Debian Codegraph @

NEWS

raw
1 1.33.6
2 ------
3 - mpg123
4 -- Prepare for const-returning strchr().
5 -- Hide seq_len debugging counter in non-debug mode (handle github PR 20).
6 -- Fix memory leak with --network internal due to inverted NULL check in
7 net123_close_internal() (handle never NULL in practice, though; bug 386).
8 - mpg123, out123: Fix strrchr() usage to be more const and correct under C99 as well
9 as C23 (handles PR 19 on the github mirror).
10 - mpg123-strip: Also use largefile API properly using mpg123config.h, but without
11 actual effect at least on Linux/x86. It is cleaner that way, though.
12 - ports/cmake: Yasm is not neeeded for Clang on MSVC (github PR 23, thanks to Alexander
13 Lobakin).
14 - libmpg123: Remove unused loop variable in layer2 left over from runtime table
15 elimination (32 bit mmx/sse code, prompted by github PR 22).
16
17 1.33.5
18 ------
19 - mpg123: Fix generic control mode for largefile-sensitive builds, where 32 bit
20 off_t was used with mpg123 API calls expecting 64 bit off_t.
21 I am appalled that it took a user on 32 bit ARM and a specific https stream
22 to notice this (bug 385, regression since 1.32.0).
23 The security impact of this could be serious, with memory corruption including
24 segfault being observed.
25 - mpg123-id3dump, out123: Enable 64 bit offset usage on largefile-sensitive
26 platforms (regression since 1.32.0).
27 - libmpg123:
28 -- Announce support for shadow stack / IBT in x86-64 assembly.
29 -- Also announce PAC/BTI for non-accurate neon64 (aarch64) synth.
30 - libout123: Add a safeguard to ensure variable-length records from buffer
31 communication are always zero-terminated.
32 - libsyn123: Use union work buffer to avoid casts that may look like breaking
33 strict aliasing.
34
35 1.33.4
36 ------
37 - mpg123: In terminal control, ignore 7-bit escape sequences to avoid spurious
38 actions, e.g. when hitting cursor keys. Inspired by Peter Tirsek.
39 - ports/cmake: Avoid possibly conflicting use of SIZEOF_OFF_T CMake variable
40 when embedding mpg123 with other projects using cmake and different off_t
41 semantics. (bug 382)
42
43 1.33.3
44 ------
45 - libmpg123:
46 -- Fix build with newer toolchain on Android by using __ANDROID__ macro (bug 380).
47 -- Consolidate and more consistently use .rodata switch in macro.
48
49 1.33.2
50 ------
51 - libmpg123:
52 -- Do not modify raw ID3v2 data while parsing (bug 379). This used to be fine
53 before MPG123_STORE_RAW_ID3 got introduced. Not anymore.
54
55 1.33.1
56 ------
57 - INSTALL updated with hints for Windows, mainly.
58 - Finally formatted README. Maybe do it in Markdown sometime, as that's
59 the fashion.
60 - build:
61 -- The ports/cmake only installs manpages for BUILD_PROGRAMS now.
62 -- The configure gives better hint if pkg-config was missing during generation
63 (bug 378).
64 - mpg123:
65 -- Replace usage of signal() in terminal code with our sigaction() wrapper
66 to fix repeated handling for --sigusr1 and --sigusr2, which got subtly
67 broken on Linux + glibc by a feature test macro change in mpg123 1.31.
68
69 1.33.0
70 ------
71 - mpg123
72 -- Fix printout of filenames at end (convert/limit text encoding).
73 -- Treat HTTP header encoding as unknown/ASCII and formally convert to UTF-8.
74 -- Make --continue mode work with --random.
75 -- Handle possible failure of __wgetmainargs on Windows (bug 375).
76 - mpg123-id3dump: Fix up command line arg handling for Windows.
77 - out123
78 -- Finally give zero exit code when generating sounds, not indicating
79 spurious failure.
80 - build:
81 -- Use CCASFLAGS for assembler tests, to enable builds that enable instruction
82 sets that way (bug 377).
83 -- PIC for compat libs (convenience libs used during build) only if building
84 shared libs (github PR 17 by Wouter Wijsman).
85 - compat:
86 -- Map strtok use to strtok_r or strtok_s (MS platforms), if possible.
87 users only in control_generic and libout123 so far. Out123 itself uses mytok.
88 Shall fix bug 376 (build with MSVC again).
89 -- Enable build on PSP by merging in the hotfix of opmitting signal code
90 (github PR 18 by Wouter Wijsman).
91 - libout123
92 -- modules/win32: Align waveOutGetDevCapsA to WAVEOUTCAPSA, in anticipation
93 of some UNICODE change.
94 - libmpg123
95 -- API version 49 with added mpg123_open_handle64(), mpg123_open64(), and
96 mpg123_open_fixed64() that are not subject to largefile renaming. This
97 means you can still access internal I/O with MPG123_PORTABLE_API. The
98 code has been there before, anyway.
99 -- With MPG123_PORTABLE_API, mpg123_open_handle() is hidden now (use
100 mpg123_open_handle64() instead).
101 -- more silence on errors (sideband limit message)
102
103 1.32.10
104 ------
105 - scripts/tag_lyrics.py: fix for python3 (thanks to cclauss,
106 github PR 16)
107 - libout123: Use strtok_r() to avoid conflicts multithreaded contexts
108 (both sides should avoid plain strtok()! Debian-bug 1089543).
109 - libmpg123:
110 -- Un-break DLL builds that need I/O functions defined in libmpg123.c
111 (like mpg123_open(), bug 374).
112 - ports/cmake: More fixup to also produce .pc files with Libs.private.
113
114 1.32.9
115 ------
116 - libmpg123:
117 -- enable 64 bit offset path for MSVCRT and avoid warnings about
118 MS's game about POSIX API with and without underscores (bug 373).
119 -- Increase the library patchlevel, as was forgotten on previous
120 release. Now you can check for distversion >= 1.32.8 or
121 mpg123 libversion >= 48 patchlevel 3 to see if you're vulnerable
122 to CVE-2024-10573.
123
124 1.32.8
125 ------
126 - libmpg123:
127 -- Add sections to assembly to support PAC/BTI code
128 for aarch64 (-mbranch-protection variants), thanks to Bill Roberts
129 (github PR 15).
130 -- Prevent premature application of header info into decoding structure,
131 at worst having triggered out-of-bounds writes of decoded PCM data
132 (bug 322, again). The bug got named CVE-2024-10573.
133 - out123: Show --quiet in --longhelp.
134
135 1.32.7
136 ------
137 - ports/cmake: Work around bug in CMake that does not detect FPU on
138 Apple ARM CPUs (github PR 14).
139 - Fix some laziness (func() to func(void)) for standards conformance.
140
141 1.32.6
142 ------
143 - build: Detect forced 64 bit offsets on a dual-mode system that used
144 to default to 32 bits and drop ambiguous suffix-less symbols in that
145 case. This avoids subtle ABI breakage (causing memory corruption)
146 with existing binaries and instead has them fail during runtime linking.
147 You trigger that when having -D_FILE_OFFSET_BITS=64 in your compiler
148 flags during mpg123 build.
149
150 1.32.5
151 ------
152
153 - build:
154 -- CMake port uses CFLAGS for pulse/jack/tinyalsa properly now (bug 366).
155 -- CMake port links libsyn123 with libm now (bug 370).
156 - libmpg123:
157 -- Fix --enable-portable (no usage of LFS_WRAP_NONE, bug 368).
158 -- Fix dct36 wrapper usage for x86-64 and NEON. Stupid (bug 367) and
159 also avoid returning void.
160 -- Make ARM builds work with nagging (missing feature macros for std=c99).
161
162 1.32.4
163 ------
164 - build:
165 -- Reorganize shared headers, API headers into src/include.
166 -- Use relative include paths, avoiding internal directories in
167 CPPFLAGS except for config.h.
168 -- Group C99 feature checks and make several standard headers
169 mandatory.
170 -- Get rid of SIZE_P, OFF_P and friends.
171 -- Only enforce dummy module together with libout123, to be able
172 to build individual modules using --disable-components logic.
173 - out123:
174 -- added --libversion
175 - libmpg123:
176 -- Avoid indirect branches into the assembly routines by using C wrappers
177 also for dct36, relieving us of the need to care for bti / endbr
178 instructions for control flow integrity.
179
180 1.32.3
181 ------
182 - ports/cmake: Only enable modules with GetThreadErrorMode() on Windows.
183 - compat: Define EOVERFLOW for ancient Windows toolchains.
184 - libmpg123, libsyn123: always ifdef LFS_LARGEFILE_64 (not just if)
185 - libsyn123: re-introduce _32 wrappers in addition to suffix-less ones
186 (regression from 1.31, bug 363)
187
188 1.32.2
189 ------
190 - libmpg123: Re-introduce _64 symbols on native 64 bit offset platforms. This
191 was a regression since 1.31 series. Sorry, too much cleanup, not enough
192 testing.
193 - build:
194 -- Better O_LARGEFILE logic, avoiding redefintion.
195 - ports/cmake:
196 -- Require C99 (bug 360, among other points, thanks to Ozkan Sezer).
197 -- Fix broken O_LARGEFILE logic (bug 360).
198 -- Typo fix and cleanup, also manual SSE switch for Android on old x86 (bug 359).
199
200 1.32.1
201 ------
202 - Include man pages again in tarball and install. We cannot avoid the empty
203 man directory when disabling programs with autoconf.
204 - Fix signal handler prototype, avoiding some justified warnings.
205 - ports/cmake:
206 -- Include CheckTypeSize, which seems to be needed sometimes (bug 357).
207 -- Avoid O_LARGEFILE redefinition, logic closer to autoconf.
208
209 1.32.0
210 ------
211 - build:
212 -- Move version handling out of configure.ac to ease other build systems.
213 -- Include "fmt123.h" instead of <fmt123.h> in main API headers to make it more likely the
214 correct one is included (at least gcc picks the one in the same directory as the
215 including header first).
216 -- All headers are build-independent now.
217 -- Fix build for picky linkers by avoiding definition of wrap_getcpuflags() where it is not
218 used (spurious linker error to non-exitent getcpuflags(), bug 353).
219 -- Handle deprecation of C99 detection macro in autoconf 2.70.
220 -- No use of AC_SYS_LARGEFILE anymore for explicit handling and differing choice for
221 the libraries and frontend programs.
222 -- Added --enable-portable and --disable-largefile to configure, removing the other
223 largefile-related options.
224 -- Added --disable-components --enable-libmpg123 to only build libmpg123 (and likewise
225 --enable-libout123, --enable-libout123-modules, --enable-libsyn123) to autoconf
226 build. CMake build has something similar with BUILD_PROGRAMS and BUILD_LIBOUT123,
227 which leave only libmpg123 and libsyn123 if disabled). (bug 351)
228 -- Consistent formatting of ./configure --help with AS_HELP_STRING().
229 - ports/Sony_PSP: removed
230 - mpg123:
231 -- Added --libversion.
232 -- Added proper A-B looping with terminal control key 'o', renamed
233 --pauseloop to --presetloop.
234 -- Really get rid of mpg123_position() usage. (It was all lies before!)
235 -- Fix terminal progress info when seeking in stopped mode (1.31 regression).
236 -- Patch up interaction of output buffer with generic remote control, adding
237 non-interruptible drain after P 3, and dropping buffer on QUIT.
238 -- Uppercase some generic control replies for consinstency: SILENCE, PROGRESS,
239 MUTE, UNMUTE
240 - libmpg123, libout123, libsyn123:
241 -- Bumped API version for version query functions.
242 -- Replaced nearly all symbol renames with explicit INT123_ prefix declarations
243 (intsym.h close to empty now).
244 - libout123:
245 -- Add sleep builtin output module (silent, but proper timing).
246 - libsyn123:
247 -- Introduced SYN123_PORTABLE_API for an API without off_t and ssize_t
248 (see NEWS.libsyn123).
249 - libmpg123:
250 -- Internal I/O using explicit largefile support via off64_t, lseek64, fallback
251 to plain 32 bit off_t.
252 -- Added explicit 64 bit API with 64 suffix (mpg123_tell64(), not mpg123_tell_64()).
253 This allows full avoidance of ambiguus off_t. The API is always using 64 bit
254 integers, regardless of internal implementation. (bug 344)
255 -- Introduced MPG123_PORTABLE_API for an API subset without off_t and
256 ssize_t.
257 -- Made mpg123_seek() and friends ignore offset sign for SEEK_END (always seeking
258 towards beginning, assuming negative offset) to make lseek()-conforming usage
259 possible. Seeking beyond the end never made sense, so no loss of valid functionality.
260 - Overall use of INT123_strerror(), trying to use thread-safe strerror_l() if possible.
261
262 1.31.3
263 ------
264
265 - build:
266 -- Fix --disable-8bit.
267 -- Fall back to generic decoder if no yasm for MSVC (bug 346).
268 -- Fix some pedantic compiler warnings, avoid breaking libtool wrappers.
269 - mpg123:
270 -- Fix verbose position printout for new resampling outside libmpg123 (where
271 output rate differs from decoding rate).
272 - libsyn123:
273 -- Fix reconfiguration of resampler to avoid double free when reducing
274 decimator stages to zero (bug 350). Thanks to Youngseok Choi for reporting
275 this fuzzed issue.
276
277 1.31.2
278 ------
279 - Fix build --with-network=internal only (configure logic error, bug 348).
280 - Fix OS/2 build with getaddrinfo() (which may support IPv6 eventually, thanks
281 to Dave Yeo).
282
283 1.31.1
284 ------
285 - Fix largefile aliases for the case of a largefile-insensitive build
286 that still does define _FILE_OFFSET_BITS from the outside
287 (sys/feature_tests.h on Illumos).
288
289 1.31.0
290 ------
291
292 - mpg123:
293 -- Finally make terminal control work on Windows, for real. Building it
294 was broken in 1.30.x.
295 -- The --control / -C switch will make mpg123 abort now if terminal
296 control cannot be enabled.
297 -- Revert to internal network code for plain HTTP to ensure continued
298 support for original shoutcast servers that do not talk proper HTTP.
299 External backends are built at the same time and can be enforced using
300 --network <backend>.
301 -- Try-witout-port for internal network code is gone. We do not need to
302 keep each ancient hack for specific hosts.
303 -- Handle redirections independently of the backend behind net123.
304 -- Set proxy environment variables when --proxy is specified, for net123
305 backends to use.
306 -- Continue reading for long commands in generic control, avoiding
307 unnecessary unfinished command errors.
308 -- Change error message from 'unknown command' to
309 'unknown command with arguments' to avoid confusion why 'help foo'
310 is unknown, as opposed to 'help'.
311 -- Reduce CPU load while just waiting for terminal input
312 (thanks to bolshoytoster on github).
313 -- Condense terminal control help output and excessive vertical whitespace
314 in printouts (inspired by Volkmar Klatt).
315 -- Fix interaction of pause (looping) with buffer, adding --pauseloop
316 to set the loop interval.
317 -- Numeric option arguments are strictly checked now for conversion errors.
318 This also catches -devbuffer, which was interpretd as -d 0 before. This
319 also applies to out123.
320 - libout123:
321 -- Add same interruption handling to out123_write() as to
322 unintr_write(), adding EAGAIN to fix bug 342 (thanks to Steffen Nurpmeso)
323 for certain ALSA setups.
324 -- Add --devbuffer support to win32 output and change default to 0.25 seconds.
325 -- Fix race condition to deadlock on buffer_sync_param() where parameters after
326 the command byte got read as more commands. This got triggered easily by
327 using the pause key in terminal mode with buffer (which was discouraged
328 before because of buffer flushing). Generally, changing parameters with
329 active buffer process was dangerous since libout123 entered the scene.
330 - some build fixes for compiler pickyness
331 - Disable largefile renames also for non-sensitive POSIX systems
332 (in some distant future, the alias symbols could go away, then …
333 bug 330).
334 - Fix Android NDK x86 builds with GLOBAL_VAR_PTR use in assembly (bug 345).
335
336 1.30.2
337 ------
338 - Only use EWOULDBLOCK if the macro is defined (FreeBSD misses it for
339 _POSIX_SOURCE, bug 339).
340
341 1.30.1
342 ------
343 - mpg123:
344 -- Show stderr of network helpers in -vvv mode.
345 -- Use curl --http0.9, if available, to support shoutcast v1 streams
346 without wget (wget not needing such switch, yet).
347 -- Support file:// URLs for local access as was intended with the last
348 release.
349 -- Give more helpful error message if neither wget nor curl are usable, also
350 allow error messages from curl to appear when not --quiet.
351 -- Update the man page.
352
353 1.30.0
354 ------
355 - build:
356 -- Use dummy as default module when no other outputs are enabled. This also
357 fixes a non-module build with just the dummy (bug 333).
358 -- Use CMAKE_CURRENT_SOURCE_DIR in CMake build to help nested use (bug 335).
