Merge branch 't/cpsi_improvement'
[paraslash.git] / NEWS
1 NEWS
2 ====
3
4 -------------------------------------------------
5 0.5.3 (to be released) "symbolic synchronization"
6 -------------------------------------------------
7
8 - Various alsa-related fixes, mostly for the raspberry pi.
9 - The test suite has been extended to include sanity checks
10 for the generated man pages.
11 - ao_writer fixes. This writer was in a quite bad shape. Many
12 serious bugs have been fixed.
13 - new audiod command: version.
14 - Minor improvements to the bitstream API.
15 - The cpsi command now prints a meaningful error message if
16 none of the given patterns matched any audio file.
17
18 ----------------------------------------
19 0.5.2 (2014-04-11) "orthogonal interior"
20 ----------------------------------------
21
22 The new sync filter, the AES_CTR128 stream cipher and the overhauled
23 network code are the highlights of this release. It also includes a
24 fair number of smaller fixes and improvements not mentioned here.
25
26 - The new sync filter synchronizes playback between multiple
27 clients.
28 - Connections between para_server and para_client are now
29 encrypted by means of AES rather than RC4 if both sides
30 support it. RC4 is still available as a fallback. This
31 feature is fully transparent, i.e. no command line options
32 are necessary, and a client linked against openssl can
33 speak with a server linked against libgcrypt and vice versa.
34 - Major cleanup of the networking subsystem.
35 - Improvements to para_fade: the new set mode, multi-channel
36 initial volumes, better error logging.
37 - The man pages of para_audiod, para_filter, para_recv, and
38 para_write contain the relevant options for receivers, filters,
39 writers. This broke in 0.5.0.
40 - ogg/vorbis latency improvements.
41 - Improved user manual.
42 - Minor fixes to avoid clang warnings.
43
44 Downloads: ./releases/paraslash-0.5.2.tar.bz2 (tarball),
45 ./releases/paraslash-0.5.2.tar.bz2.asc (signature)
46
47 ------------------------------------------
48 0.5.1 (2013-12-20) "temporary implication"
49 ------------------------------------------
50
51 Lots of fixes and improvements all over the place, and a major overhaul
52 of the build system.
53
54 - Audiod improvements and fixes.
55 - Buffer tree robustness improvements.
56 - Cleanup of the mood subsystem.
57 - Fixes and cleanups for the flac decoder.
58 - Latency improvements for the ogg/opus decoder.
59 - Crypto support is now optional. On systems without
60 openssl/gcrypt, the build succeeds but para_server,
61 para_audiod, para_client won't be built.
62 - The build system now works for cross-compile setups.
63 - The dependency tree has been flattened, which speeds up
64 builds and avoids to recreate the man pages on every change.
65 - The error code helper has been rewritten from perl to C,
66 which further improves build time.
67 - Many small bugs in the build system have been identified
68 and fixed.
69
70 Downloads: ./releases/paraslash-0.5.1.tar.bz2 (tarball),
71 ./releases/paraslash-0.5.1.tar.bz2.asc (signature)
72
73 ----------------------------------------
74 0.5.0 (2013-08-23) "invertible validity"
75 ----------------------------------------
76
77 Some API-breaking changes, one serious bug fix, and a lot of bike-shedding.
78
79 - The sideband compatibility code has been removed, hence
80 sideband connections (introduced in 0.4.11) are now mandatory.
81 - Addblob commands can produce output.
82 - The stat command no longer sends garbage when para_server was
83 compiled against libgcrypt.
84 - Dependencies for gengetopt files are computed automatically.
85 This eliminates a constant source of build bugs.
86 - The setatt command now accepts file name patterns rather than only
87 path names.
88 - overview.pdf is now based on dia, a simple diagram creation program.
89 The new version is much more detailed and contains descriptions of
90 the various programs of the paraslash package.
91 - The separator of all multi-word options has been changed from
92 underscore to dash. For example --log_color becomes --log-color.
93 - Overhauled web pages and the new logo.
94
95 Downloads: ./releases/paraslash-0.5.0.tar.bz2 (tarball),
96 ./releases/paraslash-0.5.0.tar.bz2.asc (signature)
97
98 --------------------------------------
99 0.4.13 (2013-07-29) "spectral gravity"
100 --------------------------------------
101
102 One more 0.4.x release before the API-breaking changes for 0.5.0 go
103 in. The main features of this release are the ogg/opus audio format,
104 and UTF-8 support, but it includes also tons of other improvements
105 and fixes all over the place.
