This is an informal changelog, a summary of changes in each release. (See also <ahref="documentation_bugs.html">known bugs</a>.) Particulary important for developers is the precise description of changes to the library interfaces. See also the <ahref="api/group__porting.html">porting guide</a> for specific instructions on porting to newer versions of FLAC.<br/>
<li>The <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_sector_align">--sector-align</a></span> option of <spanclass="commandname">flac</span> has been deprecated and may not exist in future versions. <ahref="http://www.etree.org/shnutils/shntool/">shntool</a> provides similar functionality.</li>
<li>The <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_ignore_chunk_sizes">--ignore-chunk-sizes</a></span> option has been added to the <spanclass="commandname">flac</span> command line tool.</li>
<li>Added support for encoding from and decoding to the RF64 format, and a new corresponding option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_force_rf64_format">--force-rf64-format</a></span>. (<ahref="http://sourceforge.net/p/flac/feature-requests/78/">SF #1762502</a>). <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_keep_foreign_metadata">--keep-foreign-metadata</a></span> is also supported.</li>
<li>Added support for encoding from and decoding to the Sony Wave64 format, and a new corresponding option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_force_wave64_format">--force-wave64-format</a></span>. (<ahref="http://sourceforge.net/p/flac/feature-requests/79/">SF #1769582</a>). <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_keep_foreign_metadata">--keep-foreign-metadata</a></span> is also supported.</li>
<li>Added new options <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_preserve_modtime">--preserve-modtime</a></span> and <spanclass="argument"><ahref="documentation_tools_flac.html#negative_options">--no-preserve-modtime</a></span> to specify whether or not output files should copy the timestamp and permissions from their input files. The default is <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_preserve_modtime">--preserve-modtime</a></span> as in previous versions. (<ahref="http://sourceforge.net/p/flac/feature-requests/85/">SF #1805428</a>).</li>
<li>Allow MM:SS:FF and MM:SS.SS time formats in non-CD-DA cuesheets. (<ahref="http://sourceforge.net/p/flac/feature-requests/95/">SF #1947353</a>, <ahref="http://sourceforge.net/p/flac/bugs/338/">SF #2182432</a>)</li>
<li>The <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_sector_align">--sector-align</a></span> option of <spanclass="commandname">flac</span> has been deprecated and may not exist in future versions. <ahref="http://www.etree.org/shnutils/shntool/">shntool</a> provides similar functionality. (<ahref="http://sourceforge.net/p/flac/feature-requests/86/">SF #1805946</a>)</li>
<li>Improved error message when user attempts to decode a non-FLAC file (<ahref="http://sourceforge.net/p/flac/bugs/341/">SF #2222789</a>).</li>
<li>Fix bug where <spanclass="commandname">flac</span> was disallowing use of <spanclass="argument">--replay-gain</span> when encoding from stdin (<ahref="http://sourceforge.net/p/flac/bugs/313/">SF #1840124</a>).</li>
<li>Fix bug with fractional seconds on some locales (<ahref="http://sourceforge.net/p/flac/bugs/309/">SF #1815517</a>, <ahref="http://sourceforge.net/p/flac/bugs/321/">SF #1858012</a>).</li>
<li>Allow MM:SS:FF and MM:SS.SS time formats in non-CD-DA cuesheets. (<ahref="http://sourceforge.net/p/flac/feature-requests/95/">SF #1947353</a>, <ahref="http://sourceforge.net/p/flac/bugs/338/">SF #2182432</a>)</li>
and <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_apply_replaygain_which_is_not_lossless">--apply-replaygain-which-is-not-lossless</a></span>
option for <spanclass="commandname">flac</span>.</li>
<li>libFLAC encoder was defaulting to level 0 compression instead of 5 (<ahref="http://sourceforge.net/p/flac/bugs/310/">SF #1816825</a>).</li>
<li>Fix bug in bitreader handling of read callback returning a short count (<ahref="http://sourceforge.net/p/flac/bugs/345/">SF #2490454</a>).</li>
<li>Improve decoder's ability to distinguish between a FLAC sync code and an MPEG one (<ahref="http://sourceforge.net/p/flac/bugs/346/">SF #2491433</a>).</li>
<li>With the new <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_keep_foreign_metadata">--keep-foreign-metadata</a></span> in <spanclass="commandname">flac</span>, non-audio RIFF and AIFF chunks can be stored in FLAC files and recreated when decoding. This allows, among other, things support for archiving BWF files and other WAVE files from editing tools that preserves all the metadata. </li>
<li>Specified 2 new APPLICATION metadata blocks for storing WAVE and AIFF chunks (for use with <ahref="documentation_tools_flac.html#flac_options_keep_foreign_metadata">--keep-foreign-metadata</a> in <spanclass="commandname">flac</span>).</li>
<li>This is not a format change, but changed default extension for Ogg FLAC from .ogg to .oga, according to new Xiph <ahref="http://wiki.xiph.org/index.php/MIME_Types_and_File_Extensions">specification</a> (<ahref="http://sourceforge.net/p/flac/bugs/283/">SF #1762492</a>).</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_no_utf8_convert">--no-utf8-convert</a></span> which works like it does in <spanclass="commandname">metaflac</span> (<ahref="http://sourceforge.net/p/flac/feature-requests/35/">SF #973740</a>).</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_keep_foreign_metadata">--keep-foreign-metadata</a></span> which can save/restore RIFF and AIFF chunks to/from FLAC files (<ahref="http://sourceforge.net/p/flac/feature-requests/9/">SF #363478</a>).</li>