359 -- some updates for OS/2 support (fixing up stdin playing, for example)
360 - mpg123:
361 -- new network backend using external tools/libraries to also support HTTPS
362 -- old network backend changed to use h_addr_list[0] instead of h_addr
363 -- terminal control keys now case-sensitive (fixing smal/big pitch controls)
364 -- additional terminal control keys for simple equalizer control (A/a for bass,
365 J/j for mids, N/n for treble, e for reset, E for printout)
366 -- terminal volume control now in decibel steps and bounded to +/- 60 dB
367 -- terminal control now also with audio from stdin (bug 338) via
368 /dev/tty or ctermid()
369 -- terminal control also available for OS/2 and Windows platforms
370 -- re-print tag info on decrease of terminal width for a bit less mess
371 -- always print an empty line after tag info for cleaner appearance
372 -- print lyrics also to stderr
373 -- remote control API v10 with "@P 3" as additonal message on track end
374 -- also added PROGRESS command as opposite of SILENCE
375 -- fix some verbosity, tweak help for --icy-interval
376 -- added --auth-file
377 -- also obscure argument to --auth for others
378 -- Cygwin/MinGW: Provide _win32_utf8_wide and _win32_wide_utf8 unconditionally.
379 It is needed by the WASAPI plugins, the underlying conversion functions
380 should be present since Windows 2000. Fixes WASAPI support on Cygwin.
381 Also needed for new network code.
382 - libout123:
383 -- pulse: initialize more error codes to avoid bogus error messages
384 -- os2: considerable fixup for proper writes of full buffers avoiding
385 nasty effects from the ... special audio system, more cleanup still
386 nice-to-have, but still lacking
387
388 1.29.4
389 ------
390 - libmpg123:
391 -- Saturate reader file position at off_t limit to satisfy
392 undefined behaviour checkers.
393 -- Avoid harmless unitialized value in ID3v1 check (filepos, later being
394 set before actual use).
395 - libout123:
396 -- Build fix for win32_wasapi output for predefined _WIN32_WINNT (bug 329),
397 thanks to Vincent Torri.
398
399 1.29.3
400 ------
401 - libmpg123: Catch more NULL pointer arguments in LFS wrappers
402 (most prominently: mpg123_feedseek(), bug 328).
403 - mpg123:
404 -- Fix regression that did _not_ enable --remote-err on -s anymore.
405 -- Fix typos in man page (thanks to Naglis Jonaitis).
406 -- Drop mixed-up value limits on remote control SEQ command. It is up to you
407 if you want to distort your sound.
408 -- Add note about equalizer frequency bands to man page.
409 - build: add BUILD_PROGRAMS option to ports/cmake
410
411 1.29.2
412 ------
413 - libmpg123: Fix non-live-decoder safeguard for mpg123_framebyframe_decode()
414 (was a no-op in practice, bug 324).
415
416 1.29.1
417 ------
418 - mpg123:
419 -- Keep default output encoding of s16 for raw and file outputs
420 also with the new resampler. This reverts the unintentional change in
421 1.26.0 of switching to f32 for forced output rate unless the NtoM
422 resampler is selected. In any case, you should make sure to specify
423 your desired --encoding if you depend on it.
424 -- Catch error in indexing (mpg123_scan() return value was ignored
425 before, bug 322).
426 - mpg123-strip: Lift the resync limit, as it should be to clean up really
427 dirty streams.
428 - mpg123-id3dump: Also lift resync limit for the same reasons.
429 - libout123: fix reporting of device property flags for buffer
430 - libmpg123: more safeguarding against attempts to decode if decoder
431 setup failed and user ignored the returned error code (bug 322)
432
433 1.29.0
434 ------
435 - build: added --enable-runtime-tables
436 - libmpg123:
437 -- Float deocder runtime table computation is back as option,
438 based on suggestion and initial patch by Ethan Halsall for a smaller
439 download size of the wasm decoder built from libmpg23. This only
440 trims the size of the binary on disk (network), for runtime
441 overhead and a bit of uneasyness about concurrency during table
442 computation, which happens implicitly on handle initialization,
443 only guarded by an integer flag. This does _not_ revive
444 mpg123_init().
445 -- The ID3v2 UTF-16 BOM check is now a straight-on loop and not a
446 recursive function.
447
448 1.28.2
449 ------
450 - libout123
451 -- Complete the fix for bug 314, reopening the device after format
452 setup failure.
453
454 1.28.1
455 ------
456 - build:
457 -- Explain --with-default-audio in configure help better.
458 -- Fix build of arm_fpu (regression of configure reorg).
459 -- Re-introduce AC_PROG_C_C99 macro for autoconf 2.69, it's only
460 obsolete after that.
461 -- Un-break CMake build for botched move of CheckCPUArch.c.in (bug 315).
462 -- Avoid conflict of warning macro with MSVC pragmas in two places. Also
463 fix UWP build with strerror check and move down inclusion of intsym.h
464 (bug 316).
465 -- Disable libout123 (and mpg123, out123) on UWP with cmake to get at least
466 the decoder lib built (317).
467 -- Hack around CMake bug(?) with QUERY_HAS_FPU to make ports/cmake also
468 work in MinGW (bug 318).
469 - libmpg123:
470 -- Make mpg123.h.in usable again with MPG123_NO_CONFIGURE,
471 for external uses (bug 313).
472 -- Use predefined MPG123_API_VERSION in mpg123.h.in for the same.
473 -- Better handle the ssize_t situation via typedef mpg123_ssize_t, less
474 likely to be broken in future MSVC versions.
475 -- Fix an integer constant definition for the most negative 32 bit numnber to
476 avoid justified compiler complaints.
477 - libsyn123:
478 -- More support for MPG123_NO_CONFIGURE.
479 -- Optionally use predefined SYN123_API_VERSION in syn123.h.in for the same.
480 -- Add a cast to silence integer sign warning for offset in muloffdiv64()
481 (bug 317)
482 - libout123:
483 -- Pulse module advertises wider format support now, not
484 just s16. This makes mpg123 -e s24 work with it, not just out123.
485 -- Optionally use predefined OUT123_API_VERSION in out123.h.in for
486 non-configure use.
487 -- Fix sndio output to properly query device format support and get
488 default fomat on FreeBSD (bug 314).
489
490 1.28.0
491 ------
492 - build:
493 -- Fix up the build to actually build all library objects with libtool
494 consistently, also ensuring no pointless static archives for output
495 modules.
496 -- Adapted things to autoconf 2.71, requiring 2.69 now (the latter tested
497 on Debian, with their patches).
498 -- Improved configure to be more useful --with-default-audio to define
499 the search order, fix static build for --with-audio being a list
500 (just choosing the first one).
501 -- Ensure consistent use of LINK_MPG123_DLL in headers.
502 - build (ports/cmake):
503 -- Thanks to Evgeni Poberezhnikov for working with us on that.
504 -- Fix up ports/cmake to really work in MSVC also for users of the lib
505 (tested in vcpkg, bug 310).
506 -- Hardcode ports/cmake CPU detection for x64 and ARM as
507 CMAKE_SYSTEM_PROCESSOR is useless crap (bug 298 for real).
508 -- Add missing io.h for _setmode() MSVC warned about (bug 311).
509 -- Added BUILD_NO_LARGENAME define to be used by MSVC builds. Note that
510 an MSVC build of libmpg123 does not support 64 bit file offsets.
511 That would need more morting to the explicit API. Thanks to MS for
512 making off_t even more messy and less useful.
513 -- Added JACK output, fixed handling of compat_str there and in win32_wasapi.
514 - libsyn123: Fix syn123_mix() to actually do intermediate conversion
515 when input and output encoding are the same but non-float. This makes
516 out123 --mix work with s16 input and output, which is not that special!
517 - libmpg123: Fix misguided handling of part2_3_length checks in
518 III_get_scale_factors_1() and III_get_scale_factors_2() which invalidated
519 decoding of a mono source encoded as ms+i-stereo (bug 312). This was
520 a regression introduced with version 1.25.7.
521 - libout123:
522 -- Print basic module loading errors only for last one in list.
523 This enables use of an output module search list that anticipates module
524 files not installed with the main package.
525 -- Fixes for win32_wasapi build with MSVC.
526
527 1.27.2
528 ------
529
530 (Trying some svn tag discipline: 1.27.1 has been tagged before, but
531 not released. Let's increment for any change.)
532
533 - Removed ports/Xcode, ports/cmake should handle that case.
534 - Ensure debug.h is included last where it matters to avoid
535 conflicts with debug/warning macros in system headers
536 (bug 308).
537 - Fix some debug/printf integer casts for 32 bit platforms (bug 309).
538
539 1.27.0
540 ------
541
542 - libmpg123:
543 -- Running on precomputed tables now, no need to call
544 mpg123_init() anymore. That and mpg123_exit() are both just
545 empty shells. You can omit them if you do not care about earlier
546 libmpg123. You can check for MPG123_API_VERSION >= 46.
547 -- Added API that avoids enums, mapped-to by default unless
548 MPG123_ENUM_API is defined.
549 - libout123:
550 -- Added API that avoids enums, mapped-to by default unless
551 MPG123_ENUM_API is defined.
552 -- Added device enumeration for win32, win32_wasapi, alsa, pulse.
553 This increments the output module ABI version to 3.
554 -- You can choose output devices now on Windows.
555 -- Changed default output module order to put pulse before alsa since
556 we now ensure that pulse is not inadvertedly started by the
557 autospawn feature. This improves the experience on desktop
558 systems with pulse where the alsa to pulse use causes glitches.
559 Note that on a modern Linux desktop (Ubuntu), you will not
560 escape an instance of pulseaudio being started, with even the
561 enumeration of the ALSA default device summoning the daemon.
562 If you _want_ sound daemon autospawn behaviour on other platforms,
563 you need to trigger it outside of libout123.
564 - examples: Update for dropped mpg123_init(), more sensible
565 copyright notes.
566 - out123:
567 -- safer limiting of maximum playback rate
568 -- Added --list-devices.
569 - mpg123:
570 -- Fix --continue output to print track_count+1 as continue
571 position after hitting the end of playlist. Makes scripts/conplay
572 go to back to the beginning again (regression in 1.24.0, bug 250).
573 -- Remote control API version 9 with @I { .. @I } wrapping of ID3 and
574 playlist display.
575 -- Added --list-devices.
576 -- Fix console printout on Windows.
577 -- Fix terminal control logic to better handle cases where stdin or
578 stderr is not a terminal, also avoid enabling control if you specify
579 stdin as input file.
580 - Updated debugging/warning/error message macros to include the function
581 name.
582
583 1.26.5
584 ------
585 - Add ./configure --enable-xdebug (for the resampler issue).
586 - Avoid denormals in the resampler by adding an alternating offset
587 (helps performance without -ffast-math, depending on platform).
588 - libmpg123:
589 -- Fix ID3v2 APIC parsing when frame length bit is set (bug 306).
590 -- Also handle the group flag (skip the group byte).
591 -- Also fix up frame flag handling for ID3v2.3. Did not crop up yet,
592 but it was just wrong. Impact was not detecting and bailing out on
593 compressed or encrypted frames properly.
594
595 1.26.4
596 ------
597 - Clarify seeking documentation regarding samples and PCM frames.
598 - Fix build on MorphOS (patch by Ozkan Sezer, bug 295).
599 - Fix cmake build to install fmt123.h.
600 - Some cmake build fixes, tinyalsa addition by Maarten (bug 299).
601 - libmpg123: explicitly handle some irrelevant corner cases in tabinit
602 (bug 279)
603
604 1.26.3
605 ------
606 - Fix accurate (--enable-int-quality) and s32 x86-64 assembly for Windows ABI
607 (thanks to Robert Kausch of fre:ac). The accurate code just crashed before.
608
609 1.26.2
610 ------
611 - Enable terminal control by default only when both input and output
612 are connected to a terminal. This avoids messing with terminal settings
613 when piping stderr to a pager, which takes over terminal input anyway,
614 while mpg123 still thinks it got control.
615 - Windows build fixes for UWP and without GetThreadErrorMode when not
616 building modules, thanks to Steve LHomme.
617 - Android build fix regarding off64_t use, thanks to Steve LHomme.
618 - More CMake build fixes thanks to David Callu (bug 290).
619 - Use PROG_LIBS for output modules, to reinstate not necessarily proper but
620 previous behaviour and fix FreeBSD port build (bug 291).
621 - Refine LFS support in libsyn123, avoiding architecture-dependent syn123.h
622 (debian bug 963205).
623
624 1.26.1
625 ------
626 - Fix cmake build by actually including the read_api_version file in the distro.
627 - Fix big-endian build, stupid omission of a variable declaration, semicolon
628 (bug 289).
629 - Silence a harmless warning for build without realtime priority.
630
631 1.26.0
632 ------
633
634 - Starting to intentionally use C99 in the codebase. API headers are still
635 supposed to be compatible to C89.
636 - There is a make check target now with some simple tests.
637 It is an open question how that should be developed in relation to the
638 external regression and compliance test suite.
639 - Finally silenced memory checkers about leaking memory from getlopt()
640 (main code overwriting values without freeing strdup() strings).
641 - AUTHORS now in UTF-8;-)
642 - CMake build files in ports/cmake, as an alternative to create MSVC
643 project files and the like (thanks to Vitaly Kirsanov)
644 - Default build with proper integer rounding (--enable-int-quality) now.
645 - Cygwin/midipix autoconf fixes (thanks to Redfoxmoon).
646 - Updated Windows build script, notably renaming .dll.def to .def.
647 Requires an argument now for build type, an optional one for
648 parallel make (not that useful on MinGW).
649 - Rework library dependency handling to avoid unnecessary linking for
650 lib*123. Also add proper Libs.private to .pc files to enable static
651 usage (especially on Windows with shlwapi).
652 - Updated support for OS/2 in the form of ArcaOS.
653 - Removed outdated Pascal port (ports/mpg123_.pas). There are others
654 out there.
655 - Updated man pages, been a while.
656 - mpg123:
657 -- Fixed-rate playback now prefers the libsyn123 resampler instead of NtoM
658 in libmpg123, see --resample parameter.
659 -- Drop --STDOUT (never properly implemented, use pipe to out123 instead).
660 -- Make --streamdump use unintr_write() to avoid inconsistencies.
661 -- Now sets non-zero exit code when any one track of the playlist
662 fails to either produce at least one frame of playback, if there
663 is data that should produce such (i.e. /dev/zero is bad, /dev/null is
664 fine). See man page for details.
665 -- Print out MPEG header info for each frame for mpg123 -vvvv.
666 -- Added --no-visual to disable cursor/inverse video games explicitly.
667 -- Clear progress bar before printing updated metadata within a stream.
668 -- Filter control/non-printable characters from user data printout, reduce
669 ID3v1 data to 7-bit ASCII (no way to know correct 8-bit encoding for sure).
670 This should cover bug 267.
671 -- Set MPG123_NO_PEEK_END when opening special file '-' (standard input).
672 That helps Windows where attempting to seek on the non-seekable stream
673 is undefined behaviour (bug 285).
674 -- Print errors in player code also for --quiet operation (just no messages
675 from the libraries).
676 -- Ignore ID3v1 once a Frankenstein stream was detected.
677 -- Prevent a cosmetic use-after-free in audio playback during program abortion just after
678 starting playback (prebuffer still in use, implication a blip of bad sound and
679 a complaining sanitizer).
680 -- Reformat audio capabilities table, more condensed, fits into 80 columns.
681 Forced rate on a separated line.
682 -- Make --pitch actually work, not just interactive changes. Pitching uses
683 a resampler now if a fixed output rate is specified.
684 -- Added --no-frankenstein.
685 -- Frameflags as long variable, 32 bits are needed since some time now.
686 - out123:
687 -- Document --STDOUT, make it more robust regarding fwrite() interruptions.
688 -- Removed the implicit phase shift that made generated waves exactly at
689 Nyquist freq non-silent, but made little sense overall.
690 -- Less high-frequency shifts to make waves fit into the table (not insisting
691 on even number of samples).
692 -- Option to work without wave table (setting the limit to zero).
693 -- Added --wave-direction to also enable backward time without phase shift.
694 -- Waves now generated by re-usable little synthethizer library dubbed
695 libsyn123.
696 -- Pink noise from libsyn123 added (using code from Phil Burk).
697 -- White noise from libsyn123.
698 -- Geiger-Mueller counter simulation from libsyn123.
699 -- Wave sweep generator from libsyn123.
700 -- Some rearrangement in help text.
701 -- Changed output of --test-encodings to list of encoding names
702 instead of raw bitmask value.
703 -- Added --endian, --inputend, and --byteswap.
704 - libsyn123:
705 -- Created the library to host some simple signal generators for testing
706 output.
707 -- It also hosts sample format conversions as a necessity to be able to
708 directly produce the format output devices need.
709 -- Well, also channel mixing while we're at it.
710 -- Oh, and a minimal-latency-and-reasonably-efficient resampler that only
711 took me over a year to figure out. I should write a paper about it.
712 - libout123:
713 -- Added out123_free() for the benefit of library wrappers. (bug 276)
714 -- Removed change of effective user ID in the WAV/RAW/AU/CDR writer.
715 This was intended as a safeguard to avoid creating files with root
716 priviledges. But: Other output modules still allowed root-level
717 access to various devices and files, so it was never safe to do
718 something awful like installing mpg123 with suid bit or configure
719 sudo to allow users to run mpg123 with arbitrary arguments.
720 You should treat out123_open just like the regular open(): You can
721 write to any file/device depending on your permissions.
722 -- Finally maybe fixed the damaged playback when using pulse hidden
723 behind the ALSA API (on Ubuntu, for example) by setting a high value
724 for device start threshold.
725 -- Fixed out_play() abortion logic to better detect fatal situations
726 (broken pipe). Needed on FreeBSD, while Linux buffers the issue away.
727 Should resolve bug 283.
728 -- Limit size of buffer block being written in out123_play to 16K, avoiding
729 unnecessary failure with ALSA at least.
730 -- Using SDL2 now if found. Output module code unchanged.