106
107 - New audio format: ogg/opus.
108 - UTF8 support for para_gui and the mp3 audio format handler.
109 - Scheduler improvements and fixes.
110 - The obsolete gettimeofday() function has been replaced
111 by clock_gettime() on systems which support it.
112 - Speed and usability improvements for para_gui.
113 - para_client now restores the fd flags of stdin and stdout
114 on shutdown.
115 - Improved manual pages.
116 - Consistent version strings for all executables.
117 - Reduced dependencies on generated files result in fewer
118 recompilations on changes.
119 - Performance improvements for the compress filter.
120 - Improved downloads web page.
121
122 -----------------------------------------
123 0.4.12 (2012-12-20) "volatile relativity"
124 -----------------------------------------
125 The new command line player, the resample filter, ALSA support for
126 para_fade, and the improved build system are the highlights of this
127 release which probably marks the end of the 0.4.x series.
128
129 - The afh receiver and the para_play executable.
130 - The resample filter: A sample rate converter based on
131 libsamplerate.
132 - The "versions" directory has been removed from the master
133 branch. The tarballs of the old releases are now available
134 in the new "releases" branch.
135 - Overhaul of the build system: All generated files are now
136 written to the "build" directory.
137 - The modular mixer API and the alsa mixer.
138 - Minor fixes for the osx writer.
139
140 --------------------------------------
141 0.4.11 (2012-07-20) "mutual diversity"
142 --------------------------------------
143
144 The major feature in this release is the new sideband API for
145 client-server communication. This API will be used exclusively starting
146 with 0.5.0, which breaks backward compatibility but allows to get rid
147 of quite some compatibility code. Other noteworthy changes include
148 decoder latency improvements and a long-standing bug fix for the
149 ALSA writer.
150
151 - Sideband connections: If both para_server and para_client
152 support this feature, data is sent as a multiplexed stream.
153 - The --no_default_filters option of para_filter has been
154 removed.
155 - Several fixes and latency improvements to various decoders.
156 - The ALSA writer now limits the prebuffer time to 500ms.
157 - Documentation improvements.
158 - Overhaul of the command_util.sh script.
159 - Fixes for some minor problems found by the clang analyzer.
160 - Compiles (almost) without warnings on gcc-3.
161 - Robustness improvements of the buffer tree code.
162
163 ------------------------------------------
164 0.4.10 (2012-03-30) "heterogeneous vacuum"
165 ------------------------------------------
166
167 Nothing earth-shaking in this release, but quite a few usability
168 improvements and the usual mix of cleanups and fixes.
169
170 - The --no_default_filters option of para_filter has been
171 deprecated. It still works but has no effect and will be
172 removed in the next version.
173 - para_gui now prints also the stderr output of the executing
174 command in the bottom window.
175 - Cleanup and consolidation of the various wrappers for
176 write(), writev(), send() and friends.
177 - The obscure error messages on mmap() failures have been
178 replaced by meaningful messages. This affects mainly
179 para_afh.
180 - para_audioc: Cleanups and memory leak fixes.
181 - Test 0004-server no longer fails if para_server is not
182 being built.
183 - New configure options: --with-id3tag-{headers,libs}.
184
185 -------------------------------------
186 0.4.9 (2011-12-06) "hybrid causality"
187 -------------------------------------
188
189 Support for another audio format, interactive mode for para_client
190 and para_audiod and many small improvements/fixes all over the place.
191
192 - Support for flac, the free lossless audio codec.
193 - Fix for an endless loop in the mp3 decoder for certain
194 (corrupt) mp3 files.
195 - When executed without specifying a command, para_client
196 and para_audioc start an interactive shell (requires
197 libreadline being installed). The interactive mode offers
198 full tab completion and command line history.
199 - autogen.sh now detects a distcc setup and adjusts the
200 parameter for the -j option of make accordingly.
201 - Shared memory areas are no longer restricted to 64K. We now
202 detect the maximal size of a shared memory area at runtime.
203 - cleanup of the internal uptime API.
204 - para_server prefaults the mmapped audio file to avoid
205 delays on slow media.
206 - A new test for the test-suite that exercises the
207 communication between para_server and para_audiod.
208 - The alsa writer eats up less CPU cycles when configured to
209 use the DMIX plugin.
