diff options
Diffstat (limited to 'nixpkgs/pkgs/development/libraries/libextractor')
3 files changed, 140 insertions, 0 deletions
diff --git a/nixpkgs/pkgs/development/libraries/libextractor/default.nix b/nixpkgs/pkgs/development/libraries/libextractor/default.nix new file mode 100644 index 000000000000..f7e9606bbe77 --- /dev/null +++ b/nixpkgs/pkgs/development/libraries/libextractor/default.nix @@ -0,0 +1,113 @@ +{ fetchurl, stdenv, substituteAll +, libtool, gettext, zlib, bzip2, flac, libvorbis +, exiv2, libgsf, rpm, pkgconfig, fetchpatch +, gstreamerSupport ? true, gst_all_1 ? null +# ^ Needed e.g. for proper id3 and FLAC support. +# Set to `false` to decrease package closure size by about 87 MB (53%). +, gstPlugins ? (gst: [ gst.gst-plugins-base gst.gst-plugins-good ]) +# If an application needs additional gstreamer plugins it can also make them +# available by adding them to the environment variable +# GST_PLUGIN_SYSTEM_PATH_1_0, e.g. like this: +# postInstall = '' +# wrapProgram $out/bin/extract --prefix GST_PLUGIN_SYSTEM_PATH_1_0 : "$GST_PLUGIN_SYSTEM_PATH_1_0" +# ''; +# See also <https://nixos.org/nixpkgs/manual/#sec-language-gnome>. +, gtkSupport ? true, glib ? null, gtk3 ? null +, videoSupport ? true, ffmpeg_3 ? null, libmpeg2 ? null}: + +assert gstreamerSupport -> gst_all_1 != null && builtins.isList (gstPlugins gst_all_1); +assert gtkSupport -> glib != null && gtk3 != null; +assert videoSupport -> ffmpeg_3 != null && libmpeg2 != null; + +stdenv.mkDerivation rec { + name = "libextractor-1.9"; + + src = fetchurl { + url = "mirror://gnu/libextractor/${name}.tar.gz"; + sha256 = "1zz2zvikvfibxnk1va3kgzs7djsmiqy7bmk8y01vbsf54ryjb3zh"; + }; + + patches = [ + ./fix-gcc8-build.patch + # Fixes build with exiv2 0.27 + (fetchpatch { + name = "libextractor-exiv2-0.27.patch"; + url = "https://git.archlinux.org/svntogit/community.git/plain/trunk/libextractor-exiv2-0.27.patch?h=packages/libextractor&id=4dc53f7fc69210ae571285dface108ed65d8ee53"; + sha256 = "0w4gc1q1m1yxsd4hv105nblmif465nw3g5nxzldy0x2rl9mdncg6"; + }) + (fetchpatch { + name = "CVE-2019-15531.patch"; + url = "https://git.gnunet.org/libextractor.git/patch/?id=d2b032452241708bee68d02aa02092cfbfba951a"; + sha256 = "01xhcjbzv6p53wz7y2ii76kb8m9iwvnm4ip9w4a0bpgaxqz4b9fw"; + excludes = [ "ChangeLog" ]; + }) + ] ++ stdenv.lib.optionals gstreamerSupport [ + + # Libraries cannot be wrapped so we need to hardcode the plug-in paths. + (substituteAll { + src = ./gst-hardcode-plugins.patch; + load_gst_plugins = stdenv.lib.concatMapStrings + (plugin: ''gst_registry_scan_path(gst_registry_get(), "${plugin}/lib/gstreamer-1.0");'') + (gstPlugins gst_all_1); + }) + ]; + + preConfigure = + '' echo "patching installation directory in \`extractor.c'..." + sed -i "src/main/extractor.c" \ + -e "s|pexe[[:blank:]]*=.