731 -- Added hex and txt (plain text) printout.
732 -- Eliminated spots where error messages would still be printed
733 also for OUT123_QUIET being effect.
734 -- Dummy output accepts any encoding now.
735 - libmpg123:
736 -- Added mpg123_open_fixed() to ease API for applications that just
737 want to decode well-behaved local files.
738 -- The user buffers for audio output data are now declared as void* for
739 mpg123_read(), mpg123_decode(), and mpg123_replace_buffer() to avoid
740 the useless need for casting your nice int16_t buffer to unsigned char
741 for decoding MPG123_ENC_SIGNED_16 data to it.
742 -- Added mpg123_free() for the benefit of library wrappers. (bug 276)
743 -- Add mpg123_format2() and mpg123_fmt2() supporting special value 0 for all
744 rates.
745 -- Fix changing of decoder (and output format along with that) after stream
746 opening. This was never recommened and only now should work at all.
747 -- Also mpg123_decode_frame() now sets return buffer to NULL and returned byte
748 count to zero in case of MPG123_NEED_MORE (or any other early abort).
749 -- MPG123_NEED_MORE not returned anymore for non-feeder streams. Got in
750 there for generic partial frame body reads, but was only intended for
751 feeder API.
752 -- Added mpg123_set_moreinfo() to support the Lame project's frame analyzer,
753 disabled by ./configure --disable-moreinfo.
754 -- Added optional storage and retrieval of raw ID3 data.
755 -- Fix skipping of ID3v2 footers (too much was attempted to be skipped). This
756 is of not much practical consequence as a tag with footer would appear on
757 the end of files anyway and files with ID3v2 tags at the end seem to be
758 rather rare.
759 -- Add mpg123_new_string() and mpg123_delete_string() to avoid confusion
760 about what mpg123_init_string() and mpg123_free_string() do.
761 -- Make mpg123_resize_string() terminate the string if shrinking (fill now
762 limited to new size).
763 -- Improve layer III frame parsing/error reporting for bad part2_3_length.
764 -- Fix crashing on stupidly low NtoM rates (exceeding downsampling factor 31).
765 This was only triggered by you specifying a forced sampling rate below 1550 Hz.
766 -- Do not remove CRC bits twice from possibly available bit reservoir.
767 This move needed recomputation of the layer3is reference data for 8 and
768 24 kHz. Old mpg123 is _wrong_ in the first few frames.
769 -- Generally more tight control and early bail out on reading bits of
770 frame data for all layers. This reduces the count of error messages
771 on badly damaged files a lot and feels a lot safer, too. Note that
772 we already silently returned zero bytes instead of actually over-reading
773 the frame buffer before, but now it happens with diagnostics and more
774 checks before it may happen.
775 -- Optionally enforce output endianess (big/little) away from native.
776 -- Fix build without error messages.
777 -- Fix build without gapless decoding.
778 -- Disable buffer when neither mmap nor shm functions detected (fixes
779 build for Android, thanks to vquicksilver).
780 -- Some support for extremely small streams (below 128 bytes). Those are
781 too short to contain anything useful besides some tiny metadata, but
782 serve to find/reproduce parser bugs.
783 -- Fix mpg123_read() for builds without feeder. It calls mpg123_decode()
784 without feeding input, which was disabled by mistake. The use of
785 mpg123_read() (instead of mpg123_decode_frame()) with mpg123_open()
786 was broken in feederless builds since those were fixed in version 1.15.
787 -- Fix ID3v2 parser logic for multiple ID3v2 tags being encountered in one
788 stream. New tags replace old data instead of appending to it when the
789 extended header update flag is not set (ID3v2.4). Update tags only
790 replace data that shall be unique. So far, I have never seen an update
791 tag in the wild, so the check for the flag is untested. The mechanism
792 of replacing parts of existing tag data has been tested, though.
793 Note that the updated libmpg123 also avoids a growing ID3 data structure
794 when repeatedly seeking back to the beginning in a file with disabled
795 seek index.
796 -- Eliminated a spots where error messages would still be printed
797 also for MPG123_QUIET being effect.
798 -- Added MPG123_NO_FRANKENSTEIN, MPG123_FLOAT_FALLBACK flags.
799 -- Now actually try floating point encoding if format matrix allows it
800 (can be disabled by unsetting MPG123_FLOAT_FALLBACK).
801 -- Added mpg123_feature2() that takes an int, as enums are not ABI-safe,
802 also added feature queries for floating point output.
803
804 1.25.13
805 -------
806
807 This is a bugfix release solely for bug 280 in the parser:
808
809 - libmpg123
810 -- Reset the flag for having a frame to decode before trying to
811 parse a new one. This prevents very unkind behaviour (crashes)
812 when combinging mpg123_scan() with decoding later on for damaged
813 streams that have a mixture of different MPEG versions.
814
815 1.25.12
816 -------
817
818 More credit to OSS-Fuzz. The ID3v2 parser code is not yet as hardened
819 as the actual MPEG decoder. The paranoid can disable it at build-time.
820 If you do not need it, this is a good idea, anyway: Code that is not
821 there, cannot be exploited. Speaking about exploits: The recent crop
822 of bugs trigger a denial of service (crash) worst-case, some invalid
823 ID3 data normally. Code injection maybe not totally ruled out (that one
824 write of a zero byte?), but does not seem easy. Update to be sure that
825 you are only suceptible to as of yet hidden bugs.
826
827 - libmpg123
828 -- Fix an out-of-bounds read of maximal two bytes for truncated RVA2 frames
829 (oss-fuzz-bug 15975). The earlier fix around the same location needed
830 one thought more. Actually, another though was needed, oss-fuzz-bug 16009
831 documents the incomplete fix.
832 -- Fix an invalid write of one zero byte for empty ID3v2 frames that demand
833 de-unsyncing (oss-fuzz-bug 16050).
834 -- Correct preprocessor syntax in mangle.h, no #error in a #define line.
835 (bug 273, thanks to nmlgc).
836 - Fix dynamic build with gcc -fsanitize=address (check for all dl functions
837 before deciding that separate -ldl is not needed).
838
839 1.25.11
840 -------
841
842 So, here is a number of bugs found by OSS-Fuzz. Credit to OSS-Fuzz
843 for the bunch, then.
844
845 - libmpg123:
846 -- Fix out-of-bounds reads in ID3 parser for unsynced frames.
847 (oss-fuzz-bug 15852)
848 -- Fix out-of-bounds read for RVA2 frames with non-delimited identifier.
849 (oss-fuzz-bug 15852)
850 -- Fix implementation-defined parsing of RVA2 values. (oss-fuzz-bug 15862)
851 -- Fix undefined parsing of APE header for skipping. Also prevent
852 endless loop on premature end of supposed APE header. (oss-fuzz-bug 15864)
853
854 1.25.10
855 -------
856
857 - libout123: Fix error messages beginning from OUT123_ARG_ERROR (bug 261).
858
859 1.25.9
860 ------
861 - mpg123: Fix --icy-interval handling to work with stream from stdin.
862 (curl | mpg123 --icy-interval=n -)
863 - libmpg123: Fix another invalid read and segfault on damaged (fuzzed)
864 files with part2_3_length == 0 (set maxband=1, pulled from upcoming 1.26.0).
865
866 1.25.8
867 ------
868 - mpg123:
869 -- Also disable cursor/video games for empty TERM (not just unset and dumb).
870 - libmpg123:
871 -- Accept changing mode extension bits when looking for next header for
872 detecting free-format streams (bug 257).
873 -- Fix compute_bpf() for free format streams (needed to estimate track
874 length and working fuzzy seeking in absence of an Info tag).
875
876 1.25.7
877 ------
878 - mpg123:
879 -- Do not play with cursor and inverse video for progress bar
880 when TERM=dumb.
881 -- Fix parsing of host port for numerical IPv6 addresses (just did
882 not work before, only for textual host names).
883 - libmpg123:
884 -- Proper fix for the xrpnt overflow problems by correctly
885 initialising certain tables for MPEG 2.x layer III. The checks that
886 catch the resulting overflow are still in place, but likely superfluous
887 now. Note that this means certain valid files would have been misdecoded
888 before, if anyone actually produced them. Thanks to Robert Hegemann for
889 the fix!
890 -- Silently handle granules with part2_3_length == 0, but
891 scalefac_compress != 0 (ignore the latter).
892
893 1.25.6
894 ------
895 - Hotfix for bug 255: Overflow reading frame data bits in layer II decoding.
896 Now, all-zero data is returned if the frame data is exhausted. This might
897 have a slight impact on performance, but not easily measurable so far.
898
899 1.25.5
900 ------
901 - Avoid another buffer read overflow in the ID3 parser on 32 bit platforms
902 (bug 254).
903
904 1.25.4
905 ------
906 - Better configure checks for i?86-apple-darwin (bug 253).
907 - libmpg123:
908 -- Prevent harmless call to memcpy(NULL, NULL, 0).
909 -- More early checking of ID3v2 encoding values to avoid bogus text being
910 stored.
911
912 1.25.3
913 ------
914 - libmpg123:
915 -- Better checks for xrpnt overflow in III_dequantize_sample() before each
916 use, avoiding false positives and catching cases that were rendered
917 harmless by alignment-enlarged buffers.
918
919 1.25.2
920 ------
921 - libmpg123:
922 -- Extend pow tables for layer III to properly handle files with i-stereo and
923 5-bit scalefactors. Never observed them for real, just as fuzzed input to
924 trigger the read overflow. Note: This one goes on record as CVE-2017-11126,
925 calling remote denial of service. While the accesses are out of bounds for
926 the pow tables, they still are safely within libmpg123's memory (other
927 static tables). Just wrong values are used for computation, no actual crash
928 unless you use something like GCC's AddressSanitizer, nor any information
929 disclosure.
930 -- Avoid left-shifts of negative integers in layer I decoding.
931
932 1.25.1: Hot Fuzz
933 -------
934 - libmpg123:
935 -- Avoid memset(NULL, 0, 0) to calm down the paranoid.
936 -- Fix bug 252, invalid read of size 1 in ID3v2 parser due to forgotten
937 offset from the frame flag bytes (unnoticed in practice for a long
938 time). Fuzzers are in the house again. This one got CVE-2017-10683.
939 -- Avoid a mostly harmless conditional jump depending on uninitialised
940 fr->lay in compute_bpf() (mpg123_position()) when track is not ready yet.
941 -- Fix undefined shifts on signed long mask in layer3.c (worked in practice,
942 never right in theory). Code might be a bit faster now, even.
943 Thanks to Agostino Sarubbo for reporting.
944
945 1.25.0: MP3 now patent-free worldwide!
946 -------
947 - Silence test for artsc-config if it is not there.
948 - Make sure -static-libgcc from LDFLAGS gets through libtool,
949 fixing 32 bit Windows builds (depend on libgcc DLL otherwise).
950 - Fix build with non-GNU make by using plain rm -f instead of silly $(RM)
951 in libout123/modules makefile fragment.
952 - Make build work on iOS, including coreaudio backend.
953 - libmpg123:
954 -- Finally provide position-independent code for x86 with assembly
955 optimisations.The textrels are gone thanks to Won Kyu Park and Taihei Momma.
956 -- Clarify some license language in files descending from the original MMX
957 optimisation.
958 -- Fix return value overflow check for MPG123_BUFFERFILL.
959 -- Introduced mpg123_getformat2() to enable the FORMAT command
960 for the generic control not stealing MPG123_NEW_FORMAT from the main
961 playback loop. The sequence LOADPAUSED-FORMAT-PAUSE (play) is supposed
962 to work now.
963 -- Enable aarch64 optimisations on *BSD by default, too. You can always
964 override that stupid OS whitelist using --with-optimization, anyway.
965 -- Use of the i486 decoder is now discouraged more prominently, in configure
966 output.
967 - out123: Fix stupid crash with verbose mode and tone generation (print
968 the string if the pointer is non-null, not if it is null).
969 - libout123: More consistent error messages for dynamic and legacy
970 (built-in) modules. Namely, you get a hint how if you choose a different
971 module than the built-in ones for a static libout123.
972
973 1.24.0:
974 -------
975 - Avoid repeating genre in metadata printout for specifications like
976 (144)Thrash Metal.
977 - In remote control mode, only enforce --quiet if no verbosity was required.
978 - Prevent --loop and --shuffle or --random from messing with the remote
979 control LOADLIST command (printout of the list would loop without reason).
980 - Fix the mpg123 command (esp. our provided binaries on Windows) to now find
981 modules again relative to the executable directory, not the current working
982 directory. This was a regression in 1.23 and might be security-relevant if
983 you called mpg123 in working directories with untrusted content.
984 Note that mpg123 1.23 looked for modules relative to the current working
985 directory only if the installation prefix for modules did not exist.
986 So, usage on an intact installation (with /usr/lib/mpg123 or the like) was
987 safe. Nevertheless this new version fixes the search to be relative to the
988 binary path as it was with 1.22 and before.
989 - At least consistent behaviour of playlist code in the face of looping.
990 Looping is about individual tracks, always. They are looped also in random
991 mode. Jumping (prev/next keys) is between tracks and resets the loop counter.
992 The display of currently playing track in the playlist is fixed for random
993 and looped play now (bug 198).
994 - Looping is now mentioned for a to-be-repeated track with --verbose.
995 - Move some compiler nagging from --enable-debug to --enable-nagging, fix up
996 some new build failures by adding some pesky feature test macros.
997 - Try not to pollute the terminal buffer with old progress bars in inverse
998 video. Only the currently live one shall be seen. That one is pretty. The
999 others are not.
1000 - Using plain dlopen()/LoadLibrary() for opening modules instead of libltdl.
1001 This also means that --with-module-suffix is gone in configure.
1002 - Windows builds only work when Unicode support is there (older than Windows
1003 2000/XP will definitely not work anymore).
1004 - The out123 tool now features tone generation, with a mix of differing
1005 wave patterns. Makes sense to be able to test the audio output by itself,
1006 and it's fun. See --wave-freq and related parameters.
1007 - libmpg123 version 43:
1008 -- Add flags MPG123_NO_PEEK_END and MPG123_FORCE_SEEKABLE, as suggested
1009 by Bent Bisballe Nyeng.
1010 -- Build fix for MSVC (consistent definition of ssize_t, spotted by manx,
1011 bug 243).
1012 -- Build fix for --with-cpu=ppc_nofpu (thanks to Michael Kostylev, bug 244).
1013 -- Add asm optimized MSVC++ Win32|x64 and UWP|x64 builds
1014 -- Remove old, broken MSVC++ builds
1015 - libout123 version 2:
1016 -- Added OUT123_BINDIR.
1017 -- New search order for output plugin directory: MPG123_MODDIR, or (relative
1018 to executable directory OUT123_BINDIR) ../lib/mpg123, plugins
1019 libout123/modules/.libs, libout123/modules, ../libout123/modules/.libs,
1020 ../libout123/modules, and at last the installation prefix $libdir/mpg213/.
1021 This shall ensure that a build inside a source tree does not try to use old
1022 modules from the system prefix. The normal libtool wrapper deals with the
1023 shared libout123 or libmpg123 only, not modules.
1024 Note that if you set MPG123_MODDIR to a non-existing directory, no modules
1025 will be found (earlier versions fell back to other choices).
1026 -- The OUT123_NAME parameter is now copied by out123_param_from(), as is
1027 the newly added OUT123_BINDIR.
1028 -- Coreaudio: Use AudioComponents API on OSX >= 10.6 (thanks to Michael Weiser).
1029 -- Coreaudio: Fix behaviour of out123_drop(), not killing the output anymore
1030 without re-opening the device (bug 236, thanks to Taihei for the fix).
1031
1032 1.23.8
1033 ------
1034 - Fix long-standing bad memory read (via integer underflow) in ID3 parser
1035 for crafted ID3v2 tags with tiny size information (bug 240). Thanks to
1036 Han Lee & Jerold Hoong for the detailed report.
1037
1038 1.23.7
1039 ------
1040 - Fix the bad RPM spec file (broken since out123 addition, thanks
1041 to Srikanth Rao for noticing).
1042 - More out123_pause() in generic control mode, working around
1043 bug 236. The coreaudio output needs to be more robust, though.
1044 - Do not undef MPG123_EXPORT in the headers, rather be prepared for it
1045 being defined already. This fixes MSVC builds, broken since 1.23.0
1046 (bug 239).
1047
1048 1.23.6
1049 ------
1050 - call out123_pause() instead of out123_stop() in remote control STOP
1051 command, fixing the regression in 1.23.x of not being able to resume
1052 playback with another LOAD unless the new track has differing audio
1053 format (bug 234)
1054
1055 1.23.5
1056 ------
1057 - libout123:
1058 -- fix build of portaudio and SDL on win32 (too many parentheses,
1059 thanks to Sandro Cumerlato for pointing it out)
1060 -- make out123_errcode() return OUT123_BAD_HANDLE as documented
1061 (adding that value to the error enum, was documented but missing),
1062 again thanks to Sandro
1063 - fix windows-builds.sh (shell quoting in copying stage) and add some
1064 bits to shipped header to make it directly usable in MSVC (define ssize_t,
1065 bug 232)
1066 - no overwriting of LIBS in Makefile (build system regression in 1.23,
1067 thanks to Peter Korsgaard for reporting that one)
1068 - no printing of message in response to SIGPIPE, as that could trigger
1069 a hang if stderr itself is the piped culprit (regression from 1.22, bug 233)
1070
1071 1.23.4
1072 ------
1073 - mpg123: Fix logic of prebuffering small bits for glitch-free start of
1074 playback. The logic as introduced in 1.23.0 actually introduced stuttering!