210 - Simplified and unified receiver code.
211 - Makefile cleanups.
212 - Commands which print a list of matching audio files now
213 emit a meaningful error message if no audio file matched the
214 given pattern(s).
215
216 --------------------------------------
217 0.4.8 (2011-08-19) "nested assignment"
218 --------------------------------------
219
220 Gcrypt support, the overhauled osx writer and regex format specifiers
221 are the highlights of this release.
222
223 - support for libgcrypt as a drop-in replacement for openssl.
224 Run configure --enable-cryptolib=gcrypt to link against
225 libgcrypt. The two crypto implementations are compatible to
226 each other, i.e. a para_client executable linked against
227 gcrypt can connect to para_server linked against libssl
228 and vice versa.
229 - Rewrite of the osx writer (output plugin for Mac OS).
230 - audiod: The format specifier for receivers, filters and
231 writers is now treated as a regular expression. This allows
232 to replace 5 lines in the config file (one for each audio
233 format) by one single line. See the manual for details.
234 - The *.cmdline.[ch] files are no longer contained in the released
235 tarballs. This reduces the size of the tarballs but requires
236 gengetopt to build the tarball.
237 - Compiles cleanly also with llvm/clang.
238 - Corrupt mp3 files are handled more gracefully.
239 - The alsa writer uses poll fds instead of computing timeouts.
240 - Cleanup of the generic writer API.
241 - sched: Optimized zero timeouts.
242 - vss timeout cleanups.
243 - oggdec fixes and improvements.
244
245 --------------------------------------
246 0.4.7 (2011-06-01) "infinite rollback"
247 --------------------------------------
248
249 The new ao writer, support for ssh RSA keys and a couple of other
250 enhancements.
251
252 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
253 platforms through the libao audio library.
254 - Support for RSA keys generated with ssh-keygen.
255 - configure: improved options for ogg/vorbis/speex.
256 - The git version reported by --version always matches HEAD.
257 - The autogen script detects the number of processors and
258 runs a parallel make if possible.
259 - Major cleanup of the crypto API.
260 - Documentation updates.
261
262 ------------------------------------------
263 0.4.6 (2011-03-31) "deterministic entropy"
264 ------------------------------------------
265
266 Lots of ogg/vorbis improvements, the new test suite, enhancements
267 for para_gui and a fair amount of other bug fixes.
268
269 - For DCCP/OGG streams the audio file header is only sent once
270 at the beginning of the stream rather than periodically
271 every five seconds. This reduces network traffic and the
272 FEC group size.
273 - The vorbis comment header is replaced by an empty dummy header
274 before the header is sent over the network. This also results in
275 less network traffic and smaller FEC groups.
276 - The new "test" make target allows to perform some sanity checks prior
277 to installing the package.
278 - ogg timing fixes and performance improvements
279 - Scheduler improvements
280 - Proper exit codes for para_write
281 - para_gui: New option --theme to select a startup theme. Several
282 other improvements and fixes.
283 - aacdec error message cleanups
284 - simplified color error handling
285
286 --------------------------------------------
287 0.4.5 (2010-12-17) "symmetric randomization"
288 --------------------------------------------
289
290 Bug fixes, internal cleanups and variable-sized FEC slices.
291
292 - Contains a fix for an invalid-free-bug in the ogg audio format
293 handler code.
294 - Switching off the DCCP sender works again.
295 - para_audiod handles crashes of para_server more robustly.
296 - Internal scheduler and writer cleanups.
297 - Reduced latency due to variable-sized FEC slices.
298 - Improved documentation and error diagnostics.
299 - The build of para_server is now optional, allowing the build
300 to succeed in case libosl is not installed.
301
302 ------------------------------------------
303 0.4.4 (2010-08-06) "persistent regularity"
304 ------------------------------------------
305
306 Support for yet another audio format, para_write improvements and
307 bug fixes.
308
309 - Support for the speex codec.
310 - Support for sample formats other than 16 bit little endian.
311 - error2.h is now created by a perl script which speeds up configure
312 considerably.
313 - Fix a bug in the aac decoder which could lead to segfaults in
314 para_filter/para_audiod.
315 - Fixes for autoconf-2.66.
316
317 ----------------------------------------
318 0.4.3 (2010-07-05) "imaginary radiation"
319 ----------------------------------------
320
321 Many improvements for the DCCP and the UDP transport, the new user
322 manual and the usual mix of bug fixes and internal improvements.
