From d10324eac4429ff3d7d38ad24a19210699229e07 Mon Sep 17 00:00:00 2001 From: Alfred Wingate Date: Wed, 11 Dec 2024 06:34:36 +0200 Subject: [PATCH 1/4] Use application/gzip as the preferred mimetype for gzip Signed-off-by: Alfred Wingate --- patoolib/__init__.py | 1 - patoolib/mime.py | 16 +++++++++++++--- tests/test_mime.py | 4 ++-- 3 files changed, 15 insertions(+), 6 deletions(-) diff --git a/patoolib/__init__.py b/patoolib/__init__.py index d665a28a..2247b615 100644 --- a/patoolib/__init__.py +++ b/patoolib/__init__.py @@ -127,7 +127,6 @@ 'application/x-cpio': 'cpio', 'application/x-debian-package': 'deb', 'application/x-dms': 'dms', - 'application/x-gzip': 'gzip', 'application/x-iso9660-image': 'iso', 'application/x-lz4': 'lz4', 'application/x-lzop': 'lzop', diff --git a/patoolib/mime.py b/patoolib/mime.py index 12405ada..c9d8894b 100644 --- a/patoolib/mime.py +++ b/patoolib/mime.py @@ -135,9 +135,11 @@ def guess_mime(filename: str) -> tuple[str | None, str | None]: Mime2Encoding: dict[str, str] = dict( [(_val, _key) for _key, _val in Encoding2Mime.items()] ) -# libmagic before version 5.14 identified .gz files as application/x-gzip -Mime2Encoding['application/x-gzip'] = 'gzip' +LegacyMimeType: dict[str, str] = { + # libmagic before version 5.14 identified .gz files as application/x-gzip + 'application/x-gzip': "application/gzip", +} def guess_mime_mimedb(filename: str) -> tuple[str | None, str | None]: """Guess MIME type from given filename. @@ -192,6 +194,10 @@ def guess_mime_file(filename: str) -> tuple[str | None, str | None]: except (OSError, subprocess.CalledProcessError) as err: log_warning(f"error executing {cmd}: {err}") mime2 = None + + if mime2 in LegacyMimeType: + mime2 = LegacyMimeType[mime2] + # Some file(1) implementations return an empty or unknown mime type # when the uncompressor program is not installed, other # implementation return the original file type. @@ -227,6 +233,10 @@ def guess_mime_file_mime( except OSError as err: # ignore errors, as file(1) is only a fallback log_warning(f"error executing {cmd}: {err}") + + if mime in LegacyMimeType: + mime = LegacyMimeType[mime] + if mime not in ArchiveMimetypes: mime, encoding = None, None return mime, encoding @@ -253,7 +263,7 @@ def get_file_mime_encoding(parts: Sequence[str]) -> str | None: "cpio archive": "application/x-cpio", "ASCII cpio archive": "application/x-cpio", "Debian binary package": "application/x-debian-package", - "gzip compressed data": "application/x-gzip", + "gzip compressed data": "application/gzip", "LZMA compressed data": "application/x-lzma", "LRZIP compressed data": "application/x-lrzip", "lzop compressed data": "application/x-lzop", diff --git a/tests/test_mime.py b/tests/test_mime.py index 40e73edf..3f292dfc 100644 --- a/tests/test_mime.py +++ b/tests/test_mime.py @@ -81,8 +81,8 @@ def test_mime_file(self): self.mime_test_file("t.cpio.foo", "application/x-cpio") self.mime_test_file("t.deb", "application/x-debian-package") self.mime_test_file("t.deb.foo", "application/x-debian-package") - self.mime_test_file("t.txt.gz", ("application/gzip", "application/x-gzip")) - self.mime_test_file("t.txt.gz.foo", ("application/gzip", "application/x-gzip")) + self.mime_test_file("t.txt.gz", "application/gzip") + self.mime_test_file("t.txt.gz.foo", "application/gzip") self.mime_test_file("t.jar", "application/zip") self.mime_test_file("t.jar.foo", "application/zip") self.mime_test_file("t.txt.lzma", "application/x-lzma") From