1075 The fix for this is really minimal and I urge everyone to at least include
1076 that one. It is the changeset of rev. 4041.
1077 - libout123:
1078 -- Actually make OUT123_KEEP_PLAYING on by default, as documented.
1079 -- Fix buffer logic with regard to draining, use common device writing
1080 code to handle resuming from pause. This fixes sdl output with buffer,
1081 possibly other sfifo-based outputs like CoreAudio (those were poorly
1082 tested with buffer, as the combination is indeed unusual, see bug 230).
1083 -- Buffer pauses devices for prebuffering, to avoid underruns.
1084 -- More reliable playback with SDL, Portaudio and CoreAudio outputs.
1085 This includes FIFO draining on close to avoid dropping of the end.
1086 It should fix parts of bug 230.
1087 -- Fixed memory leaks with some outputs where a call to the deinit()
1088 function is necessary during available/working module checking.
1089 - mpg123, out123: Use the default output flags to get OUT123_KEEP_PLAYING.
1090 Without this, there is a good chance to abort on SIGSTOP/CONT while playing.
1091 This is a regression from 1.22 .
1092 - Build fixes (bugs 227 and 228) and some build system hackery.
1093 Gone are libc feature test macros in compat.h . This means always
1094 using our own strdup() implementation. Simpler than being bugged by
1095 feature test macros in strict compiler mode (when _DEFAULT_SOURCE is not
1096 defined).
1097 - ./configure --enable-nagging does something also without --enable-debug,
1098 explictly triggers -std=c89, too (hence the above).
1099 - Some fixup to make things work in pedantic C89 mode again.
1100 It really might be time to swich on some serious C99 for the next release,
1101 since there's a followup standard to _that_ which we can ignore now.
1102 - Update list of renamed internal symbols to reduce conflicts during static
1103 linking of libmpg123 or libout123.
1104 - Internal compatibility functions now handled via libtool convenience
1105 archives, avoiding the hackish code copies.
1106
1107 1.23.3:
1108 -------
1109 - mpg123: Man page fixes by ESR.
1110 - Actually check for sys/select.h (again?) in configure so that it is used.
1111 Fixes build on a Cygwin install, at least.
1112 - libmpg123: Always default to enabled IEEE 754 hack for rounding on all
1113 platforms, use ./configure --disable-ieeefloat if you do not want it.
1114 As a protection for the exotic machine that does not use IEEE floating
1115 point storage format, mpg123_init() will error out early if the rounding
1116 does not work out.
1117
1118 1.23.2:
1119 -------
1120 - mpg123: Fix compact ID3 tag printing to take UTF-8 character count
1121 into account again (thoughtless regression in 1.23.0).
1122
1123 1.23.1:
1124 -------
1125 - libout123:
1126 -- Fix logic to prevent cases of output devices being closed more than once
1127 (stop after pausing triggered multiple close calls, which broke sndio, for
1128 example).
1129 -- Fix JACK output handling of port wishlist, was simply broken for more than
1130 two ports (-o jack -a port1,port2,port3). Now, it is ensured that each
1131 output ports gets some connection to an out123 channel, possibly splitting
1132 one to multiple ports or mixing multiple channels into one port (cyclic
1133 connection scheme). Note that this does only apply for an explicitly given
1134 list of ports. You can still use --stereo to ensure that two output ports
1135 (if found) are used also for mono tracks.
1136 -- Also use jack_free() in JACK output for ports list.
1137 -- And finally, ensure that pthread library is linked to JACK module (would need
1138 hacking to work on non-POSIX, replacing sem_t with equivalent).
1139
1140 - mpg123:
1141 -- Hide networking-related command-line parameters when disabled.
1142
1143 1.23.0:
1144 -------
1145 - libmpg123 version 42
1146 -- This adds mpg123_framelength() and makes mpg123_position() truly obsolete.
1147 -- Equalizer optional now (./configure --disable-equalizer) to save precious
1148 memory
1149 -- Clip decode tables for large amplification with fixed-point decoders.
1150 Without that, high-pitched distortion enters really quickly when
1151 trying to increase volume even if output samples would not be clipped,
1152 weren't it for fixed-point math.
1153 - libout123 version 1: First incarnation of libout123, a basic library to
1154 get audio data written to audio devices (or files) in simple blocking
1155 manner as opposed to callback APIs (which it wraps over). This collects
1156 the output modules of mpg123 and makes them available to the wider
1157 masses. Also, the separate shared memory buffer process (contained
1158 in libout123) got a lot of reworking which should be noticable in the
1159 interaction with terminal control.
1160 -- Some tuning of several output drivers (modules) regarding flexible
1161 buffer size, client names, supported formats.
1162 -- Notably the JACK output got a major overhaul and now does not swallow
1163 audio at the end anymore. Huge latency also reduced and tunable. Encoding
1164 fixed to floating point, so that libmpg123 actually chooses what is
1165 sensible.
1166 - Added mpg123 --no-infoframe.
1167 - Detect terminal on input side and enable control keys automatically.
1168 There is --no-control now to disable terminal control anyway.
1169 - Display stuff:
1170 -- Avoid spillover of status line that caused unwanted terminal scrolling.
1171 If the terminal width is smaller than the full status string, a cut is
1172 made to make it fit.
1173 -- Draw a progress bar on the terminal by inversing colors behind status
1174 line.
1175 -- Rework the format of the status line to be more compact and consistent.
1176 -- Rework smart tag printout for adapting to terminal width.
1177 -- Warning messages also start with a line break now to better fit in with
1178 verbose playback.
1179 -- Doubly-verbose mode now needed for long MPEG header print out.
1180 It is a higher level of verbosity than printing the current position.
1181 The compact header is a better fit.
1182 -- Compact header printout also compacted, example:
1183 MPEG 1.0 L III cbr128 44100 j-s
1184 -- Reporting of clipped samples also includes a line break now.
1185 -- Keep ID3v2 data around for terminal control to be able to print on
1186 demand. That was broken for some time.
1187 - Default for --preload now is 0.2 instead of 1 (fill whole buffer before
1188 playback). The maximum is 0.5 . This is mandated by corrected buffer
1189 interaction in libout123.
1190 - Improved interaction with buffer process in terminal control mode.
1191 Seeking is more transparent now, taking dropped samples from buffer into
1192 account to avoid unintended jumps. Direct seeks with number row do not
1193 pause playback anymore.
1194 - Silently skip APE tags (thanks to Hans de Goede).
1195 - Some reduction in bitrot on AIX (typos in output module, build with
1196 --disable-largefile --with-audio=aix, real test welcome).
1197 - Added '[' and ']' keys to skip to previous/next directory in playlist.
1198 - Added support for Haiku to configure script (thanks to Jerome Duval).
1199 - Remove spurious Frankenstein warning when scanning files that have no
1200 gapless info (the reported value of -1152 gapless sample length should be
1201 a giveaway anyway).
1202 - Fix a bug in the command line parser that was present from the beginning,
1203 apparently, causing certain parameters (--au, --cdr) to kill off parsing
1204 of following ones. Nobody complained in all these years. Astounding.
1205 - Forget content type from HTTP redirects (relate to error pages, not the
1206 stream redirected to), fixes web radio streams that do not send any
1207 content type for the actual stream.
1208 - Accept application/pls+xml for PLS playlists. It happens. Don't ask
1209 mpg123 to actually parse any XML!
1210 - Ignore whitespace after HTTP MIME types (esp. before a ";").
1211 - Added --name to mpg123 and out123 for providing client name p.ex. for JACK.
1212 There is no default of appending the PID to "mpg123" as client name anymore,
1213 just plain "mpg123" or "out123".
1214 - Added --devbuffer to mpg123 and out123 for influencing device buffer and
1215 possible latency a bit (perhaps more so in future when drivers are adapted
1216 to pay more attention to that).
1217 - Added out123 --channels <n>, --list-encodings, --test-format,
1218 --test-encodings, --query-format.
1219 - Calling ./configure --enable-nagging does not include --enable-debug anymore,
1220 you specify that separately.
1221 - Fix for newer glibc that wants higher POSIX feature test for getaddrinfo()
1222 stuff.
1223
1224 1.22.4:
1225 -------
1226 - Fix stupid regression introduced with a bugfix in 1.14.1 to make
1227 free format streams work again. It took 3 years for someone to notice.
1228 I added a regression test for that now!
1229
1230 1.22.3:
1231 -------
1232 - Fix UTF-16 surrogate detection because that used to falsely detect
1233 characters above 0xDFFF as surrogates (bug 216).
1234
1235 1.22.2:
1236 -------
1237 - Fix buffering for funny sample formats (namely, 24 bit), that do not
1238 fit nicely into 32768 bytes. Effect was a nasty endless loop where
1239 mpg123 needs to be externally killed.
1240
1241 1.22.1:
1242 -------
1243 - Fix mpg123-id3dump when writing images with funny (manipulated) MIME type.
1244 Stupid mistake in length computation of the fallback file extension caused
1245 junk from memory being appended to the filename if the pointer size
1246 is less than 64 bit. For 64 bit pointers (or longer) it was correct by
1247 accident.
1248 - Fix pedantic build by cleaning up out123 source, also now really showing
1249 the encoding list in --longhelp instead of possibly, again, writing junk
1250 from memory in there.
1251 - Not linking libmpg123 against libltdl anymore (bug 215).
1252 - Update MSVC++ ports a bit to make them work again.
1253
1254 1.22.0:
1255 -------
1256 - libmpg123 version 41
1257 - Build fix: Avoid name clash on OSX SDK with "normal" being defined in
1258 an enum already.
1259 - More resilience against folks who do not check errors and rather pass
1260 NULL pointers around (see NEWS.libmpg123).
1261 This changes some return values to MPG123_BAD_HANDLE instead of MPG123_ERR
1262 (always check for != MPG123_OK).
1263 - Fix WAV writing to stdout on Windows (bug 214).
1264 - Rework Xing/Info/Lame tag parsing, also detecting short ones without
1265 the TOC now.
1266 In addition, the warning about file size being off is a _lot_ more
1267 accurate now.
1268
1269 1.21.0:
1270 -------
1271 - Use LL as shortcut for LOADLIST remote command, L is taken already (bug 210).
1272 - Less namespace pollution (MPG123_EXPORT in header instead of EXPORT, bug 212).
1273
1274 1.20.1:
1275 -------
1276 - Make feed reader more efficient when skipping junk by early discarding
1277 of data from buffers to avoid re-parsing.
1278
1279 1.20.0:
1280 -------
1281 - Added NEON optimized decoder for AArch64 (ARM 64bit environment)
1282 -- enabled for --with-cpu=neon64 (NEON only) or --with-cpu=aarch64 (runtime
1283 switch between neon and generic_fpu, like arm_fpu) configuration
1284 -- compatible with aarch64-linux-gnu toolchains (from Ubuntu, debian)
1285 and Xcode 5
1286 - Added new binary out123 only with the audio output part of mpg123.
1287 This is a precursor to separating out said code into a simple audio
1288 output library. It also allows plugging in some processing in a pipe
1289 before feeding to output.
1290 - Prevent opendir(NULL) for unknown module directory when listing modules.
1291 - some build fixes, among those fixing mpg123-id3dump on Windows
1292
1293 1.19.0:
1294 -------
1295 - libmpg123 version 40
1296 - Support float and 32/24 bit integer output also with fixed-point decoders
1297 (conversion from 16 bit precision).
1298 - A-Law output encoding is actually implemented now (Michael introduced
1299 the encoding back in the old days without ever implementing theconversion
1300 table, apparently).
1301 - Added --with-cpu=arm_fpu as decoder choice for runtime switch between
1302 plain fpu code and NEON.
1303 - Fix ARM builds to work properly on debian systems without special CFLAGS
1304 (NEON decoder, specifically).
1305 - Make scripts/benchmark-cpu.pl check the return value instead of delivering
1306 crazy timings for crashed mpg123.
1307 - Trigger track initialization in mpg123_info() to avoid use of uninitialized
1308 variables (watch out for MPG123_NEED_MORE return value!).
1309 - Fix integer code to 32 bit numbers where it would waste 64 bit on
1310 modern systems.
1311 - Add non-executable stack marking also to ARM assembly.
1312
1313 1.18.1:
1314 -------
1315 - Do not close buffer twice with mpg123 -R (delete outdated code in
1316 control_generic(), fixing bug 207.
1317 - Do not mess around with first command line argument to extract program
1318 name, use a copy instead, fixing bug 205.
1319 - Prevent premature exit on timing of failure to open a track in the
1320 playlist and the user wanting to skip over it (bug 206).
1321 - Fixup of HTTP socket mess, also avoiding unnecessary FILE objects (bug
1322 204, patch by Rajeev V. Pillai).
1323 - Fix playlist behaviour on slow HTTP streams, jumping back ('d' key)
1324 from first track stays at first track (thanks to Rajeev again, sub-bug of
1325 bug 206).
1326 - Fix Frankenstein check not to stumble over ID3v1 tags. My apologies. As
1327 a bonus, the position of the concerned code in the parser is now far
1328 more logical.
1329
1330 1.18.0:
1331 -------
1332 - libmpg123 version 39
1333 - Big version increase for clear distinction regarding the 1.14.x regression
1334 fixes in libmpg123.
1335 - Regression fix: Ensure decoder reinitialization on combination of seek
1336 and resync (buffer overflow bug 201, introduced in 1.14.1).
1337 - Regression fix: Make parsing really end on reaching of junk limit
1338 (introduced in 1.14.4).
1339 - Add check for bad bit allocation value in layer I decoder (removes
1340 audible glitch in bug 201 test file).
1341 - Avoid wrong error message about Frankenstein streams when gapless
1342 decoding disabled.
1343 - Be verbose about big header changes (Frankenstein streams). Note that
1344 changes in samples per frame will confuse the seeking code, but decoding
1345 can get through.
1346 - Actually use the full routine to close tracks. Avoids socket leak when
1347 playing multiple HTTP streams from a playlist (bug 203).
1348 - Send correct host header when connecting through HTTP proxy (bug 202,
1349 figured out by Rajeev V. Pillai).
1350 - The console application now informs about changed major stream properties
1351 (prints out another header line).
1352
1353 1.17.0:
1354 -------
1355 - mpg123-id3dump --no-scan; for speedup
1356 - hide -C parameter from mpg123 --help when it is not supported
1357 - added tinyalsa Output (by Jarno Lehtinen)
1358 - build system tweaking
1359 - some care for fixing Cygwin builds
1360
1361 1.16.0:
1362 -------
1363 - Taihei strikes again:
1364 -- New Huffman decoding scheme (configurable, on by default) brings nice
1365 boost to mpg123 to make it the fastest decoder on modern CPUs again.
1366 -- new AVX decoder
1367 -- assembly-optimized DCT36 for SSE and AVX
1368 - added configs for vintage 3DNow(ext) CPUs (AMD K6) utilizing assembly DCT36
1369 - added configs for vintage SSE CPUs (Pentium M), utilizing C DCT36
1370 - added FORMAT command to generic control
1371 - added support for ID3v2 APIC frames
1372 - added mpg123-id3dump as official companion to extract meta data
1373 (including writing of album art from APIC to files)
1374 - added mpg123-strip as official companion to strip dirt out of MPEG streams
1375 (including meta data)
1376 - not deleting ID3 data right after printing in generic control
1377 - module loader a bit more flexible with relative MPG123_MODDIR
1378 - Fix chopped-off playback with pulseaudio output: draining the output
1379 now on closing.
1380 - SDL output fixup, a) not chopping off end and b) non-fatal underrun
1381 - Fix obvious errors in win32 output ... does it now play the end of
1382 tracks? Need someone to test this.
1383 - added support for screen and iris-ansi terminals with --title (as prompted
1384 by patch from sf.net user canavan)
1385 - fixes/enhancements for sgi audio output
1386 - libmpg123: proper largefile aliases for proper systems (FreeBSD with
1387 one one-and-only off_t
1388 This enables client software that insists on defining pointless
1389 _FILE_OFFSET_BITS.
1390 - wrapper script src/mpg123-with-modules to ease the pain of running mpg123
1391 with modules without installation (setting MPG123_MODDIR)
1392 - removed ALSA 0.5 code (unused for longtime, last GPL-only bit)
1393
1394 1.15.4:
1395 -------
1396 - Writing CDDA data via --cdr works again; had also been broken by attempts
1397 to catch failing writes because of full disk.
1398 - Also, the return values of WAV/CDR writing routines now properly indicate
1399 error to avoid stupid endless loop of "smart" audio writing code that wants
1400 to handle interruptions. Really, --cdr could just go and nobody would miss
1401 it ...
1402 - Fix stupid bug in mpg123_chomp_string() (introduced in 1.15.1), empty
1403 lines were badly treated. Only usage in mpg123 is for parsing HTTP redirect
1404 headers; malicious HTTP servers might exploit that (possibility to zero
1405 bytes in memory).
1406 - some cleanup for nagging mode
1407 - more resync strictness to really catch (all) cases with changing decoder
1408 structure; in practice badly needed when disabling seekbuffer (so it's
1409 your fault, somehow ...) for non-seekable streams, otherwise not really
1410 an issue because of readahead checking
1411 I repeat: This fixes possibly fatal behaviour in the parser when readahead
1412 checking is explicitly disabled (p.ex. via mpg123 --no-seekbuffer on
1413 http streams). By default, readahead checking is _always_ active!
