- assume that Python 2.4 and 2.5 are compatible and allow checking for
fallout.
- remove PYTHON_VERSIONS_COMPATIBLE that are obsoleted by the 2.3+
default. Modify the others to deal with the removals.
VERSION 1.4.9
BUGFIXES:
* The last.fm dialog did not always properly disable options when the
username was not entered.
* Fix Amazon Cover fetching by using their new web service api.
* Don't insert items into Dynamic Mode that don't exist.
* If unavailable tracks are in the Playlist and random mode is on, don't
stop those tracks if selected; continue with available tracks.
04172008 - Look for Speex includes in new directory. Converted all text
encoding from ISO-8859-1 to UTF-8. Fixed "make dist" script for
dealing with Subversion instead of CVS. Added Speex to the README.
Upped version to 1.0.2.
04112008 - Check if Speex header has bogus data (CVE-2008-1686).
08062007 - Updated my email address.
07152007 - Minor correction in Timidity resampling code (Thanks, Sam!).
07062007 - Fixed uninitialized buffer in mpglib. (Thanks, Phil!).
10292006 - Fixed bogus memory dereference when SMPEG fails init (thanks, Chris!)
10272006 - FLAC 1.1.3 breaks their API _again_, so we try to do the right
thing at build time. (Thanks, Josh!).
05122006 - Patched to get mpglib compiling again (thanks, Sam!).
12172005 - Fixed gcc4 whining in playsound_simple.c.
12062005 - Trimmed a bunch of junk out of the build system, and now it works
on Mac OS X again.
10122005 - Check for libmodplug headers in two possible places (thanks, Tyler!)
10012005 - Added playsound_simple.c.
05302005 - Backport from devtree: Fixed automake nonsense.
11122004 - Backport from devtree: fix .voc decoder crash on file open.
05082004 - Fixed "bootstrap" to work with MacOSX.
05072004 - Backed out some commits, converted repository to Subversion, and
branched off to a 1.1.0 development tree. Changed MikMod URL...old
one is now a porn site. :(
10252003 - VOC decoder was broken. Now it isn't.
10142003 - Build system fix: acinclude.m4 had some word wrapping badness.
changes:
-security fix (unchecked array index, CVE-2008-1686)
-a few bug fixes (including the 1.1.11.1 regressions)
-new version of the pulseaudio output plugin
-open-source upport for RealAudio "cook"
changes:
* Security fixes:
- Array Indexing Vulnerability in sdpplin_parse(). (CVE-2008-0073)
- integer overflow, possibly leading to buffer overflow, CVE-2008-1482
* Reworked the plugin directory naming so that external plugins don't have
to be rebuilt for every release
* Made the version parsing much more reliable; it wasn't properly coping
with four-part version numbers
* Fixed an off-by-one in the FLAC security fix patch. This breakage was
causing failure to play some files
* Support 16-bit big-endian DTS audio
* Improved frame snapshot API. (ABI extension.)
* Re-add support for # (stream parameter separator) in raw filenames
* Fixed long delay when closing stream on dual core systems
pkgsrc note: CVE-2008-0073 was already fixed by patch
There are still some bugs lurking, but on NetBSD/amd64 4.99.48
with azalia(4), both the sun and OSS drivers work fairly well!
Try it out with audio/moc-devel.
Please let me know if you encounter any problems.
Currently, I manually create the distfile with the dist-bzip2 target
in my local sources. I plan on hosting a repository of these sources
somewhere with a NetBSD/pkgsrc branch soon.
in librt and remove ffmpeg patch, which is no longer needed.
Also, I've changed options.mk so that selecting 'samplerate' implies
'sndfile', as the former is dependent on the latter.
a getline() function taken from nbftp--bothering with gnulib was too
annoying. It works for me on NetBSD-current, but please let me
know if there are any problems.
I've also extended the check for sched_get_priority_max(3)--both here
and in audio/moc--to look in librt prior to giving up. This allows
moc to run with realtime priority on NetBSD-current; unfortunately,
presently this requires root perms.
with this package. While here, correct ncurses/ncursesw options
to provide the right arguments to configure.
Changes since 1.9.1:
2008-03-16 -- Herrie 2.0:
* Added: Brasilian Portuguese translation - Eros Carvalho
* Fixed: Make vfs.lockup.chroot expand ~/ and ~user/
* Changed: Maintainer email address
* Removed: Debian package infrastructure, per maintainer's request.
2008-02-12 -- Herrie 1.9.3:
* Added: Support for Debian GNU/kFreeBSD
* Added: Debian package infrastructure
* Fixed: Shut down the application when we lose TTY input
* Added: Chinese translation - Ji ZhengYu
2008-01-19 -- Herrie 1.9.2:
* Removed: Native Windows support - it was incomplete anyway
* Improved: Manpage reorganized to make more sense - Marcela Tiznado
* Added: Finnish translation - Jorma Karvonen
* Added: Danish translation - Joe Hansen
* Improved: GUI scrolling performance when dealing with big playlists
* Added: Switch to scroll pages up and down when selection goes out of sight
Note: This may screw up the OSS driver, but if you're using NetBSD,
I recommend using the native sun audio driver. ;)
Some type fixes. Autoconf glue and functions to use netbsd atomic ops.
Conversion to pollts() for better accuracy. Bump rev.