e7501d1c7805696ff5b2ecc779f7a56ab2425c3f Mon Sep 17 00:00:00 2001 From: Alfred Wingate Date: Wed, 11 Dec 2024 06:37:54 +0200 Subject: [PATCH 2/4] Change rar mime type to application/vnd.rar https://github.com/file/file/commit/d46a1f3dbbf58eb510c1779b8bdcc59d5ee24ab9 https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1068700 Signed-off-by: Alfred Wingate --- patoolib/__init__.py | 2 +- patoolib/mime.py | 9 ++++++--- tests/test_mime.py | 16 ++++++++-------- 3 files changed, 15 insertions(+), 12 deletions(-) diff --git a/patoolib/__init__.py b/patoolib/__init__.py index 2247b615..8b13181c 100644 --- a/patoolib/__init__.py +++ b/patoolib/__init__.py @@ -136,7 +136,6 @@ 'application/x-lrzip': 'lrzip', 'application/x-lzh': 'lzh', 'application/x-ms-wim': 'wim', - 'application/x-rar': 'rar', 'application/x-redhat-package-manager': 'rpm', 'application/x-rpm': 'rpm', 'application/x-rzip': 'rzip', @@ -147,6 +146,7 @@ 'application/x-xz': 'xz', 'application/x-zip-compressed': 'zip', 'application/x-zoo': 'zoo', + 'application/vnd.rar': 'rar', 'application/zip': 'zip', 'application/zpaq': 'zpaq', "application/zstd": "zstd", diff --git a/patoolib/mime.py b/patoolib/mime.py index c9d8894b..8de9b180 100644 --- a/patoolib/mime.py +++ b/patoolib/mime.py @@ -54,8 +54,8 @@ def add_mimedb_data(mimedb: mimetypes.MimeTypes) -> None: add_mimetype(mimedb, 'application/x-lzma', '.lzma') add_mimetype(mimedb, 'application/x-xz', '.xz') add_mimetype(mimedb, 'application/java-archive', '.jar') - add_mimetype(mimedb, 'application/x-rar', '.rar') - add_mimetype(mimedb, 'application/x-rar', '.cbr') + add_mimetype(mimedb, 'application/vnd.rar', '.rar') + add_mimetype(mimedb, 'application/vnd.rar', '.cbr') add_mimetype(mimedb, 'application/x-7z-compressed', '.7z') add_mimetype(mimedb, 'application/x-7z-compressed', '.cb7') add_mimetype(mimedb, 'application/x-cab', '.cab') @@ -139,8 +139,11 @@ def guess_mime(filename: str) -> tuple[str | None, str | None]: LegacyMimeType: dict[str, str] = { # libmagic before version 5.14 identified .gz files as application/x-gzip 'application/x-gzip': "application/gzip", + # libmagic before version 5.46 identified .rar files as application/x-rar + 'application/x-rar': "application/vnd.rar", } + def guess_mime_mimedb(filename: str) -> tuple[str | None, str | None]: """Guess MIME type from given filename. @return: tuple (mime, encoding) @@ -268,7 +271,7 @@ def get_file_mime_encoding(parts: Sequence[str]) -> str | None: "LRZIP compressed data": "application/x-lrzip", "lzop compressed data": "application/x-lzop", "Microsoft Cabinet archive data": "application/vnd.ms-cab-compressed", - "RAR archive data": "application/x-rar", + "RAR archive data": "application/vnd.rar", "RPM ": "application/x-redhat-package-manager", "POSIX tar archive": "application/x-tar", "xz compressed data": "application/x-xz", diff --git a/tests/test_mime.py b/tests/test_mime.py index 3f292dfc..8f3ecb8c 100644 --- a/tests/test_mime.py +++ b/tests/test_mime.py @@ -91,10 +91,10 @@ def