1414 - build system fixed for automake 1.13 (thanks to Nix)
1415
1416 1.15.3
1417 ---
1418 - Fix WAV writing. AGAIN. People love to decode WAV to stdout in various
1419 ways that started to get broken with me starting to care for people who
1420 like to write to full disks. I frikkin' HATE this rat race! I'm even
1421 starting to SHOUT. Now, the code prefers to write no header at all (when
1422 there is no actual output) instead of possibly writing too many misleading
1423 ones. Getting the blame for breaking dir2ogg repeatedly while trying to
1424 cope with one fatal situation that you cannot really cope with sucks. Big
1425 time. I hope that's it now!
1426
1427 1.15.2:
1428 -------
1429 - build fix with older shells (== in configure)
1430
1431 1.15.1:
1432 -------
1433 - libmpg123 API version 37
1434 - Fix corner case in HTTP communication with missing path in Location header
1435 (bug 187).
1436 - Prevent nasty alignment issues by not guessing about assembler alignment
1437 when .balign is available (bug 188).
1438
1439 1.15.0:
1440 -------
1441 - Added --continue (also see scripts/conplay).
1442 - Added number row to terminal control keys for 10% jump points (for
1443 navigating long tracks).
1444 - Set xterm title always if asked, disregarding the -q switch (if you want
1445 fully quiet, don't specify --title).
1446 - Enable terminal mode in OS/2 by explicitly checking for and using
1447 os2term library.
1448 - Getting WAV/AU/CDR writing back to a sane state while still catching
1449 out-of-disk when writing headers (none in case of CDR) but removing
1450 that initial test that prevented writing on any non-seekable file (like,
1451 /dev/stdout to a pipe).
1452 - Added native audio output for QNX (thanks to Mike Gorchak), also improving
1453 portability of configure script while at that.
1454 - more robust recovery from ALSA issues (device suspend)
1455 - Fix crash in module loader on *BSD.
1456 - Extend time display to switch to hours above 60 minutes (hh:mm:ss,
1457 ditching the sub-second precision)
1458 - really include dithering in default x86-64 build.
1459 - Fix mpg123_scan() not properly seeking back for all types of files (bug
1460 3582628 ... it got a different number on sf.net now ... brilliant, folks!).
1461 - Prevent false positives for "Frankenstein!" when using mpg123_scan().
1462 - Fix printing of negative positions with buffer (bug 186).
1463 - Accept application/x-scpls as additional MIME type.
1464 - Try to handle certain Windows toolchains that miss EOVERFLOW.
1465
1466 1.14.4:
1467 -------
1468 Who reworks a parsing and data mangling without really making sure that
1469 things still work like before? I hope I won't do so in future. Not again.
1470
1471 - Fix resync logic to properly work again (used to prematurely end stream
1472 on bad headers). A regression again, this sucks, but at least we got a
1473 regression test for that now.
1474 - Fix aeon-old regression (since 1.11.0) about short seeks: Bit reservoir
1475 was reset also for short seeks that did not break continuity.
1476 - Another damned regression got fixed: Sample-accurate seek for files
1477 without full gapless info.
1478
1479 1.14.3:
1480 -------
1481 - Fix regression from 1.14.1 in parsing of bad free format streams,
1482 avoiding apparently endless loops (just very slow reading, in fact).
1483
1484 1.14.2:
1485 -------
1486 - Fix writing of WAV to stdout.
1487 It is still not the best of ideas to do so, though. At least make sure
1488 to fix rate and channels.
1489
1490 1.14.1:
1491 -------
1492 - Fix subtle bug that could trigger failure to detect a format change on
1493 resync when using libmpg123's feeder API.
1494 That bug has been there since ages but was neutralized by other (buggy)
1495 behaviour of the parser before the little overhaul. The Law of an Even
1496 Number of Bugs.
1497
1498 1.14.0:
1499 -------
1500 - libmpg123 API version 36 (see NEWS.libmpg123)
1501 - Add --ignore-streamlength.
1502 - But also deal with concatenated (Frankenstein) streams: Not applying
1503 gapless cutting once we're over the announced number of frames.
1504 - Also, gapless cutting is only applied if relevant information is
1505 present. Previously, at least the decoder delay has been cut.
1506 - Add --lyrics for printout of lyrics (from USLT in ID3v2).
1507 - Added handling of SIGUSR1 and SIGUSER2 to terminal control code,
1508 configurable via --ctrlusr1 and --ctrlusr2.
1509 - Some low-level API added (mpg123_framedata(), mpg123_framepos(), buffering
1510 and resampling control).
1511 - Added info about input buffer fill to mpg123_getstate().
1512 - ReplayGain: Try to detect LAME below 3.95 for the different reference
1513 level of 83 dB (adding preamp of 6 dB). Version 3.95 itself (not 3.95.1)
1514 I cannot detect. User-set gain values are not touched.
1515 - Keep silent about missing bit reservoir when ignoring frames. That is
1516 the whole point (mostly;-).
1517 - Add the new/old WRITE_SAMPLE variant from MPlayer that makes the generic
1518 decoder faster on older CPUs (x86 at least). Not on x86-64, though (it's
1519 not enabled per default there).
1520 - disable 3DNow and 3DNowExt DCT36, which is slow on modern CPUs
1521 - Handle out-of-disk in the file writers (a bit, see
1522 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=67259)
1523 - Added Xcode project files to build libmpg123 on Mac OS and iOS
1524 - 32 bit output for sndio output (patch by Brad Smith).
1525 - Remote control interface knows LOADLIST command now.
1526 - Now freeing meta data memory after printout.
1527 - Some documentation cleanup, including manpage fixes by quadrispro.
1528 - Re-enabled wildcard expansion on Windows (been broken by unicode file
1529 name support).
1530 - Fix MMX tabinit syntax (thanks to Marcel Müller for pointing that out).
1531 - Fix SGI audio output (been broken since change to modules).
1532 - Fix some fringe behaviour (mpg123_getformat() triggering needless reading
1533 of next frame, possibly hitting bogus MPG123_NEED_MORE).
1534 - Resync limit now also serves to increase amount of skipped junk on
1535 beginning.
1536 - Better separation of stream end and read error (in case there is no
1537 stream opened, even).
1538 - Don't always complain about failed free format header search (only with
1539 verbosity level 3).
1540
1541 1.13.8:
1542 -------
1543 - Fix build for ARM Thumb (incompatible assembly in some macros).
1544
1545 1.13.7:
1546 -------
1547 - Fix endless loop in module loader when starting mpg123 in a directory
1548 that does not exist.
1549
1550 1.13.6:
1551 -------
1552 - Fix output of one random character in -vvv mode for ID3v2.2 frames
1553 (no buffer overflow, move along, folks;-).
1554
1555 1.13.5:
1556 -------
1557 - Fix reading of ReplayGain values from Lame tag (nobody seen this before?!)
1558 (both actual values and the distinction between audiophile / radio).
1559 - Accept application/octet-stream as input from HTTP.
1560 - Update man page with encodings.
1561 - Build fixes.
1562
1563 1.13.4:
1564 -------
1565 - Fix bogus error on resync with big offsets (hits systems that have bigger
1566 off_t than int), bug 3393801.
1567 - Fix windows binary package build script again with plugdir.
1568
1569 1.13.3:
1570 -------
1571 - Fix an endless loop on bad headers, bug 3267863.
1572 - Prevent compiler warnings about unused variables with gcc 4.6 .
1573 - Fix return value handling of getaddrinfo(), bug 3288333.
1574 - Make decwin size computation work for OPT_ARM and sizeof(long) != 4
1575 (bug 3288360).
1576
1577 1.13.2:
1578 -------
1579 - Updated .NET wrapper (MPG123_SKIP_ID3V2, encsize and mpg123_strlen
1580 functions) by Malcolm Boczek.
1581 - Fix crash when calling mpg123_decoder() before opening a track -- a rare
1582 use case since one can choose a decoder with mpg123_new() already.
1583 - Fix format change detection for certain (artificial) concatenated streams
1584 (p.ex. MPEG 1 frames following MPEG 2 frames).
1585 - Prevent the nasty of bug 3154889: No change of pitch without a current
1586 audio format. No proper solution, but more like earlier behaviour.
1587 - Fix some compiler warnings.
1588 - Fix a memory leak in the module loader, also some in the examples,
1589 for being pedantic.
1590 - More use of AC_LANG_SOURCE to silence autoconf.
1591
1592 1.13.1:
1593 -------
1594 - Fix --float and --8bit (MPG123_FORCE_FLOAT, MPG123_FORCE_8BIT) regression,
1595 caused by introduction of 24 bit formats. (bug 3151597).
1596 Btw: The use of specific format choice via `mpg123 -e f32` and friends
1597 is encouraged (meaning mpg123_format_none() / mpg123_format() calls in
1598 libmpg123 API).
1599 - Loosen the parser restrictions: Allow the private header bit to vary. It
1600 is not mpg123's business anyway (finally suggested by Marcel).
1601
1602 1.13.0:
1603 -------
1604 - libmpg123 API version 29 (see NEWS.libmpg123)
1605 - New optimization: ARM NEON (Cortex A series) -- kudos to Taihei, again!
1606 - Added 24 bit integer output (truncation of 32 bit integer).
1607 - Cleanup of symbols for the static library: Only have global symbols that
1608 are actually needed, prefix them with INT123_ to avoid conflicts.
1609 - Added --streamdump to get a copy of the data libmpg123 used.
1610 - Added --icy-interval and tweaked libmpg123 to make playback of ICY stream
1611 dumps possible.
1612 - Added possibility to set module file suffix at build -- to be able to
1613 avoid .la files for plugins.
1614 - Removed compiler directives for alignment and disable associated tests.
1615 The tests don't work on a wide range of compilers anyway, and newer gccs
1616 do have force_align_arg_pointer to fix bad stacks.
1617 - There is no ChangeLog file anymore... it was autogenerated from subversion
1618 log anyway, so look there if interested.
1619
1620 1.12.5:
1621 -------
1622 - Fix bug with the new meta data printout of the mpg123 app; embedded
1623 null bytes in UTF-8 strings in non-UTF-8 environment would trigger buffer
1624 overflow.
1625 Thanks to Jakub Bogusz of PLD Linux for the tip. And yes, shame on Thomas
1626 for changing working string handling code. At all.
1627
1628 1.12.4:
1629 -------
1630 - Fix formatting of meta data printout in UTF-8 (string length computing).
1631 As a collorary, ID3v1 data is reduced to ASCII in UTF-8 environment,
1632 as we do never know what specific encoding that has.
1633
1634 1.12.3:
1635 -------
1636 - Fix a long-standing bug that subtly broke http streaming with ICY meta
1637 data when reading a bit more quickly, like when using the buffer.
1638 I was a very simple and embarrassing bug, actually ... and it made
1639 me wonder how normal streaming still worked rather well, generally:-/
1640 Thanks to Lint for staying with us and reporting bug 3022850, to finally
1641 get a grip on streaming reliability.
1642
1643 1.12.2:
1644 -------
1645 - Fix for nasty memory corruption on configurations that hand out addresses
1646 in the upper half of the memory space (the poiner alignment function broke
1647 in that case, bug 2996045).
1648 - Various build and portability fixes (including making some exotic
1649 configure switches work (again)).
1650 - Add dump_seekindex example
1651 - Sync mpg123_clr to 1.12 feature set, patch provided by Malcolm Boczek
1652 And actually include it in release tarball!
1653 - Rework mpg123.h logic for large file stuff a bit, clients can control
1654 it by defining MPG123_NO_LARGENAME or MPG123_LARGESUFFIX.
1655 - Include dumb wrappers to provide names suffixed with _64 on 64 bit
1656 machines and _32 on 32 bit machines, respectively, to help clients that
1657 insist on defining _FILE_OFFSET_BITS where it is not needed (or with a
1658 non-large value).
1659 - Tuning of the internal buffer code for feeder mode to minimize its
1660 performance impact (works in 4K blocks now).
1661 - Workaround for compiler bugs in Open64/PathScale/SunStudio (bug 3004396,
1662 suggestion by Doug Gilmore).
1663 Note that Sun Studio is still a tricky fellow, at least when it comes
1664 to our preprocessed assembly.
1665 Also, x86-Open64 recently produced another segfault in layer3.c on my
1666 box... GNU and intel compilers are what one can use as something stable.
1667 - Make mpg123_getformat() return more error codes (like MPG123_NEED_MORE).
1668 - Fix handle I/O for clients with small file offset (32 bit when libmpg123
1669 has 64 bit).
1670 - Fix 3DNow(Ext) standalone builds.
1671
1672 1.12.1:
1673 -------
1674 - Update INSTALL for the dual-mode library. No code changes.
1675
1676 1.12.0:
1677 -------
1678 Attention: This reverts the big bad libray name change. People did not
1679 like libmpg123_64.so at all ... so Thomas went along with proving that
1680 being responsible for a open source library package sucks for your life --
1681 he coded large file dual-mode into libmpg123. One library with both APIs,
1682 like glibc.
1683
1684 - libmpg123 25.0.25:
1685 -- The separate libraries are gone again, this one is dual-mode (using
1686 ca. 750 lines of not very funny code).
1687 -- The library even translates callbacks between large file support modes.
1688 -- Introduce handle-based custom I/O in libmpg123.
1689 -- The header now defines MPG123_API_VERSION to help compile-time detection
1690 of available functions (lib is version 25.0.25 now... MPG123_API_VERSION
1691 is 25).
1692 - Make the usage of AI_ADDRCONFIG dependend on a configure check. This
1693 flag is not available everywhere.
1694
1695
1696 1.11.0:
1697 -------
1698 Attention: There was a re-release of the source archive, because the
1699 initial one missed the actual changes for the library name.
1700 Sorry for the inconvenience:-/
1701
1702 - Big bad libray name change for builds with enabled large file support.
1703 That will break libmpg123 binary compatibility on Linux/x86 when upgrading
1704 from mpg123>=1.5.0 with enabled large file support. You need to relink
1705 with the new library called libmpg123_64.
1706 If you do not enable large file support, or it is implicit on your
1707 platform (x86-64), the library keeps the old name of libmpg123 and
1708 everything is fine.
1709 - Assembler cleanup:
1710 -- no use of compiler-based stack alignment (for SSE), helping portability
1711 -- mostly performance neutral, hint at faster decoding for the dynamic
1712 build and slower for static, but small change in any case
1713 - Windows networking (access http streams in mingw builds).
1714 - Do not ask for addresses that do not have matching interfaces configured
1715 (IPv6 on a IPv4-only host, for example). That can help with broken DNS
1716 servers that choke on IPv6 addresses (T-Online Speedport W502V, for example).
1717 - Fix a potential ICY reading problem were ICY metadata could be read as
1718 mp3 data (basically for MPEG frames bigger than the ICY interval).
1719
1720 1.10.1:
1721 -------
1722 - Help multilib setups that boldly assume that headers are invariant
1723 by removing the consistency check in mpg123.h and instead just adding a
1724 suffix to relevant functions when _FILE_OFFSET_BITS is defined. Invalid
1725 setups will now fail at the linking stage instead of the preprocessing stage.
1726 That should cover bug 2934488.
1727 - Fixes for gapless decoding:
1728 -- Correctly skip padding larger than one MPEG frame (strange, but occurs).
1729 -- Bug 2950218 (proper gapless cuts for seeking near the end).
1730 - Make sure the search for the first header aborts after 64K of one-byte
1731 reads, as always intended (bug 2951334).
1732 Certain files (or even random data) were able to fool mpg123 sufficiently
1733 enough that it basically reads through the whole file in 1-byte steps
1734 in search for MPEG data.
1735 The limit of 64K will very likely be configurable in a future mpg123
1736 version.
1737 - Fix subtle bug with seeking in feeder mode: A seek close to the current
1738 position (meaning: the needed data offset is already in mpg123's buffer
1739 chain) resulted in skipped bytes.
1740 - Improve the feeder mode for handling large amounts of junk at the
1741 beginning, including proper passing of the request for more data when trying
1742 to determine free format frame size. This also fixes endless looping as
1743 reported in bug 2951334.
1744 - The parser code still needs overhaul. It is not pretty.
1745 - Compile fix for OpenAL.
1746
1747 1.10.0:
1748 -------
1749 - Confirmed/fixed on OS/2 platform.
1750 At least a static build works with native audio output and generic decoder.
1751 - Make modules work with libtool-2.2.6b .
1752 - Added pitch command to control interface.
1753 - HTTP header field and MIME type parsing now case-insensitive (fixes
1754 bug 2904736).
1755 Also, the MIME-parsing for non-playlist URLs is done the same way as
1756 from playlists now.
1757 And, we interpret MIME types only to before a ";" (see bug 2836904).
1758 - There is a build-time option to enable unicode file names on windows
1759 now. UNIX (Linux) does not need that, UTF-8 works natively.
1760 An unicode-enabled build of libmpg123 expects file names as UTF-8 strings,
1761 opening of non-utf8 strings (by old applications) still works via fallback
1762 when the conversion of the string or opening of the file failed.
1763 - Also, on Windows, --realtime now works, along with added granularity of
1764 the new --priority parameter. Note that you get the latter functionality
1765 through nice/renice on UNIX, no need to add that to mpg123.
1766 - Added experimental framebyframe API.
1767 - Added a new Microsoft common language runtime wrapper (aka .NET wrapper)
1768 written by Malcolm Boczek. See ports\MSVC++\2008clr\mpg123clr\ReadMe.txt
1769 for further information
1770 - Added Visual Studio 2010 Beta 2 project files for libmpg123
1771 - Updated autotools machinery, autogen.sh is gone, use `autoreconf -iv`.
