_type_get_symbolic_icon() returns non-symbolic
fallbacks. Thus, we can't append another symbolic icon to the fallbacks.

The special case was a bit of a hack anyway. It was only applied to
themed icons and there was no generic fallback for mime types that are
not folders.

https://bugzilla.gnome.org/show_bug.cgi?id=726046
2014-03-11 11:47:57 +01:00
Daniel Mustieles
078dbda148 Updated FSF's address 2014-01-31 14:31:55 +01:00
Marc-André Lureau
2873b3c20c win32: silence build warning
glocalfileinfo.c: In function '_g_local_file_info_get':
glocalfileinfo.c:1955:11: warning: passing argument 3 of
'get_thumbnail_attributes' from incompatible pointer type [enabled by
default]
           get_thumbnail_attributes (path, info, &statbuf);
           ^
glocalfileinfo.c:1285:1: note: expected 'const struct GStatBuf *' but
argument is of type 'struct _stati64 *'
 get_thumbnail_attributes (const char     *path,

https://bugzilla.gnome.org/show_bug.cgi?id=711547
2014-01-19 08:25:33 -05:00
Marc-André Lureau
d3c729bf0e gio: fix small memory leak on local xattr
g_strconcat() allocates memory, it needs to be freed.

==10653== 1,400 bytes in 50 blocks are definitely lost in loss record
1,838 of 1,851
==10653==    at 0x4A0645D: malloc (in
/usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==10653==    by 0x54ACB22: g_malloc (gmem.c:102)
==10653==    by 0x54ACE4D: g_malloc_n (gmem.c:343)
==10653==    by 0x54C8463: g_strconcat (gstrfuncs.c:589)
==10653==    by 0x4D6ED38: get_xattrs_from_fd (glocalfileinfo.c:660)
==10653==    by 0x4D71622:
_g_local_file_info_get_from_fd (glocalfileinfo.c:2028)
==10653==    by 0x4D731A0:
g_local_file_input_stream_query_info (glocalfileinputstream.c:356)
==10653==    by 0x4C996D8:
g_file_input_stream_query_info (gfileinputstream.c:148)
==10653==    by 0x4C863F6: file_copy_fallback (gfile.c:3120)
==10653==    by 0x4C86DD2: g_file_copy (gfile.c:3398)

https://bugzilla.gnome.org/show_bug.cgi?id=722357
2014-01-16 20:27:26 -05:00
Dan Winship
158dde0507 Replace #ifdef HAVE_UNISTD_H checks with #ifdef G_OS_UNIX
In Windows development environments that have it, <unistd.h> is mostly
just a wrapper around several other native headers (in particular,
<io.h>, which contains read(), close(), etc, and <process.h>, which
contains getpid()). But given that some Windows dev environments don't
have <unistd.h>, everything that uses those functions on Windows
already needed to include the correct Windows header as well, and so
there is never any point to including <unistd.h> on Windows.

Also, remove some <unistd.h> includes (and a few others) that were
unnecessary even on unix.

https://bugzilla.gnome.org/show_bug.cgi?id=710519
2013-11-20 09:25:39 -05:00
Dan Winship
3981cddbf8 Require POSIX.1 (1990) compliance on unix
Assume unix platforms support the original POSIX.1 standard.
Specifically, assume that if G_OS_UNIX, then we have chown(),
getcwd(), getgrgid(), getpwuid(), link(), <grp.h>, <pwd.h>,
<sys/types.h>, <sys/uio.h>, <sys/wait.h>, and <unistd.h>.

Additionally, since all versions of Windows that we care about also
have <sys/types.h>, we can remove HAVE_SYS_TYPES_H checks everywhere.

Also remove one include of <sys/times.h>, and the corresponding
configure check, since the include is not currently needed (and may
always have just been a typo for <sys/time.h>).

https://bugzilla.gnome.org/show_bug.cgi?id=710519
2013-11-20 09:17:42 -05:00
Ryan Lortie
fe7069749f file-info: Add a G_FILE_ATTRIBUTE_THUMBNAIL_IS_VALID attribute
This indicates whether the thumbnail (given by G_FILE_ATTRIBUTE_THUMBNAIL_PATH)
is valid — i.e. to represent the file in its current state. If
G_FILE_ATTRIBUTE_THUMBNAIL_IS_VALID is FALSE (for a normal _or_ failed
thumbnail) it means the file has changed since the thumbnail was generated, and
the thumbnail is out of date.

Part of checking thumbnail validity (by the spec) involves parsing
headers out of the thumbnail .png so we include some (small) code to do
that in a separate file.  We will likely want to copy this code to gvfs
to do the same for GVfsFile.

Heavily based on a patch from Philip Withnall <philip.withnall@collabora.co.uk>
who suggested the feature and designed the API.

https://bugzilla.gnome.org/show_bug.cgi?id=709898
2013-10-23 11:56:28 -04:00
Sebastian Dröge
bcbaf1bef0 Fix compilation on Android with the bionic C library
https://bugzilla.gnome.org/show_bug.cgi?id=689223
2013-04-16 13:24:26 +02:00
Ryan Lortie
3902006a5b glocalfileinfo: Stop using PATH_MAX for .hidden
We were using PATH_MAX to size a static array for reading lines from
the .hidden file.  Some platforms (Hurd) don't declare a PATH_MAX.