test_mime_file(self): self.mime_test_file("t.txt.lz.foo", "application/x-lzip") self.mime_test_file("t.txt.lzo", "application/x-lzop") self.mime_test_file("t.txt.lzo.foo", "application/x-lzop") - self.mime_test_file("t.rar", "application/x-rar") - self.mime_test_file("t.rar.foo", "application/x-rar") - self.mime_test_file("t.cbr", "application/x-rar") - self.mime_test_file("t.cbr.foo", "application/x-rar") + self.mime_test_file("t.rar", "application/vnd.rar") + self.mime_test_file("t.rar.foo", "application/vnd.rar") + self.mime_test_file("t.cbr", "application/vnd.rar") + self.mime_test_file("t.cbr.foo", "application/vnd.rar") self.mime_test_file("t.rpm", "application/x-rpm") self.mime_test_file("t.rpm.foo", "application/x-rpm") self.mime_test_file("t.tar", "application/x-tar") @@ -197,8 +197,8 @@ def test_nested_gzip(self): """Test mime detection of archives with double compression""" # We won't extract this with rar, as it doesn't support archives wrapped in gzip # compression, but we will recognize the archive as a gzip-wrapped rar-file - self.mime_test_file("t.rar.gz", "application/x-rar", "gzip") - self.mime_test_file("t.rar.gz.foo", "application/x-rar", "gzip") + self.mime_test_file("t.rar.gz", "application/vnd.rar", "gzip") + self.mime_test_file("t.rar.gz.foo", "application/vnd.rar", "gzip") @needs_program('file') @needs_program('gzip') @@ -237,7 +237,7 @@ def test_mime_mimedb(self): self.mime_test_mimedb("t .bz2", "application/x-bzip2") self.mime_test_mimedb("t .bz3", "application/x-bzip3") self.mime_test_mimedb("t.cab", "application/x-cab") - self.mime_test_mimedb("t.cbr", ("application/rar", "application/x-rar")) + self.mime_test_mimedb("t.cbr", ("application/rar", "application/vnd.rar")) self.mime_test_mimedb("t.cpio", "application/x-cpio") self.mime_test_mimedb("t.deb", "application/x-debian-package") self.mime_test_mimedb("t.gz", "application/gzip") @@ -247,7 +247,7 @@ def test_mime_mimedb(self): self.mime_test_mimedb("t.txt.lz", "application/x-lzip") self.mime_test_mimedb("t.txt.lz4", "application/x-lz4") self.mime_test_mimedb("t.lzo", "application/x-lzop") - self.mime_test_mimedb("t.rar", ("application/rar", "application/x-rar")) + self.mime_test_mimedb("t.rar", ("application/rar", "application/vnd.rar")) self.mime_test_mimedb( "t.rpm", ("application/x-redhat-package-manager", "application/x-rpm") ) From 85fafd16ec01a7eb793e04011617bb47211d446a Mon Sep 17 00:00:00 2001 From: Alfred Wingate Date: Wed, 11 Dec 2024 06:40:41 +0200 Subject: [PATCH 3/4] Remove references to application/rar application/vnd.rar is the IANA assigned mime type for rar. libmagic has never refered to application/rar either, so there is no use for backwards compatibility. https://www.iana.org/assignments/media-types/application/vnd.rar Signed-off-by: Alfred Wingate --- patoolib/__init__.py | 1 - tests/test_mime.py | 4 ++-- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/patoolib/__init__.py b/patoolib/__init__.py index 8b13181c..51f8b4eb 100644 --- a/patoolib/__init__.py +++ b/patoolib/__init__.py @@ -110,7 +110,6 @@ 'application/jar': 'zip', # reported on older systems such as ubuntu 14.04 'application/java-archive': 'zip', 'application/vnd.android.package-archive': 'zip', - 'application/rar': 'rar', 'application/vnd.ms-cab-compressed': 'cab', 'application/x-7z-compressed': '7z', 'application/x-ace': 