1772
1773 1.9.2:
1774 ------
1775 - Fix logic failure in gapless code: The trailing silence was only cut on
1776 the first decode. So, seek-back and decode again never was gapless:-(
1777 - Fix linker flags on MacOS 10.6 .
1778 - Improve portability (mainly for Solaris/x86, possibly others).
1779 - Fix reading from standard input in Windows (use binary mode).
1780
1781 1.9.1:
1782 ------
1783 - Fix OpenAL output issue (bug 2840085)
1784 - Fix theoretically possible discarding of some audio samples when getting
1785 a new format.
1786 - Fix unwanted verboseness (some error messages from library even when
1787 MPG123_QUIET was specified).
1788 - Fix back-to-beginning for feed_skip (fixes bug 2859531).
1789 - Fix the CPU creep bug with NtoM decoding.
1790 That one has been nasty on you with large files since bad frame fillup
1791 has been introduced.
1792 - Fix regression (since 1.9.0) in ID3 parsing: Now Foobar2000 RVA fields
1793 are interpreted again (among others).
1794 - Fix some compiler warnings, minor cleanup.
1795
1796 1.9.0:
1797 ------
1798 - Free format support.
1799 - ARM assembly optimizations (yay Taihei again!).
1800 - Dither noise now runtime-generated, thus less bloated binary (thanks
1801 Taihei...).
1802 This means that dithered decoding is now included per default.
1803 Please note that work is ongoing to improve the dithering quality,
1804 more to come.
1805 For example: Accurate rounding to 16bits is now always used for generic
1806 dithered decoder, there are plans to adapt the noise for different
1807 sampling frequencies.
1808 - Buffer plays any audio type now.
1809 - New output module: OpenAL
1810 - Bug fixes:
1811 -- Broken frame fillup for non-signed 8 bit formats with correct zero
1812 (bug 2805665)
1813 -- Crash on invalid UTF-16 in ID3 tags fixed (thanks to Andy Hefner)
1814 -- Bad detection of separated UTF-16 strings in ID3 tags (bug 2834636).
1815 - Topic ID3 and UTF: Now we are more tolerant with UTF16, even accepting
1816 those broken tags that id3v2 produces (multiple conflicing BOMs).
1817 - Build system allows AltiVec on non-Macs now (don't enforce unsupported
1818 gcc flags).
1819 - Terminal control mode is more verbose (gives hint about pressing 'h').
1820 - The library now offers a switch and small API to defer translation of
1821 ID3 text data until client app wants to do it (or not).
1822
1823 1.8.1:
1824 ------
1825 - Include improvements to integer decoder from SVN trunk.
1826 They have been accidentally promised on 1.8.0 release announcements,
1827 though they were not there. Here it is: ISO compliant fixed point decoding
1828 (speed optimizations are still reserved for next release).
1829 - Fixed build with ./configure --enable-modules (bug in configure script).
1830 - Fix name of binary in help screen when called with path.
1831 - Build fix for old gcc with AltiVec code.
1832 - Update of Sony PSP build files.
1833
1834 1.8.0: (The Taihei Monmament)
1835 ------
1836 - New and improved SSE optimizations! For x86-64, too! Also AltiVec! Fast
1837 float output! Faster stereo!
1838 Finally, this should put mpg123 into the efficiency-leading position on
1839 current hardware! Thanks go out to Taihei Monma for pushing lots of new
1840 assembler code.
1841 - Added proper rounding modes to 16bit integer.
1842 -- Use ./configure --enable-int-quality to prefer accuracy over speed.
1843 -- Accurate code (generic, SSE, 3dnow -- i586 has always been good)
1844 delivers ISO compliant quality with 16bit output.
1845 -- Uses fast rounding for IEEE754 floats.
1846 -- Mpg123 can do fully ISO compliant MPEG audio decoding to 16bits (of
1847 course to 32bit float/int, too).
1848 - subtle performance enhancements via broader data alignment
1849 - Prefer SSE code over 3DNowExt.
1850 -- It can be tight, but generally our new SSE code is better on CPUs that
1851 can do both.
1852 - Remove HF cutoff (SFB21), has been introduced in 0.59r .
1853 - Fix layer 1 regression (been there since version 1.x).
1854 -- Layer 1 ISO compliance test is positive now.
1855 - Fix layer 2 regression (false positives header change).
1856 -- Layer 2 ISO compliance test is positive now.
1857 - more sample-accurateness
1858 -- Clear frame buffers before seek.
1859 -- Decode 4 frames in advance for layer 3 (2 for layer 1 and 2) per default.
1860 -- Number of layer 3 frames to decoded in advance (and fill bit reservoir) is
1861 runtime tunable (--preframes).
1862 - Do not stop decoding after stray ID3v1 tags in streams (bug 2793677).
1863 - Fix handling of negative RVA2 values in ID3v2 (bug 2792417).
1864 - Discard bitreservoir of previous frames if a resync is needed.
1865 - improve perceived randomness of --random (do not repeat too early)
1866 - Fix special crash case in UTF16 conversion for ID3v2.
1867 - MPG123_MODDIR environment variable for changing module directory at runtime
1868 - arts output fixes (bug 2715128)
1869 - win32 output swallowing less buffers at end (but still some audio is lost,
1870 bug 2722991)
1871 - tested mpg123 on OS/2; it works, but not the native audio output
1872 (SDL works)
1873 - WAV output announces supported output formats correctly
1874 - fix buffer startup to make sure the first bits of audio are not discarded
1875 (bug 2796802)
1876 - re-enable the --gain parameter (the move to modules disabled that),
1877 untested
1878 - Libltdl is _not_ included anymore.
1879 -- The host system is expected to provide it, otherwise plugins are not used.
1880 -- Our copy does not get in the way anymore.
1881 - enhancements of the MSVC++ builds, including optimizations using yasm
1882 - general build system fixes
1883 - Legacy build system is gone.
1884 - code cleanup
1885 -- signed/unsigned integer usage, make compilers more happy)
1886 -- simplified synth selection code (it is _less_ code now, and does more)
1887 - Fixed behaviour of MPG123_NEW_FORMAT: Now it does not stop you from
1888 just continuing decoding without querying the (new) output format, this
1889 behaviour regression had been there since somewhen after 1.4.3 .
1890
1891 1.7.3:
1892 ------
1893 - Fix a regression introduced with the --doublespeed fix. Now seeking should
1894 be a lot smoother again.
1895
1896 1.7.2:
1897 ------
1898 - Fix a possible vulnerability in the ID3 code (negative encoding values).
1899
1900 1.7.1:
1901 ------
1902 - Just one fix for making the RVA switch in terminal and generic control mode
1903 work again (instead of plain silence).
1904
1905 1.7.0:
1906 ------
1907 - Lots of output format fun!
1908 -- now supporting 8, 16, 32 bit signed/unsigned integers,
1909 32bit float (64bit float can be hacked in quickly)
1910 -- optimized decoders use fallback routines for non-optimized
1911 formats transparently
1912 -- flexible resampling is automatically chosen when output does
1913 not support standard rates (like a JACK server)
1914 -- That meant quite some restructuring in the decoder code...
1915 - Lots of core functionality now optional, enabling small-footprint
1916 builds of libmpg123.
1917 -- choosing among the MPEG layer codes
1918 -- NtoM resampling
1919 -- downsampling
1920 -- feeder API (mpg123_open_feed())
1921 -- ID3v2
1922 -- String API
1923 -- ICY
1924 -- output formats
1925 -- error/warning message printouts
1926 -- ... a layer3-only shared libmpg123 can be just 107K on x86-64
1927 or 94K on x86
1928 - --preload (tune prefill before playback for buffered mode)
1929 - win64 support (one report at least;-), modules working on Windows
1930 - Windows/MSVC++: fixed handling of unicode file names,
1931 mpg123_topen() now actually working there
1932 -- You are obliged to use mpg123_tclose() now!
1933 - using "plugins" subdir relative to mpg123 binary
1934 (in addition to ../lib/mpg123)
1935 - switch to libtool 2.2 and automake 1.10 for the build scripts
1936 -- though no update to new libtool 2 API, yet
1937 - some portability fixes
1938 - prevent user from creating non-working I/O setups
1939 (replacing timeout reader, ICY with feeding)
1940 - JACK output accepts port names for -a, stereo gets mixed down
1941 when only one port is chosen
1942 - added sndio output, contributed by Christian Weisgerber
1943 - Make --doublespeed paramter work, porting of which has been
1944 forgotten in the 1.x series!
1945 Along with that, a regression in decoding has been fixed
1946 (concerning skipping of frames in general).
1947
1948 1.6.4:
1949 ------
1950 - Fix for consistent NtoM resampled decoding and broken frames fillup.
1951
1952 1.6.3:
1953 ------
1954 - Fix sf.net bug #2413608: Make sure a new format does not slip through
1955 after early close.
1956 - Include equalizer example file and a piece of wrapper code for MSVC++
1957 in the distribution.
1958 - Do not claim to support unsigned 16 bit encoding (never been the case).
1959 - Fix default fifo detection, now it actally activates fifo support!
1960
1961 1.6.2:
1962 ------
1963 - Fix default IPv6 detection in configure.ac (a typo has been introduced
1964 before).
1965
1966 1.6.1:
1967 ------
1968 - Fix printing of largefile offsets in remote control interface.
1969 - Some portability fixes (reduced compiler warnings on Solaris, p.ex.).
1970
1971 1.6.0:
1972 ------
1973 - small things inside that I forgot to mention
1974 - some new commands for remote control interface:
1975 -- SCAN
1976 -- SAMPLE
1977 -- EQFILE
1978 -- SHOWEQ
1979 - added --keep-open command line flag (for remote control)
1980 - fixed equalizer file parsing to really handle comment lines as example file
1981 suggests
1982 - more elaborate detection of UTF-8 locale, also --utf8 switch to force it
1983 - added --fuzzy to enable inexact seeking (see below)
1984 - added --index-size parameter for tuning accurate frame seek index
1985 - enabling largefile support via configure
1986 - libmpg123 11.0.11:
1987 -- added mpg123_geteq(), mpg123_tell_stream() and mpg123_getstate()
1988 -- API cleanup (take const char* for mpg123_open())
1989 -- calm down complainling (C++) compilers by completing the mpg123_channels
1990 enum
1991 -- fixed several possible crashes when user provides NULLs where it isn't
1992 appropriate
1993 -- workaround for frames that fail decoding (bad frame body, missing bit
1994 reservoir): fill up with silence
1995 That also fixes seeking issues related to bad frames since now the
1996 fixed relation between frame and sample offsets is restored.
1997 -- Finally realized what that the resampling code for integer math
1998 (generic_nofpu) was just incomplete!
1999 Now it works, using the defined integer multiplication.
2000 -- Detect UTF-8 in ICY conversion (instead of just assuming CP-1252,
2001 provided by Torsten Glaser).
2002 -- Prevent ugly noise when the bit reservoir is not fully there
2003 (p.ex. ignorantly cut mp3 files).
2004 Concerned frames are silenced, but reservoir for following frames
2005 is kept.
2006 Thanks to Patrick Dehne for the investigation and patch!
2007 -- New option for fuzzy seeking; trading sample accuracy for speed.
2008 With that option, seeks positions are guessed using mean framesize or
2009 the Xing VBR TOC is used.
2010 -- Feeder streams work with seek index now.
2011 -- Seek frame index is runtime tunable.
2012 -- more caution with alignment issues for varying library users
2013 - new project files to build/use libmpg123 with MSVC++ 2005 and 2008,
2014 provided by Patrick Dehne
2015 - fixed a regression that made the pause/loop mode in terminal control
2016 slowly go backwards
2017
2018 Big note for libmpg123 users: Make sure your application is compiled
2019 with the same largefile setting as libmpg123! If that does not match,
2020 you will get unresolved symbols (which is better than weird errors when
2021 using a nonmatching off_t in arguments;-).
2022
2023 1.5.1:
2024 ------
2025 - fix wav/au/cdr writing to stdout (bug #2044423)
2026 - add more sensible --no-resync to replace --resync (bug #2044404)
2027 - libmpg123 7.0.7
2028 -- fix thread safety of SSE decoder (bug #2053181)
2029 -- fix robustness of length handling, added mpg123_set_filesize() (bug #2064322)
2030
2031 1.5.0:
2032 ------
2033 - libmpg123 6.0.6
2034 -- Floating point decoding back again, (build time choice for now;
2035 --with-cpu=generic_float).
2036 -- More robust string handling; extended string API.
2037 -- fixed an (so far) unnoticed ID3 bug that may have obscured comment fields
2038 -- added mpg123_feed() as shortcut; using const input buffer now
2039 -- resync is enabled again for ICY streams (the trouble in the past was
2040 due to a reader bug, not the streams)
2041 -- decode ICY text data to UTF-8 (thanks, Thorsten)
2042 -- preserve file flags for timeout mode
2043 - More elaborate (helpful?) error message for output format problems.
2044 - Reworked HTTP code... and added IPv6 support!
2045 - Fix the --proxy parameter (has been accidentally ignored before).
2046 - Added --ignore-mime to work with web streams with funny playlist MIME
2047 types like "audio/megurl".
2048 - Know MIME audio/mpegurl.
2049 - --timeout is now also used for the connection itself (not just reading).
2050 - Now trying to resume ALSA device after suspend.
2051 - subtle build changes (your CFLAGS rule now)
2052
2053 1.4.3:
2054 ------
2055 - Fix mpg123_tell() and mpg123_seek() return values with respect to the
2056 samples in decoding buffer (bug 1947373).
2057 - Fix (bogus) 3DNow detection on IDT C6 (bug 1947982).
2058 - Fix mono playback of win32 output (bug 1955157).
2059 - Fix ABI compatibility with compilers that don't align the stack like
2060 gcc does (fixing SSE crashes on win32).
2061
2062 1.4.2:
2063 ------
2064 - Fix build problems (for Cygwin and Win32, perhaps others, too).
2065
2066 1.4.1:
2067 ------
2068 - Fix a serious regression of terminal control mode (forgot to remove a
2069 term_restore(), bug 1936548).
2070
2071 1.4.0:
2072 ------
2073 - Enable runtime check for available and working output modules.
2074 Now mpg123 tries all built modules (or a list specified in
2075 mpg123 -o moda,modb,modc) in turn to find a working output. This minimizes the
2076 situation of having to specify a device when default output doesn't work for
2077 some reason. See bug 1910500.
2078 - libmpg123 2.2.2
2079 -- Fixing an endless loop for ICY streams that suddenly end without error
2080 (just EOF).
2081 -- Further fixes around EOF handling in (buffered/ICY) streams.
2082 - Enable user to interrupt mpg123 when stuck in a loop of failing to open a
2083 track (with --loop).
2084 - This time with correct RPM spec (hopefully... sorry, Michael!).
2085
2086 1.3.1:
2087 ------
2088 - Enable snd_pcm_sw_params_set_xfer_align() for alsa-lib < 1.0.16 (bug 1908603).
2089 - libmpg123 2.1.2 (no-op change)
2090
2091 1.3.0:
2092 ------
2093 This is a _major_ fix release for http streaming!
2094 Much more usable than earlier mpg123 1.x versions.
2095
2096 - libmpg123 libtool-version 2.0.2
2097 -- Fix nasty bug in ICY reader that failed to account for partial reads
2098 from slower network.
2099 -- Now non-seekable streams can get peek-ahead for MPEG sync, too, triggered
2100 by the MPG123_SEEKBUFFER flag.
2101 -- Do not try to resync on streams with ICY metadata, that won't work.
2102 Consequently, if there's unreliable network, use --loop -1 to make
2103 mpg123 reconnect to the server on errors.
2104 -- Applying a few layer 3 decoder fixes from old mhipp CVS (hail to the
2105 creator;-) That includes the fix for bug 1895025.
2106 -- Fix ID3v2 parser with very invalid claimed ID3v2 frame sizes (could
2107 have triggered DOS before).
2108 -- Oh, and fixed a steady memory leak in ID3 handling. Should mention that;-)
2109 - Print ICY-NAME and ICY-URL again, as mpg123-0.6x did.
2110 - Added --smooth option for not flusing buffer between tracks but at program
2111 end only.
2112 - Included some contributed ports/bindings of libmpg123 under ports/ .
2113
2114 1.2.1:
2115 ------
2116 - libmpg123 libtool-version 1.2.1 (coincidence!)
2117 -- Earlier check for free format headers to prevent them from screwing up
2118 resync (free format still not supported)
2119 -- bug 1889051: more flexible resync, not insisting on the format of the
2120 first frame (which could have been bogus)
2121 - Add an output write loop to handle SIGSTOP/CONT operation (bug 1890563).
2122 - Fixed a bug where playback status for some http streams was not shown.
2123
2124 1.2.0:
2125 ------
2126 - libmpg123 libtool-version 1.0.1
2127 -- See NEWS.libmpg123 for details.
2128 -- This adds --resync-limit parameter to mpg123.
2129 - Some build system work, including:
2130 -- It is now possible to build a mpg123 rpm from the source tarball via
2131 rpmbuild -tb .
2132 -- configure: Accept --with-audio=foo,bar,bla (comma-separated list) in addition
2133 to --with-audio="foo bar bla".
2134 -- autogen.sh included in distribution.
2135 -- Include the forgotten i486 files in the distribution (now that's embarrasing...).