<li>Changed default extension for Ogg FLAC from .ogg to .oga, according to new Xiph <ahref="http://wiki.xiph.org/index.php/MIME_Types_and_File_Extensions">specification</a> (<ahref="http://sourceforge.net/p/flac/bugs/283/">SF #1762492</a>).</li>
<li>Fixed bug where using <spanclass="argument">--replay-gain</span> without any padding option caused only a small PADDING block to be created (<ahref="http://sourceforge.net/p/flac/bugs/282/">SF #1760790</a>).</li>
<li>Fixed bug where encoding from stdin on Windows could fail if WAVE/AIFF contained unknown chunks (<ahref="http://sourceforge.net/p/flac/bugs/290/">SF #1776803</a>).</li>
<li>Fixed bug where importing non-CDDA cuesheets would cause an invalid lead-out track number (<ahref="http://sourceforge.net/p/flac/bugs/286/">SF #1764105</a>).</li>
<li>Changed default extension for Ogg FLAC from .ogg to .oga, according to new Xiph <ahref="http://wiki.xiph.org/index.php/MIME_Types_and_File_Extensions">specification</a> (<ahref="http://sourceforge.net/p/flac/bugs/283/">SF #1762492</a>).</li>
<li>Fixed bug where importing non-CDDA cuesheets would cause an invalid lead-out track number (<ahref="http://sourceforge.net/p/flac/bugs/286/">SF #1764105</a>).</li>
<li>New configure option <spanclass="argument">--disable-cpplibs</span> to prevent building libFLAC++ (<ahref="http://sourceforge.net/p/flac/patches/23/">SF #1723295</a>).</li>
<li>Fixed bug compiling <spanclass="commandname">flac</span> without Ogg support (<ahref="http://sourceforge.net/p/flac/bugs/281/">SF #1760786</a>).</li>
<li>Fixed bug where sometimes an existing installation of flac could interfere with the build process (<ahref="http://sourceforge.net/p/flac/bugs/285/">SF #1763690</a>).</li>
<li>OS X fixes (<ahref="http://sourceforge.net/p/flac/patches/25/">SF #1786225</a>).</li>
<li>Added support for encoding the residual coding method introduced in libFLAC 1.2.0 (RESIDUAL_CODING_METHOD_PARTITIONED_RICE2) which will encode 24-bit files more efficiently.</li>
<li>One of the reserved bits in the FLAC frame header has been assigned for future use; make sure to refer to the <ahref="api/group__porting__1__1__4__to__1__2__0.html">porting guide</a> if you parse FLAC streams manually.</li>
<li>Added a new undocumented option <spanclass="argument">--ignore-chunk-sizes</span> for ignoring the size of the 'data' chunk (WAVE) or 'SSND' chunk (AIFF). Can be used to encode files with bogus data sizes (e.g. with WAV files piped from foobar2000 to flac.exe as an external encoder). <b>Use with caution</b>: all subsequent data is treated as audio, so the data/SSND chunk must be the last or the following data/tags will be treated as audio and encoded.</li>
Interface changes (see also the <ahref="api/group__porting__1__1__4__to__1__2__0.html">porting guide</a> for specific instructions on porting to FLAC 1.2.0):
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_warnings_as_errors">-w,--warnings-as-errors</a></span> for treating all warnings as errors.</li>
<li>Allow <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_picture">--picture</a></span> option to take only a filename, and have all other attributes extracted from the file itself.</li>
<li>Fixed a bug that caused suboptimal default compression settings in some locales (<ahref="http://sourceforge.net/p/flac/bugs/237/">SF #1608883</a>).</li>
<li>Fixed a bug where FLAC-to-FLAC transcoding of a corrupted FLAC file would truncate the transcoded file at the first error (<ahref="http://sourceforge.net/p/flac/bugs/241/">SF #1615019</a>).</li>
<li>Fixed a bug where using <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_decode_through_errors">-F</a></span> with FLAC-to-FLAC transcoding of a corrupted FLAC would have no effect (<ahref="http://sourceforge.net/p/flac/bugs/242/">SF #1615391</a>).</li>
<li>Fixed a bug where new PICTURE metadata blocks specified with <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_picture">--picture</a></span> would not be transferred during FLAC-to-FLAC transcoding (<ahref="http://sourceforge.net/p/flac/bugs/246/">SF #1627993</a>).</li>
<li>Allow <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_import_picture_from">--import-picture-from</a></span> option to take only a filename, and have all other attributes extracted from the file itself.</li>
<li>Fixed a bug in the XMMS plugin where Ctrl-3 (file info) would cause a crash if the file did not exist (<ahref="http://sourceforge.net/p/flac/patches/20/">SF #1634941</a>).</li>
<li>Fixed a makefile linkage bug with libogg (<ahref="http://sourceforge.net/p/flac/bugs/239/">SF #1611414</a>).</li>
<li>Added pkg-config files for libFLAC and libFLAC++ (<ahref="http://sourceforge.net/p/flac/patches/21/">SF #1647881</a>).</li>
<li>Added <spanclass="argument">--disable-ogg</span> option for building without Ogg support even if libogg is installed (<ahref="http://sourceforge.net/p/flac/bugs/165/">SF #1196996</a>).</li>
<li>Completely rewritten bitbuffer which uses native machine word size instead of bytes for dramatic speed improvements. The speedup should be most dramatic on CPUs with slower byte manipulation capability and big-endian machines.</li>
<li>Fixed a bug with default apodization settings that were erroneous in some locales (<ahref="http://sourceforge.net/p/flac/bugs/237/">SF #1608883</a>).</li>
<li>(behavior only) FLAC__stream_encoder_set_metadata() now makes a copy of the "metadata" array of pointers (but still not copies of the objects themselves) so the client does not need to maintain its copy of the array after the call.</li>
<li><spanclass="commandname">flac</span> now supports FLAC and Ogg FLAC as input to the encoder (e.g. can re-encode FLAC to FLAC) and preserve all the metadata like tags, etc.</li>