Switch to using g_file_get_contents() and g_str_split('\n') instead.

Also take the time to clean up a bit with a switch to using a 'set mode'
GHashTable (since this code was originally written before we had those).

This patch is largely based on a patch from Emilio Pozuelo Monfort (who
also reported the bug).

https://bugzilla.gnome.org/show_bug.cgi?id=695147
2013-03-13 09:41:03 -04:00
Colin Walters
f398bec5bc Add g_close(), use it
There are two benefits to this:

1) We can centralize any operating system specific knowledge of
   close-vs-EINTR handling.  For example, while on Linux we should never
   retry, if someone cared enough later about HP-UX, they could come by
   and change this one spot.
2) For places that do care about the return value and want to provide
   the caller with a GError, this function makes it convenient to do so.

Note that gspawn.c had an incorrect EINTR loop-retry around close().

https://bugzilla.gnome.org/show_bug.cgi?id=682819
2013-01-29 09:46:04 -05:00
Colin Walters
8fe8dcda1e glocalfile: Only check for .hidden files if standard::is-hidden is requested
For OSTree, I use Gio and also really care about performance.  It's
disturbing to see open('.hidden') all over my straces and such.  At
the moment I have an explicit set of things to query, as opposed to
"standard::*", since even before this that also implies an lstat() of
the parent directory.

This matches up with what we do for all the other attributes.

See https://bugzilla.gnome.org/show_bug.cgi?id=587806
https://bugzilla.gnome.org/show_bug.cgi?id=691558
2013-01-11 11:14:12 -05:00
David Zeuthen
d77948eadf fileinfo: mark lost+found/ root directory as hidden
This was reported in bug 689800.

https://bugzilla.gnome.org/show_bug.cgi?id=689800

Signed-off-by: David Zeuthen <zeuthen@gmail.com>
2012-12-06 16:21:58 -05:00
Timothy Arceri
510ba9b4ef Support for .hidden files
https://bugzilla.gnome.org/show_bug.cgi?id=587806
2012-12-06 16:18:19 -05:00
Kalev Lember
ce586ba991 Unconditionally use g_content_type_get_symbolic_icon()
... and g_content_type_get_generic_icon_name(). The new functions are
implemented for Win32 since commit dace477c, so we no longer need to
guard them with G_OS_UNIX.
2012-11-28 23:25:52 +01:00
Sebastian Dröge
c9affa778e Revert "Fix compilation on Android with the bionic C library"
This reverts commit cb0ed84d40.

It wasn't meant to be pushed yet.
2012-11-28 16:55:12 +01:00
Sebastian Dröge
cb0ed84d40 Fix compilation on Android with the bionic C library 2012-11-28 16:32:48 +01:00
Debarshi Ray
51d0830102 file-info: Don't leak the filename if thumbnailing failed
Fixes: https://bugzilla.gnome.org/686895
2012-11-20 00:24:52 +01:00
Cosimo Cecchi
d681b581ff file-info: catch thumbnail files in large directory as well
When building the file attribute table info for local files, use
thumbnail paths in $XDG_CACHE_DIR/thumbnails/large in addition to
$XDG_CACHE_DIR/thumbnails/normal.

Failing to do this would cause an application that creates large
thumbnails by default to never find any value for
G_FILE_ATTRIBUTE_THUMBNAIL_PATH, with no
G_FILE_ATTRIBUTE_THUMBNAILING_FAILED set, which might cause the
application to either think thumbnailing is still in progress, or
blindly requeue thumbnail operations in a loop.

Large thumbnails are generally preferred, so we now default to the path
of a large thumbnail (in case both are present).

https://bugzilla.gnome.org/show_bug.cgi?id=686895
2012-10-26 10:48:09 -04:00
Chun-wei Fan
3dd65859a7 gio: Fix build on Windows
The newly-introduced functions, g_content_type_get_symbolic_icon() and
g_content_type_get_generic_icon_name() don't seem to be for Windows, at
least for now.  So filter them out from gio.symbols on Windows.

Also, glocalfileinfo.c calls g_content_type_get_symbolic_icon() in
get_icon(), so only build that code when on Unix, for the time being.

https://bugzilla.gnome.org/show_bug.cgi?id=684278
2012-09-19 10:32:05 +08:00
William Jon McCann
40b4fae42e Add ability to get symbolic icon for content type
https://bugzilla.gnome.org/show_bug.cgi?id=682101
2012-08-30 11:04:43 -04:00
William Jon McCann
a15a071f35 Add symbolic icon support to gfileinfo
https://bugzilla.gnome.org/show_bug.cgi?id=682101
2012-08-30 11:04:43 -04:00
Colin Walters
6e64ba58b9 Switch all open() calls to use g_open()
Because it now handles EINTR.  And we should do so.  While most people
use Linux, which tries very hard to avoid propagating EINTR back up
into userspace, it can still happen.

https://bugzilla.gnome.org/show_bug.cgi?id=682833
2012-08-28 13:56:59 -04:00
Matthias Clasen
04af05fd1f Drop unneeded includes 2012-07-03 21:14:41 -04:00
William Jon McCann
ee19ddcb14 Update to use XDG cache home for thumbnails
https://bugzilla.gnome.org/show_bug.cgi?id=675168
2012-06-13 14:46:13 -04:00