2136 Also a little fix for wrong --list-cpu on i486 build.
2137 -- Fix build with MinGW32.
2138 - Fix -v stat output (time display) for MinGW32 build.
2139
2140 1.1.0:
2141 ------
2142 - Added arts output.
2143 - More verbosity on lame/xing/info tags (on demand) instead of compile-time
2144 debugging switch.
2145 - XMMS2 plugin cleanup (submitted to XMMS2 for inclusion)
2146 - Improved ID3v2 robustness (let bad tags not affect playback of the important
2147 part).
2148 - Terminal control:
2149 -- Key for going to previous track ([d], because it's left of [f] usually...
2150 no config file yet for that).
2151 -- Key for printing current playlist, [l].
2152 -- Now possible to switch on verbosity also for mpg123 -q -C
2153 -- Key for (re)printing ID3 tag info ([t]).
2154 -- Key for (re)printing MPEG frame header info ([m]).
2155 - Soften the -q option: Now you are able to increase verbosity after -q
2156 again, like mpg123 -q -v (this fixes cplay's progress display).
2157
2158 1.0.1:
2159 ------
2160 - Fix module loader to prevent crash in working paths with more than 49
2161 characters length (bug 1859413).
2162
2163 1.0.0:
2164 ------
2165 - Fixed Pulseaudio.
2166 - Cleaned up configure.
2167 - Pushed remote control version to v3.
2168 - Updated documentation.
2169 - Fix VPATH build for generated mpg123.h .
2170
2171 1.0rc3:
2172 -------
2173 - Fixed JACK output.
2174 - Disable buffer when statically building CoreAudio in (MacOSX); with
2175 CoreAudio module the buffer works, with static output it doesn't.
2176 This is the situation because of fork() discrimination by Apple Inc. .
2177 - Added verbosity to tell about ID3v2 text encodings (level 4).
2178 - minor cleanup
2179
2180 1.0rc2:
2181 -------
2182 - Fixed some bugs, the ID3v2 interface of libmpg123 is safer now (there
2183 were possibly stale pointers in there).
2184 - Buffer mode got vital improvements (yeah, that is sort of feature creep
2185 during release process, but worth it)
2186 The improved communication makes it possible to skip tracks in terminal
2187 control mode without the hack to immediately restart playback.
2188 To be able to do seeking in terminal mode with buffer properly, buffered
2189 playback stops on seek keys. You restart playback after you are done
2190 with seeking.
2191 - Support either portaudio v18 or v19 API (v18 works for me, v19 works
2192 according to user report).
2193 - Audio output handling got safer and more clean, a byproduct of implementing
2194 the above. I stumbled over some things and had to fix it (yeah, feature
2195 creep in parts again, but still: Worth it!;-)
2196 Any audio access is now restricted to either that standalone mpg123
2197 process or the buffer process, the output being opened _after_ the fork().
2198 There is hope that this fixes buffered output on MacOSX.
2199 - More feature creep, this time because of something like that was already
2200 included in the Debian mpg123:
2201 --delay parameter to issue a pause before each track (does not make
2202 sense for file writing outputs, though).
2203 - Even more feature creep: remote control can now give ID3v2 data, also
2204 in more raw form via new "tag" command.
2205 This sort of qualifies as bugfix, actually, since it's item 1786430 on
2206 the tracker;-)
2207 - Make flexible rate computations more safe (especially on 32bit platforms)
2208 by switching to the looping code instead of direct multiplication/division
2209 (which blows up with integers). A possible effect was premature track
2210 ending with a forced rate.
2211 - Flexible rate mode works _at_all_ again with fixed output support
2212 detection code (comparison of -1 with a size_t ended a loop all too early).
2213 - check for error in flushing output (this ends mpg123 for a broken pipe,
2214 for example)
2215
2216 1.0rc1:
2217 -------
2218 - In one word: LOTS!
2219 - Many small things that happened while migrating to the new setup.
2220 - The decoder is now a library (libmpg123) which can be installed also
2221 for other programs to use. This finally provides a heavily improved
2222 replacement for the aged mpglib. We have doxygen API doc, example codes.
2223 You can still choose to compile the decoder statically into the mpg123
2224 binary, though.
2225 - We have a plugin system for audio output! This should be relieving to
2226 package maintainers for binary distributions, drastically reducing the
2227 number of builds they have to provide.
2228 Here, too, you can choose to build one audio output into the binary
2229 like before; you can have your old-school one-piece mpg123 or the hip
2230 I-am-legion shared-lib-and-plugin flexible mpg123;-)
2231 One exception: For now, you have to compile without modules in
2232 MinGW32/MSYS.
2233 - ID3v2 support has been enhanced and, in the case of specific comments
2234 (those with a description), fixed.
2235 The ID3 parsing is part of libmpg123, offering support for virtually
2236 all textual information in ID3v2.2, ID3v2.3 and ID3v2.4 tags (along with
2237 the usual 128 bytes of ID3v1 or ID3v1.1 in a block).
2238 All tag texts are converted to UTF-8, allowing mpg123 to properly display
2239 special chars in an Unicode terminal without fuzz (for non-UTF locales,
2240 there is still the ASCII reduction).
2241 - Gapless decoding (of layer 3 streams) is now considered to be a standard
2242 feature and is enabled per default both at build- and runtime.
2243 - ...
2244
2245 0.68:
2246 -----
2247 - Fix bug 1807964: distorted playback of MPEG 2.5 files
2248
2249 0.67:
2250 -----
2251 Thanks go out to the ROM 106.5FM team (http://www.rom.lu/) for supporting
2252 this release with a donation on sf.net and a bottle of finest Gewuerztraminer
2253 wine from Luxembourg!
2254
2255 - verified that it builds on SGI IRIX 6.5 for MIPS32 and MIPS64
2256 - enable flexible rate for stdout decoding, too
2257 - UTF8 filtering to ASCII finally works thanks to Mika Tiainen, see bug 1745847
2258 - corrected frame offset; -k was off by one
2259 This means now that -k disagrees with mpg123 0.59r on many files because
2260 the old version does play a frame of silence when encountering LAME tag
2261 instead of dropping it for audio.
2262 - added --timeout <n>: Possibility to end playback on a stream that doesn't give
2263 data for <n> seconds.
2264 - added --loop <n>: for ... well... looping playback of a track; motivation was
2265 persistent internet streaming with unreliable connection (together with --timeout)
2266 - build fixes
2267 - use movap instruction in SSE assembly when compiler supports the needed
2268 alignment (idea from MPlayer)
2269 - show decoder options in their priority order in --list-cpu/--test-cpu
2270 - added Perl script to benchmark each of mpg123's CPU optimizations
2271 - portability enhancements and cleanups along with testing on SunOS 4.1.2
2272 with gcc-2.7.0
2273 - workaround for ugly flexible rate for MMX/SSE/3DNowExt code:
2274 --with-cpu=mmx also includes i586 code as fallback; similar with the others
2275
2276 0.66:
2277 -----
2278 - support for a combined build of x86 optimizations and runtime selection
2279 among these The (supposedly) "highest" CPU-supported optimization is chosen
2280 by default, --cpu, --list-cpu and --test-cpu parameters are there for you to
2281 take control.
2282 - default build for x86 systems is combined "x86" with generic, i386,
2283 i586, MMX, 3DNow, SSE, 3DNowExt; "x86_dither" also includes the dithering
2284 i586 decoder
2285 - no automatic -march compiler flags anymore; the --with-cpu configure option
2286 selects specific assembler code but special CFLAGS are up to you
2287 - extended 3DNow and SSE optmizations from MPlayer's mp3lib
2288 - x86 assembler optimizations adapted for MacOS X/Intel
2289 - straight support for Windows using MinGW32, reactivated the native
2290 WinMM output. Core functionality including assembler optmimizations
2291 and WAV writing is there.
2292 - assembler optimizations also for cygwin
2293 - set remote-err automatically when putting audio to stdout
2294 - a bit more general portability in the new build system
2295 - fix file creation mode for -O
2296 - experimental decoding to 32bit float (raw and WAV file): ./configure
2297 --with-cpu=generic_float
2298 - generic control interface (-R) can create and read from a named pipe,
2299 parameter --fifo <path>
2300 - audio bugfix: playback of mono file on stereo-only hardware, better
2301 error reporting there
2302 - ALSA output handling underruns
2303 - MMX, SSE and 3DNowExt fallback to i586 and below or 3Dnow for forced
2304 sample rate
2305 - more terminal control: volume, RVA and verbosity controls
2306
2307 0.65:
2308 -----
2309 - now allow crc bit to vary over stream (fixes files with bad lame tag
2310 and some internet streams)
2311 - do not try to parse id3v2.2 (as opposed to 2.3 and 2.4)
2312 - correct id3v2.3 parsing for not expecting syncsafe integers for framesize,
2313 2.4 introduced that
2314 - omit gcc specific compiler flags when gcc is not there
2315 - initital attempt to support sgi/irix with configure (please test)
2316
2317 0.64:
2318 -----
2319 - fixing a potential denial of service problem; broken connection to server
2320 could trigger an endless loop
2321
2322 0.63:
2323 -----
2324 - build fixes, only relevant to developers and people who had problems building
2325
2326 0.62:
2327 -----
2328 - JUMP command in generic frontend accepts offsets/positions in seconds
2329 (JUMP 4s vs. JUMP 4)
2330 - initial support for ICY meta data in SHOUTcast streams
2331 - a bit of fixing with rva code, added rva and general volume control to
2332 generic frontend
2333 - accepting more MIME types for http streaming (application/pls and
2334 audio/x-mpeg)
2335 - ID3v2 info output taking less space per default - more resembling the
2336 old ID3v1 view for short tags, full view with always one line per item
2337 via extra --long-tag switch
2338 - generally less verbosity (i.e. noise on console) per default
2339 - ID3v2 now having some kind of support for unicode; trying to print what's
2340 latin1-compatible and prevent messing up of terminal with unknown bytes
2341 - several bug fixes and small improvements
2342
2343 0.61:
2344 -----
2345 - improved AltiVec code from Taihei Monma
2346 - an i486 optimization by Petr Salinger
2347 - improving interactive experience (terminal, frontends)
2348 -- prevent/lessen bad effect after seeking while paused
2349 -- show targeted position while seeking in terminal mode while paused
2350 -- not messing up terminal on unexpected errors
2351 - decoder robustness - better file sync (weird junk at beginning, some vbr
2352 frames)
2353 - build system improvements:
2354 -- fixing configure isses on some platforms
2355 -- adding cygwin, kfreebsd
2356 - extended fix for CVE-2006-1655
2357 - generic code should compiling again...
2358
2359 0.60 final: (The Team!)
2360 -----------
2361 - no functional changes
2362
2363 0.60-beta6: (Thomas Orgis)
2364 -----------
2365 - more use of ID3v2 tags, mpg123 now displays both (merged) info from
2366 ID3v2 and ID3v1; trying to support both ID3v2.3 and ID3v2.4, but ignoring
2367 encoding (UTF8 should still work if your terminal supports it, straight
2368 unicode should work except for parsing genre numbers and multiline strings)
2369
2370 0.60-beta5: (Thomas Orgis)
2371 -----------
2372 - more robust playlist support:
2373 -- path/url length not limited to 1023 characters anymore
2374 -- reading in of whole playlist at program start (old operation was not
2375 really safe with playlist from http)
2376 -- fixed bug where reading of playlist form stdin caused segfault (introduced
2377 with the pls format support)
2378 - smaller cleanups, screen output cosmetics
2379
2380 0.60-beta4: (Nicholas Humfrey & Thomas Orgis)
2381 -----------
2382 - proper seeking, also for VBR streams
2383 - real two-way seeking in remote control mode at all
2384 - better estimates of remaining frames (on vbr files), exact if Xing header
2385 is found
2386 - Removed libao support (as it is GPL and we aren't - and we don't need
2387 it with native Alsa)
2388 - Added ALSA 0.9/1.0 support by Clemens Ladisch (old alsa is in
2389 audio_alsa05.c but will most likely vanish soon)
2390 - Added PortAudio audio-output support
2391 - Added SDL audio-output support
2392 - Fixed NAS audio-output support
2393 - re-enabled MPEG 2.5, code just seems to work
2394 - improved stream checking; being more willing in general (favor resync
2395 instead of bail out) but directly give up if free format stream is obvious
2396 - Whee! Initial Relative Volume Adjustment (buzz) support with reading of
2397 ReplayGain values in lame/info tag, the RVA2 frames in ID3v2.4, ID3v2 TXXX
2398 tags from foobar2000 (please test, someone!) and custom ID3v2.3 comments
2399 entitled RVA / RVA_ALBUM / RVA_AUDIOPHILE / RVA_RADIO / RVA_MIX
2400 - more persistence in trying to get http streams
2401
2402 0.60-beta3: (Thomas Orgis)
2403 -----------
2404 - fixed bug with segfault in http code on stream from live365.com -
2405 bug id 1529266
2406 - made http request HTTP/1.0 compilant again (tested only without proxy) -
2407 bug id 1529319
2408 - now using http content-type header to check for correct types mpg123 can
2409 understand
2410 -- you can now always use -@ for a single http resource (direct stream or
2411 playlist) since a stream will be reopened based on MIME type,
2412 though omitting it will save you one http request to the server in the
2413 direct stream case...
2414 - now simple parsing of shoutcast pls playlists (how did you people use
2415 shoutcast before?)
2416 - playlist mode enhanced with selective playback of one entry in list
2417 (--listentry parameter)
2418
2419 0.60-beta2: (Thomas Orgis)
2420 -----------
2421 - fixed bug in terminal control mode that caused the f key to quit program
2422 isntead of advance to next file (introduced by the first attempt to restore
2423 "normal" Ctrl+C behaviour in terminal control mode)
2424
2425 0.60-beta1: (Thomas Orgis & Nicholas Humfrey)
2426 -----------
2427 On the way to the first new offical realease since years!
2428
2429 - new license: all LGPL except the old audio_alsa.c
2430 - merge with Nicholas' sourceforge package
2431 -- autooled configure/make system
2432 -- MacOSX, Jack, libao (alsa 0.9/1.x support through that)
2433 - experimental gapless playback for layer 3
2434 - merged in MMX and AltiVec optimizations
2435 - security bug fix of GLSA 200607-01 and catching possible crashes found by fuzzing
2436 - fixed many issues here and there
2437 - polished help messages, updated man page...
2438
2439 UNDERGROUND 0.59r-thor7: (Thomas Orgis <[email protected]>) (well, I'm going to
2440 ------------------------ release a _really_ official mpg123 in future)
2441 - applied relevant stuff from Debian's 22:
2442 -- fix for CVE-2006-1655 (buffer overflow in III_i_stereo())
2443 -- strict aliasing hack in WRITE_SAMPLE of decode_i386.c
2444 -- fixed http streaming through proxy by actually using the request url...
2445 -- buffer resync on Ctrl-C (cleaning, reducing pauses with multiple files)
2446 -- removed unused esdserver variable from mpg123.h
2447 -- did NOT apply the Debian Makefile/manpage changes (going to rewrite that
2448 stuff anyway)
2449 - made the netbsd-i386 targets visible as netbsd-x86
2450
2451 UNOFFICIAL 0.59r-thor6 (Thomas Orgis <[email protected]>)
2452 ----------------------
2453 - split nedbsd targets for realtime priority selection code; since one
2454 needs -lpthread with it. standard netbsd target doesn't use realtime code --
2455 I won't recommend running mpg123 as root anyway!
2456
2457 UNOFFICIAL 0.59r-thor5 (Thomas Orgis <[email protected]>)
2458 ----------------------
2459 - Separated the the use of char* equalfile as filename from the use as
2460 TRUE/FALSE flag by introducing int have_eq_settings. Compilers kept nagging
2461 about assigning and integer (TRUE) to char* ... and it feels more right
2462 this way.
2463 - Made control_generic aware of real being possibly different than float (seq
2464 command used to raise an FPE on OSF1 when this assumption was not the case.
2465 - Fixed the command line arg parsing to work properly on Alpha (32 and 64
2466 bit mixup)
2467
2468 UNOFFICIAL 0.59r-thor4 (Thomas Orgis <[email protected]>)
2469 ----------------------
2470 - merged with 0.59r-19 from Debian project (Daniel Kobras
2471 <[email protected]>), hopefully fixing all those security vulnerabilities
2472 -> my dirty http_open and open_stream variants now replaced by more
2473 proper code
2474 -> still my (rather workloaded) parsing/processing in control_generic,
2475 but now including the idea to watch out for \n, \r\n and \r (or \n\r,
2476 for that matter;-)
2477
2478 UNOFFICIAL 0.59r-thor3: (Thomas Orgis <[email protected]>)
2479 -----------------------
2480 - hacked additional variants of http_open and open_stream that return
2481 errors instead of exit()ing; makes more sense when used by a frontend
2482 - further cleanup (and return to using continue instead of nested
2483 if(){}else{if()else{}}) in command processing, better error messages there
2484
2485 UNOFFICIAL 0.59r-ThOr(remote-err,eq,flushing): (Thomas Orgis <[email protected]>)
2486 ----------------------------------------------
2487 - added audio_flush() calls to various places in control_generic.c to
2488 make sure there is nothing left in there that hurts later (not sure if
2489 this has effects)
2490 - (re?)enabled (?) code for stream info
2491 - corrected answer of JUMP (now gives actual reached position, +/- 1 frame
2492 since I'll still have to figure out what frame 0 really is...)