<li>New <spanclass="code"><ahref="format.html#def_PICTURE">PICTURE</a></span> metadata block for storing things like cover art, new <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_picture">--picture</a></span> option to <spanclass="commandname">flac</span> and <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_import_picture_from">--import-picture-from</a></span> option to <spanclass="commandname">metaflac</span> for importing pictures, new <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_export_picture_to">--export-picture-to</a></span> option to <spanclass="commandname">metaflac</span> for exporting pictures, and metadata API <ahref="api/group__flac__metadata__level0.html#ga3">additions</a> for searching for suitable pictures based on type, size and color constraints.</li>
<li>Fixed a bug in Ogg FLAC encoding where metadata was not being updated properly. Existing Ogg FLAC files should be recoded to fix up the metadata, e.g. <spanclass="command">flac -Vf -S 10s --ogg file.ogg</span></li>
<li>In the developer libraries, the interface has been simplfied by merging the three decoding layers into a single class; ditto for the encoders. Also, libOggFLAC has been merged into libFLAC and libOggFLAC++ has been merged into libFLAC++ so there is a single API supporting both native FLAC and Ogg FLAC.</li>
<li>Speaker assignments and channel orders for 3-6 channels (see <ahref="format.html#frame_header">frame header</a>).</li>
<li>Further restrictions on the <ahref="format.html#subset">FLAC subset</a> when the sample rate is <=48kHz; in this case the maximum LPC order is now 12 and maximum blocksize is 4608. This is to further limit the processing and memory requirements for hardware implementations while not measurably affecting compression.</li>
<li>Improved the <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_decode_through_errors">-F</a></span> option to allow decoding of FLAC files whose metadata is corrupted, and other kinds of severe corruption.</li>
<li>Encoder can now take FLAC and Ogg FLAC as input. The output FLAC file will have all the same metadata as the original unless overridden with options on the command line.</li>
<li>Encoder can now take WAVEFORMATEXTENSIBLE WAVE files as input; decoder will output WAVEFORMATEXTENSIBLE WAVE files when necessary to conform to the latest Microsoft specifications.</li>
<li>Now properly supports AIFF and WAVEFORMATEXTENSIBLE multichannel input, performing necessary channel reordering both for encoding and decoding. WAVEFORMATEXTENSIBLE channel mask is also saved to a tag on encoding and restored on decoding for situations when there is no natural mapping to FLAC channel assignments.</li>
<li>Expanded support for "odd" sample resolutions to WAVE and AIFF input; all resolutions from 4 to 24 bits-per-sample now supported for all input types.</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_tag_from_file">--tag-from-file</a></span> for setting a tag from file (e.g. for importing a cuesheet as a tag).</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_picture">--picture</a></span> for adding pictures.</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_apodization">--apodization</a></span> for specifying the window function(s) to be used in LPC analysis.</li>
<li>Added support for encoding from non-compressed AIFF-C (<ahref="http://sourceforge.net/p/flac/bugs/143/">SF #1090933</a>).</li>
<li>Importing of non-CDDA-compliant cuesheets now only issues a warning, not an error (see <ahref="http://www.hydrogenaud.io/forums/index.php?showtopic=31282">here</a>).</li>
<li>MD5 comparison failures on decoding are now an error instead of a warning and will also return a non-zero exit code (<ahref="http://sourceforge.net/p/flac/bugs/221/">SF #1493725</a>).</li>
<li>Fixed a bug that caused a crash when <spanclass="argument">-a</span> and <spanclass="argument">-t</span> were used together (<ahref="http://sourceforge.net/p/flac/bugs/173/">SF #1229481</a>).</li>
<li>Fixed a bug with --sector-align where appended samples were not always totally silent (<ahref="http://sourceforge.net/p/flac/bugs/179/">SF #1237707</a>).</li>
<li>Fixed a bug printing out unknown AIFF subchunk names (<ahref="http://sourceforge.net/p/flac/bugs/186/">SF #1267476</a>).</li>
<li>Fixed a bug where WAVE files with "data" subchunks of size 0 where accepted (<ahref="http://sourceforge.net/p/flac/bugs/190/">SF #1293830</a>).</li>
<li>Fixed a bug where sync error at end-of-stream of truncated files was not being caught (<ahref="http://sourceforge.net/p/flac/bugs/183/">SF #1244071</a>).</li>
<li>Fixed a problem with filename parsing if file does not have extension but also has a . in the path (<ahref="http://sourceforge.net/p/flac/bugs/159/">SF #1161916</a>).</li>
<li>Fixed a problem with fractional-second parsing for <spanclass="argument">--skip</span>/<spanclass="argument">--until</span> in some locales (<ahref="http://sourceforge.net/p/flac/bugs/125/">SF #1031043</a>).</li>
<li>Increase progress report rate when -p and -e are used together (<ahref="http://sourceforge.net/p/flac/bugs/229/">SF #1580122</a>).</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_set_tag_from_file">--set-tag-from-file</a></span> for setting a tag from file (e.g. for importing a cuesheet as a tag).</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_import_picture_from">--import-picture-from</a></span> for importing pictures.</li>
<li>Added a new option <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_export_picture_to">--export-picture-to</a></span> for exporting pictures.</li>
<li>Added shorthand operation <spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_remove_replay_gain">--remove-replay-gain</a></span> for removing ReplayGain tags.</li>
