Ivo Emanuel Gonçalves
2008-Sep-18 19:52 UTC
[Vorbis-dev] vorbis-tools 1.3.0 BETA - Help testing.
Dear vorbis-dev lurkers, I bring you the first beta release of vorbis-tools 1.3.0. It has quite a few new features and the final version will have even more, so I am sure it is to everyone's benefit that it gets as much tested as possible so we can do a quick release. Here's the changelog so far: * Fixed an error in configure.ac; --with-speex/flac work again (#1319) * Corrected problems in the Visual Studio project files * Updated po files from the Translation Project * Added new en_GB.po, eo.po, sk.po and vi.po translation files * Added AC_USE_SYSTEM_EXTENSIONS to configure.ac; no more autogen.sh warnings * Corrected SUBLANG values in intl/localename.c (#1415) * Modify -v to -V on oggenc and oggdec for consistency (#1112) * Fix for utf8_decode in Windows; improves behavior in vorbiscomment (#268) * ogg123: backported fix from libfishsound to patch the Speex decoder (#1347) * ogg123: fixed CPU issue when outputting to a closed pipe (#1357) * ogg123: return value to stop decoding after buffer is shut down (#1357) * oggdec: gettextized help text (#1385) * oggenc: fixed a core dump while resampling from FLAC (#1316) * oggenc: fixed a typo in the Skeleton handling routine * oggenc: fixed remapping channels bug (#1326) * oggenc: support for WAVE_FORMAT_EXTENSIBLE headers (#1326) * oggenc: support for 32 bit Wave files (#1326) * oggenc: --ignorelength; support for Wave files > 4 GB (#1326) * oggenc: split help text into manageable chunks to help translators (#1385) * oggenc: --utf8 command similar to vorbiscomment's --raw (#268) * oggenc: fixed the encoding of extended characters in Windows (#268) * oggenc: validate raw UTF-8 sent to oggenc (#268) * ogginfo: support for information in Kate streams (#1360) * vcut: 64 bit fixes (#1366) * vorbiscomment: correct memory allocation (#472) * vorbiscomment: validate raw UTF-8 sent to vorbiscomment (#268) Those numbers at the end (#xxx) are Trac tickets in case you'd like to know more about those issues. Grab the source code at[1]. To compile in Windows use Visual Studio or mingw. To compile in Linux, automake will do. If you don't want to try to compile for yourself you'll have to wait for someone from the community to do it for you. Known bugs so far: * oggdec, if compiled with Visual Studio, will not work in Windows. This will (hopefully) be fixed before the final release. * a display problem in ogg123 if the terminal window is too small Features expected to be in the final 1.3.0: * ReplayGain support in ogg123 (by William Poetra Yoga Hadisoeseno) * Lyrics (Ogg Kate) encoding in oggenc (by ogg.k.ogg.k) * Support for multi-line comments in vorbiscomment (by three different authors since 2002) * And possibly a big rewrite of vcut (by Michael Gold) if it's finished on time -Ivo [1] http://spreadopenmedia.org/tmp/vorbis-tools-1.3.0b1.tar.gz sha1 checksum: 7e0d6e487d0be7bd1926962b0a225362626f6477
Ivo Emanuel Gon?alves wrote:> Dear vorbis-dev lurkers, I bring you the first beta release of > vorbis-tools 1.3.0. It has quite a few new features and the final > version will have even more, so I am sure it is to everyone's benefit > that it gets as much tested as possible so we can do a quick release.I tried compiling it in environment of WindowsXP/Mingw+msys. However, libogg and libvorbis are done dynamic link of even if I use the "--enable-static" option. I can make the binary that was linked static to without a problem if I appoint "libogg.a, libvorbis.a etc." with manual operation. Aoyumi
On Thu, Sep 18, 2008 at 20:52:47 +0100, Ivo Emanuel Gon?alves wrote:> Dear vorbis-dev lurkers, I bring you the first beta release of > vorbis-tools 1.3.0. It has quite a few new features and the final > version will have even more, so I am sure it is to everyone's benefit > that it gets as much tested as possible so we can do a quick release.While testing vcut I occasionally run into a problem with oggenc where two output files end up with the same serial number because they were encoded in the same second. I've filed this in trac: https://trac.xiph.org/ticket/1432 There's a patch attached to that ticket to use (time(NULL) ^ getpid()) as the RNG seed, which fixes the problem (and is the same thing I do in my new vcut patch) -- but I'm not sure how portable it is. -- Michael -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: Digital signature Url : http://lists.xiph.org/pipermail/vorbis-dev/attachments/20080927/a9b6ed27/attachment.pgp