2493
2494 UNOFFICIAL 0.59r-realeq-remoteerr: (Thomas Orgis <[email protected]>)
2495 ----------------------------------
2496 - applied patch from Cool Feet Audio project ([email protected], see
2497 README.cfa) to use real-time equalizer in remote mode
2498 - applied patch from AMF (amf.sourceforge.net) by Steve Grundell for remote
2499 communication not polluting the STDOUT where the audio data is read from
2500 (instead using STDERR) ... and faster seeking, maybe
2501 The patches go not cleanly together, but rejects are trivial. Thus,
2502 a combined patch with both would makes sense.
2503 - modified (IMHO improved) the control_generic command reading: now
2504 always making sure that the full command is read up to a mandatory \n;
2505 also reworked the application of the commands afterwards
2506 - added SEQ command: simple eq with only 3 bands
2507 - added SILENCE command: I don't want the playback messages when I don't
2508 read them (since my frontend is reading the STDOUT it knows what is going on)
2509 - added SLJ command: SILENCE-LOAD-JUMP... a hack for my hacked
2510 mixplayd... maybe one should remove that again.
2511
2512 0.59r: (MH)
2513 ------
2514 - applied audio_alib changes from Ralf Hildebrandt <[email protected]>
2515 - some 'paetsches' from DK
2516 - minor optimizations in layer3 (dequantize/huffman)
2517 - BSDOS 4.0 with gcc added to Makefile (<[email protected]>)
2518 - Solaris-ESD patch from Ryan Prosser <[email protected]>
2519 - Added Frontend changes from Bertrand Petit <[email protected]>
2520 - more DK 'paetsches' (final dk7-clubhaus edition), includes extended term
2521 control
2522 - SGI audio fix for non RAD machines from Erik Mouw <[email protected]>
2523 - fixed ugly bug in layer3 dequant for bt==2 (may cause high frequency clitchs)
2524 - AMD-3DNow optimizations in decode and dct64
2525 - OS/2 reworked from Samuel Audet (see audio_os2.c for more)
2526 - added freebsd-esd make-entry from Daniel O'Connor ([email protected])
2527 - added enhanced head_check from D. Skarda <[email protected]>
2528 - applied 'xterm-title-patch' from "Wilson, Jeff D" <[email protected]>
2529 - more esd changes from Robert Bihlmeyer <[email protected]>
2530 - applied Hannu Napari's <[email protected]> SGI audio patches
2531 - minor changes in httpget.c and mpg123.c
2532 - added native AIX support from [email protected]
2533 - applied playlist patch from <[email protected]>
2534 - more SGI audio patches from Gilles Zunino <[email protected]>
2535 - applied esdserver patch from Matthew Parslow <[email protected]>
2536 - added some fixes for control_generic
2537
2538 0.59q: (MH)
2539 ------
2540 - ESD fixes
2541 - minor bugfixes (DK patch ;)
2542 - au and cdr 'support' (it's a hack)
2543 - Fixed 'VBR' Bug in get_scale_fac1(). Thanx to <[email protected]> for
2544 this. (Wasn't really a VBR Bug. The new encoders just triggered it.)
2545 - Added equalizer patch from <[email protected]>
2546
2547 0.59p: (MH)
2548 ------
2549 - Added mpglib (see mpglib directory)
2550 - 'make generic' is more generic now
2551 - Added HPUX/ALib support from Ducroquet Erwan ([email protected])
2552 - Added EsoundD support from <fill it in>
2553 - Added exp. relatime support (compile with -DSET_RT)
2554 - Added linux-sparc patch from <hmm lost email)
2555 - Added genrepatch from Shane Wegner <[email protected]>
2556 - Added wav-File patch Samuel Audet <[email protected]>
2557 - Added SGI-RAD support from "J. Dean Brederson" <[email protected]>
2558 - Added sajberplay/FreeBSD patch from Chou Ye-chi <[email protected]>
2559 - Added 486 optimizations from Fabrice Bellard <[email protected]>
2560 - Added combined output devices on SUN (ie Speaker and headphones together)
2561 - Added audio_alsa.c (under GPL license) for the ALSA sound system from
2562 A. Hermansen <[email protected]> and J. Kysela <[email protected]>
2563 - Added K6-3DNow patches from KIMURA Takuhiro <[email protected]>
2564 - Added MINT patches from Petr Stehlik <[email protected]>
2565 - Fixed minor bug for linux-i486 and mono
2566
2567 0.59o: (MH)
2568 ------
2569 - fixed Bug with stereo streams on mono audio hardware for OSS
2570 - fixed a bug with mono output and 1:1 (no) rate conversion
2571 - again changes in the probe code
2572 - fix in the buffer code for the case, that audio parameters are changing
2573 - more changes to the flexibel rate converter, should now work when
2574 writing to STDOUT.
2575 - Added Andy's (<[email protected]>) float2int speed up proposal
2576 (check WRITE_SAMPLE in decode_i386.c) .. Maybe someone wanna take the
2577 idea and put it into the decode_i586.s part?
2578 - Added TK3Play support from Brian Foutz <[email protected]>
2579 (compiles but untested)
2580 - RIFF Header fix (according to Thomas Niederreiter's
2581 <[email protected]> patch)
2582 - m68k patch from Stefan Gybas <[email protected]> applied
2583 - minor additional PPC changes applied
2584 - resync and EOF hang fixed
2585 - resync and bitrate correction relaxed
2586 - Slightly change verbose output; added countdown
2587
2588 0.59n: (MH)
2589 ------
2590 - fixed shuffle bug
2591 - added memory mapped IO for file reading
2592 - added MPEG 2.5 12kHz and 8kHz table entry
2593 - started flexibel rate conversion ..
2594 - new '-Z' / '--random' option
2595 - minor clean-up
2596 - splitted audio file in audio system dependend smaller parts
2597 - started system stream support (external program: system.c)
2598 - unsigned 8 bit (instead of ulaw) is now default for 8bit operation
2599 - Linux PPC patch from Grant Erickson <[email protected]> applied
2600 - added BSDi patch from Peter Berger <[email protected]>, who reported
2601 a failure when catching signals
2602 - minor bug fix in the control interface for sajber jukebox
2603 - fixed the ugly 'Ctrl-Z in buffer mode' bug .. YES ;)
2604 - added buffer preload when underrun occurs (preload = 1/8 buffer size)
2605 - minor SUN audio clean up
2606
2607 0.59m: (MH)
2608 ------
2609 - Bug with RIFF header streams fixed.
2610 - HTTP auth option patch from Henrik P Johnson <[email protected]>
2611 - NetBSD patch(s) added from <[email protected]> and <[email protected]>
2612 - Added advanced shuffle support from Steven Tiger Lang <[email protected]>
2613 - fixed bug with too large big_value field
2614 - fixed (hopefully) the bug with the '-a' option
2615 - work-around for some pedantic compilers in fr->synth init
2616 - some changes for the 'generic' version
2617 - applied NAS bug-fix patch
2618 - new (good) MPEG 2.5 tables (from Niklas Beisert)
2619 - added WIN32 support from Tony Million
2620 - added some bug fixes from Niklas Beisert
2621 - fixed another ugly bug in layer3 dequantize (wrote over the border
2622 of the 'xr' field)
2623 - a minor HPUX clean up
2624 - added BSD patch from Kevin Brintnall <[email protected]>
2625 - added Makefile.win32
2626 - added NAS big endian fix
2627
2628 0.59l: (MH)
2629 ------
2630 - simple equalizer support (to be continued)
2631 - added NAS patch from Martin Denn <[email protected]>
2632 - real mono support
2633 - added pentium optimized decode_i586 from Stefan Bieschewski <[email protected]>
2634 - fixed a bug in MS dequantize function
2635
2636 0.59k: (OF)
2637 ------
2638 - fixed "Accept" header in httpget.c
2639 - fixed buffer overflow (causing core dumps) in httpget.c
2640 - fixed display of song length
2641 - fixed -b (buffer) problem on SGI / IRIX -- it definitely works now
2642 - fixed problem when files with different sample rates where played,
2643 it now works with -b (buffer), too
2644 - fixed order of local variables in decode.c to work around a
2645 compiler bug in ultrix
2646 - fixed Makefile for ultrix
2647 - reverted to old getbits.s (the new one could cause coredumps)
2648 - more compact MPEG header display in non-verbose mode
2649 - small fixes in buffer code and elsewhere
2650
2651 0.59j: (MH)
2652 ------
2653 - added sajber jukebox control interface (it's currently a big hack)
2654 (see 'JUKEBOX' file for more)
2655 - added simple rewind facility .. (only used by sajber interface at the moment)
2656
2657 0.59i: (MH)
2658 -----------
2659 - added front-end (remote) patch from "Brian J. Swetland" <[email protected]>
2660 - added OS2 patches from Niclas Lindstrom <[email protected]>
2661 - fixed 'rewind bug'
2662 - added patches from Leo Broukhis <[email protected]>
2663 - added shuffle patch from Tillmann Steinbrecher <[email protected]>
2664 - added i386-getbits from M.Stekelenburg <[email protected]>
2665 - added fallback to 8bit if no 16bit samples support (currently linux only)
2666 - added (slightly changed) outburst patch
2667 from Antti Andreimann <[email protected]>
2668 - added 'http accept' patch from Hur TaeSung <[email protected]>
2669 - fixed buffer mode bug
2670
2671 0.59h: (MH)
2672 ------
2673 - added '--8bit' option to force 'ULAW 8 BIT' playing
2674 - pcm buffer type changed to 'unsigned char *'
2675 - plays 2:1 and 4:1 with the real frequency (ie 11025/22050 instead of 44100)
2676 - program (re)sets (changed) frequency for every frame
2677 - buffer mode is broken
2678
2679 0.59g: (MH)
2680 ------
2681 - added support for MPEG 2.5 (it's a HACK!) .. mpg123 understands and plays
2682 2.5 but it does it wrong. (you will hear this!)
2683 - Fixed wrong values in BandInfo struct for 32kHz mode
2684 - Some performance improvements and minor changes and bug-fixes.
2685 - Tried the new imdct from the maplay1.2+ (version 1.81) package. (but
2686 it was slightly slower on my system, currently disabled)
2687
2688 0.59f: (OF)
2689 ------
2690 - HP-UX audio fix (provided by Damien Clermonte)
2691 - Buffering (-b option) hopefully fixed. Please test!
2692
2693 0.59e:
2694 ------
2695 - added Audio MPEG 2.0 support (but near untested) (MH)
2696
2697 0.59d: (MH)
2698 ------
2699 - minor modifications to dct64_i386.c
2700 - removed duplicated dependency 'dct64.o' in Makefile
2701
2702 0.59c: (OF)
2703 ------
2704 - Another complete rework of the buffer system (-b option) (normally uses
2705 mmap(), but it can also use SYSV-ShMem because mmap() is broken on linux).
2706
2707 0.59: (MH)
2708 -----
2709 - extracted dct64() from decode*.c files.
2710 - some changes to dct64() and decode*.c to allow auto-increment/decrement
2711 - removed bit-counter in getbit-functions. Layer3 part2 now counts it itself.
2712
2713 0.58: (OF)
2714 -----
2715 - Bugfix: not closing the audio device caused a few samples to be lost.
2716 - Bugfix: audio_set_rate() was not called (!?!).
2717 - Additional command line options "--doublespeed" and "--halfspeed" (couldn't
2718 find better names). See the manpage.
2719 - Complete redesign of command line options. See the manpage.
2720 - Changed the buffer algorithm (-b). I'm not sure whether it's better or worse
2721 now.
2722 - Supports multiple files on the command line. Pressing Ctrl-C will skip to the
2723 next MPEG file, pressing it twice within one second will abort mpg123
2724 completely. See the "INTERRUPT" section in the manpage.
2725 - Support for http URLs. See also the -p option in the manpage.
2726 - Support for list files. See also the -@ option in the manpage.
2727
2728 0.57: (MH)
2729 -----
2730 - Put the downsampling code together with the full quality decode into one
2731 binary. (new options -2to1,-4to1)
2732 - Optimizations in the dct12.
2733 - Removed the 'int' version.
2734
2735 0.56: (OF)
2736 -----
2737 - Additional command line option "-rs" to try to resync on "broken" MPEG streams
2738 (now called "--resync" or "-y").
2739 - Additional command line option "-b" to use audio output buffering. See the
2740 manpage for additional information.
2741 - Changed a few variable names and some other minor fixes, so the
2742 source compiles more cleanly (without warnings) on most platforms (for
2743 example, the FreeBSD port compiles without warnings using "-Wall -ansi
2744 -pedantic").
2745 - Makefile changes: moved "solaris" to "solaris-gcc", new target
2746 "solaris" (using Sun's SparcWorks cc, faster tha gcc!), new targets
2747 "aix" and "generic".
2748
2749 0.55:
2750 -----
2751 - removed some obsolete code from layer3.c (MH)
2752 - changed scalefac stoing in layer3 (MH)
2753 - decode_4to1.c added again. Renamed decode_lp.c to decode_2to1.c (MH)
2754 (both not optimized .. still expterimental)
2755 - Minor cleanups and optimizations in layer3. (MH)
2756 - Removed some (probable) minor bugs. (MH)
2757 - Changed 2**(...) scaling (and omitted a few muls) (MH)
2758 - Changed Makefile (OF)
2759 - Changed dequantize to reorder with a mapping table
2760 (you can disable this by undefining MAP in layer3.c) (MH)
2761 - removed the 'dummy' scale step for mixed mode (MH)
2762 - Changed copy policy (MH)
2763 - Modified dequant-reorder-mapping (MH)
2764
2765 0.54: (MH)
2766 -----
2767 - Minor changes in the documents ..
2768 - huffman.c: x->len,y->len removed .. always '15' for the important cases
2769 - audio.c: some additions to the SOLARIS audio code
2770
2771 0.53b: (OF)
2772 ------
2773 - Makefile:
2774 -- Renamed SPARC10 to SUNOS (which is more appropriate).
2775 Tested with SunOS 4.1.4 on a SPARCstation SLC, 20 MHz (322% CPU,
2776 so no realtime decoding).
2777 -- Added SOLARIS. Tested with Solaris 2.5.1 on an Ultra 1, 167 MHz (14%
2778 CPU), a SPARCstation 5, 110 MHz (47% CPU), and a SPARCstation 10, 33 MHz
2779 (68% CPU).
2780 -- Added FreeBSD. Tested with 2.1.7 on a Pentium-90 (38% CPU) and 2.2 on a
2781 Pentium-166 (20% CPU).
2782 -- Added Digital Unix. Tested with DU 3.2 on a DEC 3000/500 (Alpha AXP
2783 150 MHz, 33% CPU). Does not support DEC's audio interface ("AudioFile"),
2784 decodes to stdout only.
2785 -- Improved "clean" target to remove core files and editor backup files.
2786 -- Added "install" target.
2787 -- Added "dist" target to create a .tar.gz file of the source directory.
2788 -- In the distributed Makefile, all system sections are commented out. When
2789 you type "make" without uncommenting one of the sections, an appropriate
2790 error message will be displayed.
2791 - audio.c
2792 Fixed problem when opening /dev/audio on Solaris: I had to remove O_NDELAY,
2793 because writing to /dev/audio didn't block until the requested number of
2794 samples could be written, but instead it wrote only as much samples as
2795 it was able too. Sounded funny. Removing O_NDELAY fixed the problem.
2796 I don't think that it hurts SunOS either, so I didn't make another #ifdef.
2797 - common.c
2798 Fixed bug by adding function finish_output(): The decoder writes the audio
2799 samples into a buffer, and each time the buffer is full, it is written
2800 to the audio device (or to stdout). But when the decoding is done,
2801 there are some samples left in the buffer which aren't written yet --
2802 they were just discarded. This additional function fixes the problem by
2803 writing the remaining samples from the buffer. It is called immediately
2804 after the main frame loop in mpg123.c.
2805 - decode*.c
2806 Fixed compiler warning on symbol "div": Some compilers issue warnings
2807 because the symbol div is already used for something else (see "man div"),
2808 so I changed it to "divv".
2809 - layer*.c Functions do_layer*:
2810 -- Moved the declaration for audio_play_samples into mpg123.h. Makes the
2811 code cleaner.
2812 -- I moved the variables pcm_samples and pcm_point to common.c, so they are
2813 shared between all modules. This was necessary because the finish_output()
2814 function (see common.c above) needs to access them.
2815 -- Renamed loop variables "i" and "j" to "k" in order to fix compiler warnings
2816 for duplicate/hidden symbols (similar to "div" in decode*.c, see above).
2817 Changed the preprocessor symbol AUDIOBUFSIZE into a global variable
2818 "audiobufsize". This makes it possible to change it by a command
2819 line option.
2820 - layer1.c
2821 Fixed error in function I_step_one, second parameter: unsigned int
2822 scale_index[2][] --> ...[2][SBLIMIT] This caused an error with the DU
2823 compiler (which is ok, because [2][] is incorrect according to ANSI).
2824 I don't know whether [2][SBLIMIT] actually works correctly (I have no
2825 layer-1 streams to test it), but at least it compiles without error, and
2826 it appears to be correct (looking at the rest of the source). Heaven knows
2827 why gcc doesn't issue an error, and what it actually makes from it.
2828 - mpg123.1
2829 A simple manual page. Type "nroff -man mpg123.1 | more" to view it, or
2830 install it in /usr/local/man/man1 (this is done automatically by "make
2831 install").
2832 - mpg123.c
2833 New boolean variable "quiet": If this variable is set to TRUE (command line
2834 option -q), no messages are printed.