*$|pexe = strdup(\"$out/lib/\");|g" + ''; + + buildInputs = + [ libtool gettext zlib bzip2 flac libvorbis exiv2 + libgsf rpm + pkgconfig + ] ++ stdenv.lib.optionals gstreamerSupport + ([ gst_all_1.gstreamer ] ++ gstPlugins gst_all_1) + ++ stdenv.lib.optionals gtkSupport [ glib gtk3 ] + ++ stdenv.lib.optionals videoSupport [ ffmpeg_3 libmpeg2 ]; + + configureFlags = [ + "--disable-ltdl-install" + "--with-ltdl-include=${libtool}/include" + "--with-ltdl-lib=${libtool.lib}/lib" + "--enable-xpdf" + ]; + + # Checks need to be run after "make install", otherwise plug-ins are not in + # the search path, etc. + # FIXME: Tests currently fail and the test framework appears to be deeply + # broken anyway. + doCheck = false; + #postInstall = "make check"; + + meta = { + description = "Simple library for keyword extraction"; + + longDescription = + '' GNU libextractor is a library used to extract meta-data from files + of arbitrary type. It is designed to use helper-libraries to perform + the actual extraction, and to be trivially extendable by linking + against external extractors for additional file types. + + The goal is to provide developers of file-sharing networks or + WWW-indexing bots with a universal library to obtain simple keywords + to match against queries. libextractor contains a shell-command + extract that, similar to the well-known file command, can extract + meta-data from a file an print the results to stdout. + + Currently, libextractor supports the following formats: HTML, PDF, + PS, OLE2 (DOC, XLS, PPT), OpenOffice (sxw), StarOffice (sdw), DVI, + MAN, FLAC, MP3 (ID3v1 and ID3v2), NSF(E) (NES music), SID (C64 + music), OGG, WAV, EXIV2, JPEG, GIF, PNG, TIFF, DEB, RPM, TAR(.GZ), + ZIP, ELF, S3M (Scream Tracker 3), XM (eXtended Module), IT (Impulse + Tracker), FLV, REAL, RIFF (AVI), MPEG, QT and ASF. Also, various + additional MIME types are detected. + ''; + + license = stdenv.lib.licenses.gpl2Plus; + + maintainers = [ ]; + platforms = stdenv.lib.platforms.linux; + }; +} diff --git a/nixpkgs/pkgs/development/libraries/libextractor/fix-gcc8-build.patch b/nixpkgs/pkgs/development/libraries/libextractor/fix-gcc8-build.patch new file mode 100644 index 000000000000..e04d09be8998 --- /dev/null +++ b/nixpkgs/pkgs/development/libraries/libextractor/fix-gcc8-build.patch @@ -0,0 +1,16 @@ +diff --git a/src/plugins/ole2_extractor.c b/src/plugins/ole2_extractor.c +index 072ffc5..a105840 100644 +--- a/src/plugins/ole2_extractor.c ++++ b/src/plugins/ole2_extractor.c +@@ -345,9 +345,8 @@ process_star_office (GsfInput *src, + gsf_input_read (src, size, (unsigned char*) buf); + if ( (buf[0] != 0x0F) || + (buf[1] != 0x0) || +- (0 != strncmp (&buf[2], +- "SfxDocumentInfo", +- strlen ("SfxDocumentInfo"))) || ++ (0 != strcmp (&buf[2], ++ "SfxDocumentInfo")) || + (buf[0x11] != 0x0B) || + (buf[0x13] != 0x00) || /* pw protected! */ + (buf[0x12] != 0x00) ) diff --git a/nixpkgs/pkgs/development/libraries/libextractor/gst-hardcode-plugins.patch b/nixpkgs/pkgs/development/libraries/libextractor/gst-hardcode-plugins.patch new file mode 100644 index 000000000000..73ecf5f252f6 --- /dev/null +++ b/nixpkgs/pkgs/development/libraries/libextractor/gst-hardcode-plugins.patch @@ -0,0 +1,11 @@ +--- a/src/plugins/gstreamer_extractor.c ++++ b/src/plugins/gstreamer_extractor.c +@@ -2215,6 +2215,7 @@ void __attribute__ ((constructor)) + gstreamer_init () + { + gst_init (NULL, NULL); ++ @load_gst_plugins@ + g_log_set_default_handler (&log_handler, NULL); + g_log_set_handler (NULL, G_LOG_LEVEL_MASK | G_LOG_FLAG_FATAL | G_LOG_FLAG_RECURSION, + &log_handler, NULL); + |