323
324 - FEC support for the DCCP sender (Gerrit Renker). The new
325 --dccp_max_slice_size, --dccp_data_slices_per_group and
326 --dccp_slices_per_group options can be used to set the FEC
327 parameters for the DCCP transport.
328 - DNS lookups for UDP targets (Gerrit Renker).
329 - The new user manual replaces the README, README.afs, REQUIREMENTS
330 and INSTALL documents.
331 - Fix an end-of-file detection bug in the oggdec filter.
332 - The new nonblock API.
333 - Both options of the oggdec filter have been removed.
334 - New debug mode for the internal scheduler.
335
336 ------------------------------------------
337 0.4.2 (2010-04-23) "associative expansion"
338 ------------------------------------------
339
340 It's been some time since the last release, but finally here is
341 paraslash-0.4.2. The bulk of the changes comes from the new buffer
342 tree API, but there are changes all over the tree. Mainly performance
343 and usability improvements, but also quite some bug fixes.
344
345 - The new buffer tree API.
346 - DCCP: Support for CCID negotiation (Gerrit Renker).
347 - UDP robustness fixes.
348 - The --bufsize option for mp3dec is gone as it no longer makes sense
349 for the new buffer tree API.
350 - Fix audible buffer underruns for wma streams.
351 - The alsa writer no longer prints meaningless underrun durations.
352 - audiod: Defaults work also for udp streams. If no filter is
353 given for an audio format that is received via upd, fecdec is
354 automatically added as the first filter (along with the decoder).
355
356 ---------------------------------------
357 0.4.1 (2009-12-22) "concurrent horizon"
358 ---------------------------------------
359
360 Support for another audio format, minor feature enhancements and lots of bug
361 fixes. All fixes that have been accumulated in the maint branch (in particular
362 those mentionened in the 0.3.6 release notes) appear in this release as well.
363
364 - wma support.
365 - new afh option: --human to activate human-readable output.
366 - new server/audiod option: --log-timing to print timing information.
367 - build system improvements.
368 - source code documentation updates.
369
370 -------------------------------------
371 0.3.6 (2009-12-07) "cubic continuity"
372 -------------------------------------
373
374 Quite a few bugs have been found and fixed since 0.3.5, so here's
375 another 0.3.x release. No new features.
376
377 - Always check return value of malloc().
378 - ogg vorbis/FEC: Do not write garbage after the audio file header.
379 - exit if root privileges could not be dropped.
380 - FEC: Fix computation of extra slices.
381 - oss: Fix check for empty input buffer.
382 - Avoid buffer underruns due to filter chain output buffer constraints.
383 - server: Fix assignment of afs_pid.
384 - Don't panic if the afs database contains unknown audio formats.
385 - http/dccp: Do not send the audio file header twice.
386 - FEC: Timing improvements.
387
388 ----------------------------------------------
389 0.4.0 (2009-11-10) "simultaneous independence"
390 ----------------------------------------------
391
392 Two significant changes which require the new version number: The
393 improved authentication dialog and the fact that the database code
394 has been moved to a library, libosl. To use the new version, you have
395 to generate new RSA keys, see INSTALL for details. A shell script is
396 provided for conversion of the 0.3 database to the new 0.4 format.
397
398 - stronger crypto for client authentication
399 - the database code has been moved to a library
400 - improved status item handling
401 - cleanup of the build system
402 - The "-V" option now also prints the git version
403 - the new parser-friendly listing mode for the ls and stat commands
404 - mandatory rc4 encryption
405 - major audio format handler cleanups
406 - (id3,...) tags are no longer stored as a combined string in the database
407 - new mood methods: artist_matches, title_matches, comment_matches,
408 album_matches, year_maches, year.
409
410 --------------------------------------------
411 0.3.5 (2009-09-21) "symplectic separability"
412 --------------------------------------------
413
414 Full client support for *BSD Unixes, complete re-write of the ogg
415 vorbis audio format handler, various improvements all over the place
416 and the usual mix of bugfixes. This release marks the end of the 0.3
417 series if no serious problems show up.
418
419 - the new oss writer (supported on *BSD and Linux)
420 - rewrite of the ogg vorbis audio format handler. It's
421 recommended to replace the chunk tables of existing ogg
422 vorbis files in the afs database by re-adding these files
423 with "add -f".