<li><spanclass="argument"><ahref="documentation_tools_metaflac.html#metaflac_shorthand_export_cuesheet_to">--export-cuesheet-to</a></span> now properly specifies the FLAC file name (<ahref="http://sourceforge.net/p/flac/feature-requests/46/">SF #1272825</a>).</li>
<li>Fixed UTF-8 decoder to disallow non-shortest-form and surrogate sequences (see <ahref="http://www.unicode.org/versions/corrigendum1.html">here</a>).</li>
<li>Added support for building on OS/2 with EMX (<ahref="http://sourceforge.net/p/flac/bugs/174/">SF #1229495</a>)</li>
<li>Added support for building with Borland C++ (<ahref="http://sourceforge.net/p/flac/patches/17/">SF #1599018</a>)</li>
<li>Added a <spanclass="argument">--disable-xmms-plugin</span> option to <spanclass="command">configure</span> to prevent building the XMMS plugin (<ahref="http://sourceforge.net/p/flac/feature-requests/33/">SF #930494</a>).</li>
<li>Added a <spanclass="argument">--disable-doxygen-docs</span> option to <spanclass="command">configure</span> for disabling Doxygen-based API doc generation (<ahref="http://sourceforge.net/p/flac/patches/12/">SF #1365935</a>).</li>
<li>Added a <spanclass="argument">--disable-thorough-tests</span> option to <spanclass="command">configure</span> to do the basic library, stream, and tool tests in a reasonable time (<ahref="http://sourceforge.net/p/flac/feature-requests/40/">SF #1077948</a>).</li>
<li>Added large file support with <spanclass="argument">AC_SYS_LARGEFILE</span>; use <spanclass="argument">--disable-largefile</span> with <spanclass="command">configure</span> to disable.</li>
<li>libFLAC: Fixed a bug where missing STREAMINFO fields (min/max framesize, total samples, MD5 sum) and seek point offsets were not getting rewritten back to Ogg FLAC file (<ahref="http://sourceforge.net/p/flac/bugs/197/">SF #1338969</a>).</li>
<li>libFLAC: Fixed UTF-8 decoder to disallow non-shortest-form and surrogate sequences (see <ahref="http://www.unicode.org/versions/corrigendum1.html">here</a>).</li>
<li>libFLAC: Fixed a bug in the return value for <spanclass="argument">FLAC__stream_decoder_set_metadata_respond_application()</span> and <spanclass="argument">FLAC__stream_decoder_set_metadata_ignore_application()</span> when there was a memory allocation error (<ahref="http://sourceforge.net/p/flac/bugs/176/">SF #1235005</a>).</li>
Interface changes (see also the <ahref="api/group__porting__1__1__2__to__1__1__3.html">porting guide</a> for specific instructions on porting to FLAC 1.1.3):
<li><b>Changed</b> FLAC__stream_encoder_finish() now returns a FLAC__bool to signal a verify failure, or error processing last frame or updating metadata.</li>
<li><b>Changed</b> FLAC__StreamDecoderState: removed state FLAC__STREAM_DECODER_UNPARSEABLE_STREAM</li>
<li><b>Changed</b> FLAC__StreamDecoderErrorStatus: new error code FLAC__STREAM_DECODER_ERROR_STATUS_UNPARSEABLE_STREAM</li>
<li>The above two changes mean that when the decoder encounters what it thinks are unparseable frames from a future decoder, instead of returning a fatal error with the FLAC__STREAM_DECODER_UNPARSEABLE_STREAM state, it just calls the error callback with FLAC__STREAM_DECODER_ERROR_STATUS_UNPARSEABLE_STREAM and leaves the behavior up to the application.</li>
<li><b>Changed</b> FLAC::Encoder::Stream::finish() now returns a bool to signal a verify failure, or error processing last frame or updating metadata.</li>
<li>Sped up decoding by a few percent overall.</li>
<li>Sped up encoding when not using LPC (i.e. when using <spanclass="commandname">flac</span> options <spanclass="argument">-0</span>, <spanclass="argument">-1</span>, <spanclass="argument">-2</span>, or <spanclass="argument">-l 0</span>).</li>
<li>Fixed a decoding bug that could cause sync errors with some ID3v1-tagged FLAC files.</li>
<li>New option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_input_size">--input-size</a></span> to manually specify the input size when encoding raw samples from stdin.</li>
<li>Added support for HTTP streaming in XMMS plugin. <b>NOTE</b>: there is a bug in the XMMS mpg123 plugin that hijacks FLAC streams; to fix it you will need to add the '.flac' extension to the list of exceptions in <spanclass="code">Input/mpg123/mpg123.c:is_our_file()</span> in the XMMS sources and recompile.</li>
<li>libFLAC: Sped up Rice block decoding in the bitbuffer, resulting in decoding speed gains of a few percent.</li>
<li>libFLAC: Sped up encoding when not using LPC (i.e. <spanclass="code">max_lpc_order == 0</span>).</li>
<li>libFLAC: Trailing NUL characters maintained on Vorbis comment entries so they can be treated like C strings.</li>
<li>libFLAC: More FLAC tag (i.e. Vorbis comment) validation.</li>
<li>libFLAC: Fixed a bug in the logic that determines the frame or sample number in a frame header; the bug could cause sync errors with some ID3v1-tagged FLAC files.</li>
<li>libFLAC, libOggFLAC: Can now be compiled to use only integer instructions, including encoding. The decoder is almost completely integer anyway but there were a couple places that needed a fixed-point replacement. There is no fixed-point version of LPC analysis yet, so if libFLAC is compiled integer-only, the encoder will behave as if the max LPC order is 0 (i.e. used fixed predictors only). LPC decoding is supported in all cases as it always was integer-only.</li>
</ul>
</li>
<li>
Interface changes:
<ul>
<li>
libFLAC:
<ul>
<li><b>Changed:</b> Metadata object interface now maintains a trailing NUL on Vorbis comment entries for convenience.</li>
<li><b>Changed:</b> Metadata object interface now validates all Vorbis comment entries on input and returns false if an entry does not conform to the Vorbis comment spec.</li>
