| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
| |
Formerly known as "Atom Shell Archive Format" [1].
Unofficial MIME type also used by NPM "file-type" library [2].
[1] https://github.com/electron/asar
[2] https://github.com/sindresorhus/file-type/blob/main/core.d.ts#L274
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
GLib takes priorities >= 80 to mean a short-circuit to a magic
match, overriding any conflicting matching globs. More generally, high
priority magic should only be used if it’s really unlikely that the
magic will match anything except the given MIME type. That’s evidently
not true here.
Some Python files which import Qt modules can look like valid QML
files, matching the magic but not the glob.
Lower the priority of the magic so that it doesn’t trigger the
short-circuit code in GLib. This means the name match is guessed as the
right one instead of the magic match.
See
https://gitlab.gnome.org/GNOME/glib/-/blob/6ee71750a532d8de5eaed5cf12e8718738ddd5c6/gio/gcontenttype.c#L840
and https://gitlab.gnome.org/GNOME/glib/-/issues/2599.
Tested with `XDG_DATA_DIRS=$buildroot/share gio info ./not-qml.py`.
Signed-off-by: Philip Withnall <pwithnall@endlessos.org>
Helps: #84
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Make `image/fits` an alias, matching its designation of a subset of
the FITS standard, as per https://datatracker.ietf.org/doc/html/rfc4047#section-5 -
"The general nature of the full FITS standard requires the use of the
media type "application/fits". Nevertheless, the principal intent
for a great many FITS files is to convey a single data array in the
PHDU, and such arrays are very often 2-dimensional images. Several
common image viewing applications already display single-HDU FITS
files, and the prototypes for virtual observatory projects specify
that data provided by web services be conveyed by the data array in
the PHDU. These uses justify the registration of a second media
type, namely "image/fits", for files which use the subset of the
standard described by the original FITS standard paper."
The existing alias `image/x-fits` was common before the standardization
of `application/fits` and `image/fits`, so it should stay.
Current standards and historical contexts can be found here:
- https://fits.gsfc.nasa.gov/fits_standard.html
- https://listmgr.nrao.edu/pipermail/fitsmime/2002-December/thread.html
- https://www.ucolick.org/~sla/fits/mime/comments.html
Add missing eight-dot-three-style `*.fit` and `*.fts` globs for legacy files,
matching the list supported by `libvips`, i.e.,
https://github.com/libvips/libvips/blob/757d03100566368e5d06c7355e73c1215dd34968/libvips/foreign/fits.c#L132
And as discussed in the FITS standard definition (`fits_standard40aa-le.pdf`) -
"In the absence of other information it is reasonably safe to
presume that a file name ending in ‘.fits’ is intended to be
a FITS file. Nevertheless, there are other commonly used ex-
tensions; e.g., ‘.fit’, ‘.fts’, and many others not suitable for
listing in a media type registration."
|
| |
|
|
|
|
|
|
|
| |
FlashForge's FlashPrint software uses gcode with a binary header
containing extra data to save sliced models. Seeing as they have a known
magic and that we might want to use them outside of FlashPrint itself,
add a mime-type.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This MIME type name is used in Qt Creator [1] and KDevelop [2]. A
different name - text/x-objcpp-src - is used in KSyntaxHighlighting [3].
The name text/x-objc++src is consistent with the existing MIME type
names text/x-c++src and text/x-objcsrc. So KSyntaxHighlighting's name
will have to be replaced later.
The Objective-C Wikipedia article and the KSyntaxHighlighting's data
file [3] claim that *.M is one of the file extensions of Objective-C++
source files. But since macOS's file system is case-insensitive by
default, *.mm is used almost universally. Thus there is no need to
complicate the MIME type database with case-sensitive attributes.
Make text/x-troff-mm a subclass of text/troff to reuse its magic
element. As far as I can tell, these two file formats share the same
comment syntax, which is encoded in text/troff's magic.
Add Objective-C, Objective-C++, Troff, Troff MM and C tests to verify
correct detection of the new MIME type and the absence of regressions.
[1] https://code.qt.io/cgit/qt-creator/qt-creator.git/tree/src/plugins/cppeditor/CppEditor.json.in?id=2cc28dea65846153d16c213283ffac7af530590d#n100
[2] https://invent.kde.org/kdevelop/kdevelop/-/blob/ebd207cca2387b5734a591803381880505e39a30/plugins/astyle/astyle_plugin.cpp#L284
[3] https://invent.kde.org/frameworks/syntax-highlighting/-/blob/0b2c0dc5368de519c4e9517a4dec69017df4cd6c/data/syntax/objectivecpp.xml#L3
|
|
|
|
| |
https://gitlab.freedesktop.org/xdg/shared-mime-info/-/merge_requests/151#note_1098689
|
| |
|
|
|
|
| |
https://nokiatech.github.io/heif/technical.html
|
|
|
|
|
|
|
| |
De-facto standard mentioned by Apple in [1] and [2].