424 - support for netmask subsets (Gerrit Renker)
425 - the new prebuffer filter
426 - improved signal handling
427 - variable fec output buffer size
428 - improved FEC timing fixes audible buffer underruns in UDP mode
429 - --log_color actually works
430 - new ls option: -d (print dates as seconds after the epoch)
431 - update to gengetopt 2.22.2
432 - support for RSA keys of size > 512 bits
433 - new option "mixer_channel" for para_fade
434
435 -----------------------------------------
436 0.3.4 (2009-05-07) "elliptic inheritance"
437 -----------------------------------------
438
439 The new udp sender, forward error correction, colored logs and various
440 other improvements. As the udp sender does not depend on any special
441 libraries, it is built unconditionally.
442
443 - The udp sender replaces the ortp sender. The new code uses forward
444 error correction to protect against packet losses. Many thanks to
445 Gerrit Renker for providing ipv6 support.
446 - The default port for udp streaming now defaults to 8000, like
447 for the http and the dccp senders/receivers.
448 - Loglevels are now specified as symbolic names, e.g.
449 "--loglevel info".
450 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
451 - para_server/para_audiod: Color support for log messages.
452 - new options for mp3dec: --ignore-crc, --bufsize
453 - new audiod option: --config-file.
454 - gengetopt cleanups.
455 - Improved help/man pages: The documentation of para_audiod,
456 para_recv, para_filter and para_write now also contains
457 all options of the available receivers/filters/writers. The
458 man page of para_fade contains a description of the different
459 modes of operation.
460 - More source code documentation.
461 - vss timing fixes.
462
463 --------------------------------------------
464 0.3.3 (2008-12-01) "axiomatic perspectivity"
465 --------------------------------------------
466
467 Internal code cleanups, bug fixes, improved tag handling and the new
468 amplification filter.
469
470 - para_server uses the generic scheduling code.
471 - overhaul of the virtual streaming system.
472 - mp3: id3 version 2 support via libid3tag (optional)
473 - ogg: vorbis comment support.
474 - aac meta info support.
475 - mp3 audio format handler cleanups.
476 - new filter: "amp" to amplify the amplitude of the audio stream
477 - new status item/database entry: amplification. It is
478 used by the amp filter to pre-amplify the audio stream.
479 - fix a close-without-open bug in para_write.
480 - fix a bug in com_init() which was introduced in 0.3.2.
481 - better error diagnostics for para_client.
482
483 -----------------------------------------
484 0.3.2 (2008-04-11) "probabilistic parity"
485 -----------------------------------------
486
487 The new para_afh executable, scheduling and documentation improvements.
488
489 - new ls option: -lc (list chunk table)
490 - new executable: para_afh, the stand-alone audio file handler tool
491 - afs commands can send output more than SHMMAX (32MB on Linux). This
492 also reduces the memory usage of commands that produce large amounts
493 of output.
494 - major scheduler and audiod cleanups.
495 - more detailed and much nicer man pages.
496
497 ---------------------------------------
498 0.3.1 (2008-02-23) "liquid interaction"
499 ---------------------------------------
500
501 A mix of cleanups, bug fixes, improvements, and some new features. No
502 significant changes to the new database (osl) code, which is generally
503 a good sign.
504
505 - Share some similar/duplicate code between the http and the
506 dccp sender.
507 - Generic access control lists for paraslash senders.
508 - dccp sender: Access control lists, connection limiting and support
509 for the allow,deny,on,off,help sender commands.
510 - The default dccp port changed from 5001 to 8000 (suggested by
511 Gerrit Renker).
512 - para_server starts even if not all public keys could be loaded.
513 - Audiod performance improvements.
514 - fix a bug in the "off" command of the http sender.
515 - fix some fd and memory leaks.
516 - Update to gengetopt-2.22.
517
518 -------------------------------------
519 0.3.0 (2008-01-12) "solar saturation"
520 -------------------------------------
521
522 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
523
524
525 Usually one might expect lots of new features AND a big increase in size
526 for a major release like this.
527
528 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
529 ever. The decrease in size is mostly due to the removal of some
530 graphical tools (which were only quick hacks anyway). But also the
531 fact that the mysql code is gone cuts down the size a bit.
532
533 Being independent of mysql comes at a cost: The fact that paraslash
534 now contains its own database (the object storage layer, osl) increases
535 the (stripped) binary size of para_server by ~50K on i386.
536
537 - no more restrictions on unique basenames.