<li><b>Changed</b> the signature of FLAC__metadata_object_vorbiscomment_entry_matches(): the first argument is now <spanclass="code">FLAC__StreamMetadata_VorbisComment_Entry entry</span> (was <spanclass="code">const FLAC__StreamMetadata_VorbisComment_Entry *entry</span>), i.e. <spanclass="code">entry</span> is now pass-by-value.</li>
</ul>
</li>
<li>
libFLAC++:
<ul>
<li><b>Changed:</b> Metadata object interface now maintains a trailing NUL on Vorbis comment values for convenience.</li>
<li><b>Changed:</b> Metadata object interface now validates all Vorbis comment entries on input and returns false if an entry does not conform to the Vorbis comment spec.</li>
<li><b>Changed:</b> All Metadata objects' operator=() methods now return a reference to themselves.</li>
<li><b>Added</b> methods to FLAC::Metadata::VorbisComment::Entry for setting comment values from null-terminated strings:
<li><b>Changed</b> the signature of FLAC::Metadata::VorbisComment::get_vendor_string() and FLAC::Metadata::VorbisComment::set_vendor_string() to use a UTF-8, NUL-terminated string <spanclass="code">const FLAC__byte *</span> for the vendor string instead of <spanclass="code">FLAC::Metadata::VorbisComment::Entry</span>.</li>
<li><b>Added</b> FLAC::Metadata::*::assign() to all Metadata objects.</li>
<li>New optimizations almost double the decoding speed on PowerPC (e.g. Mac G4/G5)</li>
<li>A native OS X release thanks to updated Project Builder and autotools files</li>
</ul>
</li>
<li>
FLAC format:
<ul>
<li>Made invalid the metadata block type 127 so that audio frames can always be distinguished from metadata by seeing 0xff as the first byte. (This was also required for the Ogg FLAC mapping.)</li>
<li>First official FLAC->Ogg bitstream mapping standardized (see new <ahref="ogg_mapping.html">FLAC-to-Ogg mapping specification</a>). See the documentation for the <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_ogg">--ogg</a></span> switch about having to re-encode older Ogg FLAC files.</li>
<li>New option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_force">-f</a></span> (<spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_force">--force</a></span>) to force overwriting if the output file already exists.</li>
<li>New option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_cue">--cue</a></span> to select a specific section to decode using cuesheet track/index points.</li>
<li>New option <spanclass="argument"><ahref="documentation_tools_flac.html#flac_options_totally_silent">--totally-silent</a></span> to suppress all output.</li>
<li>New (but undocumented) option <spanclass="argument">--apply-replaygain-which-is-not-lossless</span> which applies ReplayGain to the decoded output. See <ahref="http://www.hydrogenaud.io/forums/index.php?showtopic=17293&st=11">this thread</a> for usage and caveats.</li>
<li>When encoding to Ogg FLAC, use a random serial number (instead of 0 as was done before) when a serial number is not specified.</li>
<li>When encoding multiple Ogg FLAC streams, <spanclass="argument">--serial-number</span> or random serial number sets the first number, which is then incremented for subsequent streams (before, the same serial number was used for all streams).</li>
<li>Decoder no longer exits with an error when writing to stdout and the pipe is broken.</li>
<li>Better explanation of common error messages.</li>
<li>Default extension when writing AIFF files is .aif (before, it was .aiff).</li>
<li>Write more common representation of SANE numbers in AIFF files.</li>
<li>Bug fix: calculating ReplayGain on 48kHz streams.</li>
<li>Bug fix: check for supported block alignments in WAVE files.</li>
<li>Bug fix: "offset" field in AIFF SSND chunk properly handled.</li>
<li>Bug fix: small memory leak in XMMS plugin.</li>
</ul>
</li>
<li>
build system:
<ul>
<li><spanclass="code">ordinals.h</span> is now static, not a build-generated file anymore.</li>
</ul>
</li>
<li>
libraries:
<ul>
<li>libFLAC: PPC+Altivec optimizations of some decoder routines.</li>
<li>libFLAC: Make stream encoder encode the blocksize and sample rate in the frame header if at all possible (not in STREAMINFO), even if subset encoding was not requested.</li>
<li>libFLAC: Bug fix: fixed seek routine where infinite loop could happen when seeking past end of stream.</li>
<li>libFLAC, libFLAC++: added methods to skip single frames, useful for quickly finding frame boundaries (see interface changes below).</li>
<li>libOggFLAC, libOggFLAC++: New seekable-stream and file encoder and decoder APIs to match native FLAC APIs (see interface changes below).</li>
</ul>
</li>
<li>
Interface changes:
<ul>
<li>
libFLAC:
<ul>
<li><b>Added</b> FLAC__metadata_get_tags()</li>
<li><b>Added</b> callback-based versions of metadata editing functions:
<li>All code is now <ahref="http://valgrind.org/">Valgrind</a>-clean!</li>
<li>New <ahref="format.html#def_CUESHEET">CUESHEET</a> metadata block for storing CD TOC and index point information. Now a CD can be completely backed up to a single FLAC file for archival.</li>
<li>Complete MSVC build system with .dsp projects for everything, which can build both static libs and DLLs, and in debug or release mode, all in the same source tree.</li>
</ul>
<spanclass="commandname">flac</span>:
<ul>
<li>Can now decode FLAC to AIFF; new <spanclass="argument">--force-aiff-format</span> option.</li>
<li>New <spanclass="argument">--cuesheet</span> option for reading and storing a cuesheet when encoding a whole CD. Automatically creates seek points for track and index points unless <spanclass="argument">--no-cued-seekpoints</span> is used.</li>
<li>New <spanclass="argument">--replay-gain</span> option for calculating ReplayGain values and storing them as tags.</li>