[1] https://developer.apple.com/library/archive/documentation/UserExperience/Conceptual/PassKit_PG/DistributingPasses.html
[2] https://developer.apple.com/documentation/walletpasses/send_an_updated_pass
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
b969e1c74e7c905a96c1590f5350a264dbaa83ae added a reference to this file
and a command to generate it into tests/mime-detection/list, but didn't
commit the generated file itself. Now when test-mime fails for some
other reason (e.g. after adding a failing test case), the following
error message is printed:
Could not open /path/to/shared-mime-info/tests/mime-detection/test-not-dtshd.txt
Presumably when the test file does not exist, the intended 'data' test
does not run. The testing system should be improved to report such
nonexistent file issues as test failures.
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Look for `<key>_SPCommandLineArguments</key>` in addition to plist root
element to distinguish from other property list XML files.
Our offset specifiers aren't that fancy, so this can trigger on files
that have the above string ´_before_ `<plist version="1.0"`, but those
should be quite pathological cases.
Start offset 34 is length of `<plist version="1.0"><array><dict>`, 384
is arbitrary guesswork.
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/169
|
|
|
|
| |
https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/169
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Matching just 0x64582025 within the first 18725 bytes of any file is
prone to false positives.
The priority of the magic match here being bumped to 60 seems to
indicate intent to make it take precedence over the parent
`audio/vnd.dts`. But we should take the parent magic into account here
as it does not "inherit" with `sub-class-of`.
As the comment for this match says "16bits HD be", accept only "16bits
core be" as the parent at least for now, not all possible parent
magics -- it would not seem consistent to have it in "14bits"
variants, or the "16bits core le" one. The test case we have for this
is one matching "16bits core be".
Refs https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/166
Refs https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/167
|
|
|
|
| |
https://www.iana.org/assignments/media-types/model/gltf+json
|
|
|
|
| |
https://www.iana.org/assignments/media-types/model/gltf-binary
|
|
|
|
|
|
| |
For GRUB modules.
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/165
|
|
|
|
| |
https://wiki.multimedia.cx/index.php/Protracker_Module#File_Format
|
| |
|
|
|
|
|
|
|
|
| |
I failed to include the test files in the list in
238f20f82d485dafa16904c5bd1b07c8034ddacd, and unfortunately can no
longer remember their origin. I likely took an existing one from the
test suite and converted from it, but the details and tools used escape
me.
|
|
|
|
|
|
| |
My editor keeps adding it (which is nice), but I need to remove it
manually all the time not to include it as unrelated whitespace in other
commits (which is not). Just add it.
|
|
|
|
|
|
| |
"BEGIN/END CA CERTIFICATE" seems to be a thing, although not that
common. "BEGIN/END TRUSTED CERTIFICATE" is an OpenSSL marker for trusted
certificates, see x509(1).
|
| |
|
| |
|
| |
|
|
|
|
| |
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/159
|
|
|
|
|
|
|
| |
JSON Schema files are distinguished from other JSON files by a "$schema"
property which is typically at the beginning of the file.
https://datatracker.ietf.org/doc/html/draft-bhutton-json-schema-00
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
Stub files are files containing type hints, they are
written in normal Python 3 syntax, but
generally leaving out runtime logic like variable
initializers, function bodies, and default arguments.
https://www.python.org/dev/peps/pep-0484/#stub-files
https://mypy.readthedocs.io/en/stable/stubs.html#stub-file-syntax
|
|
|
|
| |
https://en.wikipedia.org/wiki/X_PixMap#XPM2
|
| |
|
|
|
|
|
|
| |
https://toml.io/en/v1.0.0#mime-type
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/156
|
|
|
|
| |
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/155
|
|
|
|
| |
These are the first files of a multivolume archive.
|
|
|
|
| |
https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/124#note_788636
|
|
|
|
| |
https://en.wikipedia.org/wiki/Open_Virtualization_Format
|
|
|
|
| |
https://wiki.qemu.org/Features/QED
|
|
|
|
| |
https://en.wikipedia.org/wiki/VirtualBox#VirtualBox_Disk_Image
|
|
|
|
| |
https://en.wikipedia.org/wiki/VMDK
|
|
|
|
| |
https://en.wikipedia.org/wiki/VHD_(file_format)
|
|
|
|
|
| |
Numbers 3.5 and Pages 5.5 appear to place an Index/Document.iwa, not
index.xml there.
|
|
|
|
|
|
| |
Dart is a programming language by Google.
https://dart.dev/
|
|
|
|
|
|
| |
Fixed by https://gitlab.freedesktop.org/xdg/xdgmime/-/merge_requests/3
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/127
|
|
|
|
| |
Closes https://gitlab.freedesktop.org/xdg/shared-mime-info/-/issues/46
|