'ace', diff --git a/tests/test_mime.py b/tests/test_mime.py index 8f3ecb8c..287d8208 100644 --- a/tests/test_mime.py +++ b/tests/test_mime.py @@ -237,7 +237,7 @@ def test_mime_mimedb(self): self.mime_test_mimedb("t .bz2", "application/x-bzip2") self.mime_test_mimedb("t .bz3", "application/x-bzip3") self.mime_test_mimedb("t.cab", "application/x-cab") - self.mime_test_mimedb("t.cbr", ("application/rar", "application/vnd.rar")) + self.mime_test_mimedb("t.cbr", "application/vnd.rar") self.mime_test_mimedb("t.cpio", "application/x-cpio") self.mime_test_mimedb("t.deb", "application/x-debian-package") self.mime_test_mimedb("t.gz", "application/gzip") @@ -247,7 +247,7 @@ def test_mime_mimedb(self): self.mime_test_mimedb("t.txt.lz", "application/x-lzip") self.mime_test_mimedb("t.txt.lz4", "application/x-lz4") self.mime_test_mimedb("t.lzo", "application/x-lzop") - self.mime_test_mimedb("t.rar", ("application/rar", "application/vnd.rar")) + self.mime_test_mimedb("t.rar", "application/vnd.rar") self.mime_test_mimedb( "t.rpm", ("application/x-redhat-package-manager", "application/x-rpm") ) From bab80a04a72941b3c53e5fce1f96f5a2fc531280 Mon Sep 17 00:00:00 2001 From: Alfred Wingate Date: Wed, 11 Dec 2024 07:04:19 +0200 Subject: [PATCH 4/4] Remove references to application/x-redhat-package-manager libmagic has never referred to x-redhat-package-manager and has had x-rpm since 3.30. Red Hat themselves use x-rpm in their mailcap. https://pagure.io/mailcap/blob/master/f/mime.types Signed-off-by: Alfred Wingate --- patoolib/__init__.py | 1 - patoolib/mime.py | 2 +- tests/test_mime.py | 4 +--- 3 files changed, 2 insertions(+), 5 deletions(-) diff --git a/patoolib/__init__.py b/patoolib/__init__.py index 51f8b4eb..9d8bc849 100644 --- a/patoolib/__init__.py +++ b/patoolib/__init__.py @@ -135,7 +135,6 @@ 'application/x-lrzip': 'lrzip', 'application/x-lzh': 'lzh', 'application/x-ms-wim': 'wim', - 'application/x-redhat-package-manager': 'rpm', 'application/x-rpm': 'rpm', 'application/x-rzip': 'rzip', 'application/x-shar': 'shar', diff --git a/patoolib/mime.py b/patoolib/mime.py index 8de9b180..ffad9cd0 100644 --- a/patoolib/mime.py +++ b/patoolib/mime.py @@ -272,7 +272,7 @@ def get_file_mime_encoding(parts: Sequence[str]) -> str | None: "lzop compressed data": "application/x-lzop", "Microsoft Cabinet archive data": "application/vnd.ms-cab-compressed", "RAR archive data": "application/vnd.rar", - "RPM ": "application/x-redhat-package-manager", + "RPM ": "application/x-rpm", "POSIX tar archive": "application/x-tar", "xz compressed data": "application/x-xz", "Zip archive data": "application/zip", diff --git a/tests/test_mime.py b/tests/test_mime.py index 287d8208..a337f87a 100644 --- a/tests/test_mime.py +++ b/tests/test_mime.py @@ -248,9 +248,7 @@ def test_mime_mimedb(self): self.mime_test_mimedb("t.txt.lz4", "application/x-lz4") self.mime_test_mimedb("t.lzo", "application/x-lzop") self.mime_test_mimedb("t.rar", "application/vnd.rar") - self.mime_test_mimedb( - "t.rpm", ("application/x-redhat-package-manager", "application/x-rpm") - ) + self.mime_test_mimedb("t.rpm", "application/x-rpm") self.mime_test_mimedb("t.tar", "application/x-tar") self.mime_test_mimedb("t.cbt", "application/x-tar") self.mime_test_mimedb("t.tar.bz2", "application/x-tar", "bzip2")