<li>New <spanclass="argument">--until</span> option complements <spanclass="argument">--skip</span> to stop decoding at a specified point in the stream.</li>
<li><spanclass="argument">--skip</span> and <spanclass="argument">--until</span> now also accept mm:ss.ss format.</li>
<li>New <spanclass="argument">-S #s</span> flavor to specify seekpoints every '#' number of seconds.</li>
<li><spanclass="commandname">flac</span> now defaults to <spanclass="argument">-S 10s</span> instead of <spanclass="argument">-S 100x</span> for the seek table.</li>
<li><spanclass="commandname">flac</span> now adds a 4k PADDING block by default (turn off with <spanclass="argument">--no-padding</span>).</li>
<li>Fixed a bug with --skip and AIFF-to-FLAC encoding.</li>
<li>Fixed a bug where decoding a FLAC file whose total_samples==0 in the STREAMINFO would corrupt the WAVE header.</li>
</ul>
<spanclass="commandname">metaflac</span>:
<ul>
<li>New <spanclass="argument">--import-cuesheet-from</span> option for reading and storing a cuesheet to a FLAC-encoded CD. Automatically creates seek points for track and index points unless <spanclass="argument">--no-cued-seekpoints</span> is used.</li>
<li>New <spanclass="argument">--export-cuesheet-to</span> option for writing a cuesheet from a FLAC file for use with CD authoring software.</li>
<li>New <spanclass="argument">--add-replay-gain</span> option for calculating ReplayGain values and storing them as tags.</li>
<li>New <spanclass="argument">--add-seekpoint</span> option to add seekpoints to an existing FLAC file. Includes new <spanclass="argument">--add-seekpoint=#s</span> flavor to add seekpoints every '#' number of seconds.</li>
</ul>
XMMS plugin:
<ul>
<li>Configurable sample resolution conversion with dither.</li>
<li>ReplayGain support with customizable noise shaping, pre-amp, and optional hard limiter.</li>
<li>New Vorbis comment editor.</li>
<li>File info now works.</li>
<li>Bitrate now shows the smoothed instantaneous bitrate.</li>
<li>Uses the ARTIST tag if there is no PERFORMER tag.</li>
</ul>
Winamp2 plugin:
<ul>
<li>Configurable sample resolution conversion with dither.</li>
<li>ReplayGain support with customizable noise shaping, pre-amp, and optional hard limiter.</li>
<li>File info now works.</li>
<li>Uses the ARTIST tag if there is no PERFORMER tag.</li>
</ul>
Libraries (developers take note!):
<ul>
<li>All code and tests are instrumented for Valgrind. All tests run Valgrind-clean, meaning no memory leaks or buffer over/under-runs.</li>
<li>Separate 64-bit datapath through the filter in <spanclass="commandname">libFLAC</span> for better compression of >16 bps files.</li>
<li><spanclass="code">FLAC__metadata_object_new(FLAC__METADATA_TYPE_VORBIS_COMMENT)</span> now sets the vendor string.</li>
<li>The documentation on the usage of <spanclass="code">FLAC::Iterator::get_block()</span> in <spanclass="commandname">libFLAC++</span> has an important correction. If you use this class make sure to read <ahref="api/group__flacpp__metadata__level2.html">this</a>.</li>
<li>AIFF input support (thanks to Brady Patterson).</li>
<li>Small decoder speedup.</li>
<li><spanclass="argument">--sector-align</span> now supported for raw input files.</li>
<li>New -T, --tag options for adding Vorbis comments while encoding.</li>
<li>New --serial-number option for use with --ogg.</li>
<li>Automatically writes vendor string in Vorbis comments.</li>
<li>Drastically reduced memory requirements.</li>
<li>Fixed bug where extra fmt/data chunks that were supposed to be skipped were not getting skipped.</li>
<li>Fixed bug in granulepos setting for Ogg FLAC streams.</li>
<li>Fixed memory leak when encoding multiple files with -V.</li>
</ul>
<spanclass="commandname">metaflac</span>:
<ul>
<li>UTF-8 support in Vorbis comments.</li>
<li>New --import-vc-from and --export-vc-to commands for importing/exporting Vorbis comments from/to a file. For example, the following can be used to copy tags back and forth:<br/>
<li>Fixed <ahref="http://sourceforge.net/p/flac/bugs/54/">bug #606796</a> where <spanclass="commandname">metaflac</span> was failing on read-only files.</li>
</ul>
Libraries:
<ul>
<li>All APIs now meticulously documented via Doxygen. <ahref="api/index.html">See here</a>.</li>
<li>New <spanclass="commandname">libOggFLAC</span> and <spanclass="commandname">libOggFLAC++</span> libraries. These wrap around <spanclass="commandname">libFLAC</span> to provide encoding and decoding of Ogg FLAC streams, providing interfaces similar to the ones of the native FLAC libraries. These are also documented via Doxygen.</li>
<li>New FLAC__SeekableStreamEncoder and FLAC__FileEncoder in <spanclass="commandname">libFLAC</span> simplify common encoding tasks.</li>
<li>New verify mode in all encoders.</li>
<li>FLAC__stream_encoder_finish() now resets the defaults just like the stream decoders.</li>
<li>Drastically reduced memory requirements of encoders and decoders.</li>
<li>Encoder now automatically writes vendor string in VORBIS_COMMENT block.</li>
<li>Encoding speedup of fixed predictors and MD5 speedup for 16bps mono/stereo signals on x86 (thanks to Miroslav Lichvar).</li>
<li>Fixed bug in metadata interface where a bps in STREAMINFO > 16 was incorrectly parsed.</li>
<li>Fixed bug where aborting stream decoder could cause infinite loop.</li>
<li>24-bit input support restored in <spanclass="commandname">flac</span>.</li>
<li>Decoder speedup in <spanclass="commandname">libFLAC</span>, which is directly passed on to the command-line decoder and plugins.</li>
<li>New <spanclass="argument">-F</span> option to <spanclass="commandname">flac</span> to continue decoding in spite of errors.</li>