Based on patch provided by Zafer Aydogan via private mail.
pkgsrc changes:
* Marked as DESTDIR ready.
ChangeLog:
Jan 08 2007 Pascal Brochart <pbrochart@tuxfamily.org>
* v 0.8.0: Rewriting of the glthreads effects.
Use mutex when really created (fix deadlock).
Fix memory leak with SDL_ttf.
New translation (pt).
Pause and unpause display works again.
Aug 18 2004 Pascal Brochart <pbrochart@tuxfamily.org>
* v 0.7.0: Bug fixes.
Add sinus wave to spectrum analyzer.
Use GL_TRIANGLES instead of GL_LINE_STRIP in glthreads.
New translations (fr fr_FR).
This update was long overdue from version 0.100.0. Especially of note:
native audio(4) support (OSS driver initially converted by
jakemsr@openbsd.org), with certain changes/fixups by me (both in the
driver code and elsewhere ... it is very much still a work in progress,
but it is a work that does work, heh).
Also, some patches to do the "right thing" throughout, i.e. press the
"we are not linux" button. More fixes/enhancements coming soon.
One outstanding issue I know of: an mmap failure when trying to run
with realtime scheduling that I haven't been able to look into enough yet.
A good player which supports this is audio/moc ... I hope to add
a gstreamer plugin and much more shortly.
I am in the planning stages of a rewrite of the driver using kqueue, also.
vorbis-tools 1.2.0 -- 2008-02-21
* FLAC support now relies on libFLAC
* Support for large files
* Fixed acinclude.m4 to properly support --no switches
* ogg123: added remote control support
* ogg123: fixed a bug in esd when pressing CTRL + C
* ogg123: fixed a type mismatch in option parsing for 64 bit systems
* ogg123: configuration no longer hardcoded to /etc
* oggdec: limited support for chained Ogg bitstreams
* ogg123: compiles with older versions of libcurl
* oggdec: support decoding of multiple files into a single one
* oggenc: -k, switch for Skeleton bitstream encoding
* oggenc: fixed issues with Skeleton on big endian systems
* oggenc: proper 5.1 channel mapping support
* oggenc: FLAC module does not confuse every Ogg file as its own
* oggenc: compiles with older versions of libvorbis
* ogginfo: recognizes Skeleton, Dirac, FLAC and Kate bitstreams
* vcut: solved issues described in ticket #1313
* vorbiscomment: support for creation of long comments
* vorbiscomment: support for multiplexed Vorbis
* Several minor bug fixes
2.1.5 - January 26th, 2008 :
============================
* Added an option in the "Process Fields" scanner to enable/disable
detection of Roman numerals,
* Disabled unsynchronisation when writing ID3v2.4 tags (as Taglib doesn't
support them) to avoid problem with pictures on some applications (thanks
to Alexey Illarionov),
* Fixed compilation problem with previous versions of flac older than 1.1.3,
* Fixed the size of log view when resizing the main window,
* Fixed browser problem with the hidden directories filter: directories
starting with severals dots were also hidden,
* Fixed problem when renaming severals directories and files with the
'Rename File and Directory' scanner (thanks to vdaghan) ,
* Fixed detection of valid roman numerals in the "Process Fields" scanner
(was used some parts of the Roman library from David M. Syzdek),
* Brazilian Portuguese translation updated (thanks to doutor.zero),
* Polish translation updated (thanks to Michał Smoczyk),
* Swedish translation updated (thanks to Anders Strömer),
* German translation updated (thanks to Götz Waschk),
pkgsrc change:
Fix installed .pc file to include rpath.
Changes from 1.4 to 1.5
* Support for Mac OS X and Microsoft Windows
* Distributed under the MPL (in addition to the previous LGPL license)
* Added support for Speex files
* Added support for TrueAudio files
* Added support for WavPack files
* Added support for ID3v2 general encapsulated object frames
* Added support for ID3v2 unsynchronized lyrics frames
* Added support for ID3v2 URL frames
* Propper exports of all public classes / functions
* Updated the APE::Item API to work with value lists
* Added support to the FileRef class for new Xiph (Ogg) extensions
* Made the samples per frame for MPEG headers accessible
* Made MP3 Xing headers accessible
* Prevent invalid encodings from being written to ID3v1 tags
* Non-Latin1 ID3v2 text frames are automatically converted to UTF-8 on
write (if they are not explicitly set to UTF-16)
* Added support for reading ID3v2.2/3 unsynchronized tags
* Made it possible to search for ID3v2 comment frames by description
* Fixed a number of bugs in ID3v2 relative volume adjustment reading and
writing
* Added work arounds for iTunes writing invalid ID3v2 frame lengths
* Added work arounds for iTunes not being able to correctly parse
numerical ID3v2 genres
* Added work arounds for iTunes putting non-text information in ID3v2
comment frames
* Added a function to export strings to std::wstring
* Added a function to check ASCII compatibility of strings
* Added a function to check Latin1 compatibility of strings
changes:
-Bug fixes and minor improvements
-Voice variants have names rather than numbers
-Added languages
-Add options for different intonation styles
-Added option for gaps (pauses) between words
"ncurses" option. "wide-curses" now just toggles whether we use
wide or narrow curses, which is a much simpler knob for users.
Bump the PKGREVISION to 1.