<li>Correctly set granulepos in Ogg packets so seeking Ogg FLAC streams will be easier.</li>
<li>New <ahref="format.html#metadata_block_vorbis_comment">VORBIS_COMMENT</a> metadata block for tagging with Vorbis-style comments.</li>
<li>Vastly improved <spanclass="commandname">metaflac</span>, now with many editing and tagging options.</li>
<li>Partial id3v1 support in Winamp plugins.</li>
<li>Updated Winamp 3 plugin.</li>
<li>Note: new semantics for -P option in <spanclass="commandname">flac</span>.</li>
<li>Note: removed -R option in <spanclass="commandname">flac</span>.</li>
</ul>
New library features:
<ul>
<li>Previously mentioned decoder speedup in <spanclass="commandname">libFLAC</span>.</li>
<li>New metadata interface to <spanclass="commandname">libFLAC</span> for manipulating metadata in FLAC files.</li>
<li>New <spanclass="commandname">libFLAC++</span> API, an object wrapper around <spanclass="commandname">libFLAC</span>.</li>
<li>New <ahref="format.html#metadata_block_vorbis_comment">VORBIS_COMMENT</a> metadata block for tagging with Vorbis-style comments.</li>
<li>Customizable metadata filtering by type in decoders.</li>
<li>Stream encoder can take an arbitrary list of metadata blocks, instead of just one SEEKTABLE and/or PADDING block.</li>
</ul>
Bugs fixed:
<ul>
<li>Fixed bug with using pipes under Windows.</li>
<li>Fixed several bugs in the plugins and made them more robust in general.</li>
<li>Fixed bug in <spanclass="commandname">flac</span> where decoding to WAVE of a FLAC file with 0 for total_samples in the STREAMINFO block yielded a WAVE chunk of 0 size.</li>
This release is only to fix a bug that was causing some of the plugins to crash sporadically. It can also affect <spanclass="commandname">libFLAC</span> users that reuse one file decoder instance for multiple files
<li>Support for Ogg-FLAC, i.e. <spanclass="commandname">flac</span> can now read and write FLAC streams using Ogg as the transport layer.</li>
<li>New Winamp 3 plugin based on the Wasabi Beta 1 SDK.</li>
<li>New utilities for adding FLAC support to the Monkey's Audio GUI (see <ahref="documentation_tasks.html#monkey">how</a>).</li>
<li>Mac OS X support. The download area now contains an OS X binary release.</li>
<li>Mingw32 support.</li>
<li>Better handling of MS-specific 'fmt' chunks in WAVE files.</li>
</ul>
New features for developers:
<ul>
<li>Added a SeekableStreamDecoder layer between StreamDecoder and FileDecoder. This makes it easier to use libFLAC in situations where files have been abstracted away. See the latest <ahref="api/index.html">documentation</a> for more. The interface for the StreamDecoder and FileDecoder remain the same and are still binary-compatible with libFLAC 1.0.</li>
<li>Drastically reduced the stack requirements of the encoder.</li>
</ul>
Bug fixes:
<ul>
<li>Fixed a serious bug with <spanclass="commandname">flac</span> and raw input where the encoder was trying to rewind when it shouldn't, which would add 12 junk samples to the encoded file. This was not present in WAVE encoding.</li>
<li>Fixed a minor bug in <spanclass="commandname">libFLAC</span> with setting the file name to stdin on a file decoder.</li>
<li>Fixed a minor bug in <spanclass="commandname">libFLAC</span> where multiple calls to setting the file name on a file decoder caused leaked memory.</li>
<li>Fixed a minor bug in <spanclass="commandname">metaflac</span>, now correctly skips an id3v2 tag if present.</li>
<li>Fixed a minor bug in <spanclass="commandname">metaflac</span>, now correctly skips long metadata blocks.</li>
It's finally here. There are a few new features but mostly it is minor bug fixes since 0.10:
<ul>
<li>New '--sector-align' option to <spanclass="commandname">flac</span> which aligns a group of encoded files on CD audio sector boundaries.</li>
<li>New '--output-prefix' option to <spanclass="commandname">flac</span> to allow the user to prepend a prefix to all output filenames (useful, for example, for encoding/decoding to a different directory).</li>
<li>Better WAVE autodetection (doesn't rely on ungetc() anymore).</li>
This is probably the final beta. There have been many improvements in the last two months:
<ul>
<li>Both the encoder and decoder have been significantly sped up. Aside from C improvements, the code base now has an assembly infrastructure that allows assembly routines for different architectures to be easily integrated. Many key routines have now have faster IA-32 implementations (thanks to Miroslav).</li>
<li>A new metadata block <ahref="format.html#def_SEEKTABLE">SEEKTABLE</a> has been defined to hold an arbitrary number of seek points, which speeds up seeking within a stream.</li>
<li><spanclass="commandname">flac</span> now has a command-line usage similar to 'gzip'; make sure to see the latest <ahref="documentation.html">documentation</a> for the new usage. It also attempts to preserve the input file's timestamp and permissions.</li>
<li>The -# options in <spanclass="commandname">flac</span> have been tweaked to yield the best compression-to-encode-time ratios. The new default is -5.</li>
<li><spanclass="commandname">flac</span> can now usually autodetect WAVE files when encoding so that -fw is usually not needed when encoding from stdin.</li>
<li>The WAVE reader in <spanclass="commandname">flac</span> now just ignores (with a warning) unsupported sub-chunks instead of aborting with an error.</li>
<li>Added an option '--delete-input-file' to <spanclass="commandname">flac</span> which automatically deletes the input after a successful encode/decode.</li>
<li>Added an option '-o' to <spanclass="commandname">flac</span> to force the output file name (the old usage of "flac - outputfilename" is no longer supported).</li>
<li>Changed the XMMS plugin to send smaller chunks of samples (now 512) so that visualization is not slow.</li>
<li>Fixed a bug in the stream decoder where the decoded samples counter got corrupted after a seek.</li>
<li>FLAC's sync code has been lengthened to 14 bits from 9 bits. This should enable a faster and more robust synchronization mechanism.</li>
<li>Two reserved bits were added to the frame header.</li>
<li>A CRC-16 was added to the FLAC frame footer, and the decoder now does frame integrity checking based on the CRC.</li>
<li>The format now includes a new subframe field to indicate when a subblock has one or more 0 LSBs for all samples. This increases compression on some kinds of data.</li>
<li>Added two options to the analysis mode, one for including the residual signal in the analysis file, and one for generating gnuplot files of each subframe's residual distribution with some statistics. See the latest <ahref="documentation.html#analysis_options">documentation</a>.</li>
<li>XMMS plugin now supports 8-bit files.</li>
<li>Fixed a bug in the Winamp2 plugin where the audio sounded garbled.</li>
<li>Fixed a bug in the Winamp2 plugin where Winamp would hang sporadically at the end of a track (c.f. <ahref="http://sourceforge.net/projects/flac/&atid=113478">bug #231197</a>).</li>
<li>Created a new utility called <spanclass="commandname">metaflac</span>. It is a metadata editor for .flac files. Right now it just lists the contents of the metadata blocks but eventually it will allow update/insertion/deletion.</li>
<li>Added two new metadata blocks: PADDING which has an obvious function, and APPLICATION, which is meant to be open to third party applications. See the <ahref="format.html#def_APPLICATION">latest format docs</a> for more info, or the new <ahref="id.html">id registration page</a>.</li>
<li>Added a <spanclass="argument">-P</span> option to <spanclass="commandname">flac</span> to reserve a PADDING block when encoding.</li>
<li>Added support for 24-bit files to <spanclass="commandname">flac</span> (the FLAC format always supported it).</li>
<li>Started the Winamp3 plugin.</li>
<li>Greatly expanded the test suite, adding more streams (24-bit streams, noise streams, non-audio streams, more patterns) and more option combinations to the encoder. The test suite runs about 30 streams and over 5000 encodings now.</li>
<li>Fixed a bug in <spanclass="commandname">libFLAC</span> that happened when using an exhaustive LPC coefficient quantization search with 8 bps input.</li>
<li>Fixed a bug in <spanclass="commandname">libFLAC</span> where the error estimation in the fixed predictor could overflow.</li>
<li>Fixed a bug in <spanclass="commandname">libFLAC</span> where LPC was attempted even when the autocorrelation coefficients implied it wouldn't help.</li>
<li>Reworked the LPC coefficient quantizer, which also fixed another bug that might occur in rare cases.</li>
<li>Really fixed the '-V overflow' bug (c.f. <ahref="http://sourceforge.net/p/flac/bugs/5/">bug #231976</a>).</li>
<li>Fixed a bug in <spanclass="commandname">flac</span> related to the decode buffer sizing.</li>
</ul>
FLAC is very close to being ready for an official release. The only known problems left are with the Winamp plugins, which should be fixed soon, and pipes with MSVC.
The encoder is now much faster. The -m option has been sped up by 4x and -r improved, meaning that in the default compression mode (-6), encoding should be at least 3 times faster. Other changes:
<ul>
<li>Some bugs related to <spanclass="commandname">flac</span> and pipes were fixed</li>
<li>A "loose mid-side" (<spanclass="argument">-M</span>) option to the encoder has been added, which adaptively switches between independent and mid-side coding, instead of the exhaustive search that <spanclass="argument">-m</span> does.</li>
<li>An analyze mode (<spanclass="argument">-a</span>) has been added to <spanclass="commandname">flac</span>. This is useful mainly for developers; currently it will dump info about each frame and subframe to a file. It's a text file in a format that can be easily processed by scripts; a separate analysis program is in the works.</li>
<li>The source now has an autoconf/libtool-based build system. This should allow the source to build "out-of-the-box" on many more platforms.</li>
This is the first beta version of FLAC. Being beta, there will be no changes to the format that will break older streams, unless a serious bug involving the format is found. What this means is that, barring such a bug, streams created with 0.5 will be decodable by future versions. This version also includes some new features:
<ul>
<li>An <ahref="http://userpages.umbc.edu/~mabzug1/cs/md5/md5.html">MD5 signature</a> of the unencoded audio is computed during encoding, and stored in the Encoding metadata block in the stream header. When decoding, <spanclass="commandname">flac</span> will now compute the MD5 signature of the decoded data and compare it against the signature in the stream header.</li>
<li>A test mode (<spanclass="argument">-t</span>) has been added to <spanclass="commandname">flac</span>. It works like decode mode but doesn't write an output file.</li>
<li>This version fixes a bug in the constant subframe detection. More importantly, a verify option (-V) has been added to <spanclass="commandname">flac</span> that verifies the encoding process. With this option turned on, <spanclass="commandname">flac</span> will create a parallel decoder while encoding to make sure that the encoded output decodes to exactly match the original input. In this way, any unknown bug in the encoder will be caught and <spanclass="commandname">flac</span> will abort with an error message.</li>