538 - independent of mysql: The new self-contained object
539 storage layer (osl) replaces the mysql database.
540 - New executable para_fsck: Check integrity of osl tables.
541 - Lyrics support.
542 - Reliable audio file move/rename detection.
543 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
544 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
545 Solaris (sparc64).
546 - the new osl-based audio file selector (afs) replaces the random,
547 the playlist and the mysql selector of paraslash-0.2.x.
548 - IPv6 support (thanks to Gerrit Renker).
549 - paraslash-0.2.x streams are now called "moods". Writing
550 0.3.x-mood definitions should be both easier and more
551 powerful than writing 0.2.x-stream definitions.
552 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
553 been removed. The world is a better place without them. However,
554 para_gui is still there.
555 - afs tracks audio file selection also in playlist mode.
556 - few easy-to-use afs commands replace the many not-so-easy-to-use
557 mysql commands (and are available also in playlist mode).
558 - Improved error subsystem.
559 - The earth-shaking new logo.
560
561 -----------------------------------------
562 0.2.17 (2007-11-20) "isotropic threshold"
563 -----------------------------------------
564
565 Mainly bugfixes and cleanups in this version which marks the end of
566 the 0.2.x series if no serious bugs show up after the release.
567
568 - mysql_selector: fix a locking bug.
569 - universal chunk queueing.
570 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
571 to Gerrit Renker).
572 - be more carful wrt. signed vs. unsigned argument passing.
573 - cleanup error.h and fix some references to invalid error
574 codes.
575 - update to gengetopt-2.21.
576 - update to ortp-0.13.1.
577 - autoconf: extend checks for headers, library functions and
578 compiler characteristics.
579 - Fix streaming of large mp3 files.
580 - Fix an off-by-one bug in playlist handling.
581
582 --------------------------------------
583 0.2.16 (2007-04-05) "neural discharge"
584 --------------------------------------
585
586 The main change in this release is the major audio format handler
587 cleanup which removes some similar/duplicate code and makes it easier
588 to implement plugins for other audio formats. Of course, the usual mix
589 of other improvements/changes/bugfixes also made it into the release.
590
591 - simplified audio format handlers (most of the handling functions
592 were moved one layer up to the virtual streaming system).
593 - para_server uses mmap to read audio files
594 - repositioning of mp3 streams is much faster, in particular for
595 jumping near the end of large mp3 files.
596 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
597 and AFS_WRITE.
598 - fix a bug in para_filter that caused decoding of aac files
599 to start only after a few seconds.
600 - fix osx_writer hangs
601 - simplified dccp code (thanks to Gerrit Renker)
602 - the compress filter works also on big endian systems (ppc)
603
604 -----------------------------------------
605 0.2.15 (2007-02-16) "inductive resonance"
606 -----------------------------------------
607
608 Minor improvements, more documentation and a bunch of bug fixes.
609
610 - para_server: The server.users file is only read once on server
611 startup rather than for each connection
612 - mp3dec: Fix decoding of corrupt mp3 files
613 - afs (audio file sender) is now called vss (virtual streaming
614 system). Consequently, the permission flags specified in
615 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
616 become VSS_READ and VSS_WRITE respectively.
617 - para_audiod/para_filter: Fix a bug that caused the last chunk
618 of audio data not being written under certain circumstances
619 - audiod: compute the difference of server time and local time
620 correctly
621 - para_server/para_audiod: Fix some memory leaks
622 - documentation improvements
623 - configure.ac: fix checks for para_krell
624 - new man pages
625
626 -------------------------------------------
627 0.2.14 (2006-10-15) "transient singularity"
628 -------------------------------------------
629
630 The only major enhancement of this version is the osx writer which completes
631 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
632 until now for reasons beyond the scope of this changelog entry.
633
634 - new output plugin for Mac Os: the osx writer
635 - rename configure command line options from --enable-xxx-headers to
636 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
637 - configure: new command line options: --with-mad-headers,
638 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
639 - some robustness fixes
640 - dymamic audio format recognition for audiod
641 - para_server: new command line option: --autoplay_delay
642 - para_audiod: new command line option: --clock_diff_count
643
644 ---------------------------------------
645 0.2.13 (2006-07-14) "sonic convolution"
646 ---------------------------------------
647
648 A bunch of new features and core changes.
649
650 - the new paraslash scheduler, short and sweet.
651 - Support for m4a/mp4 files via the new aac audio format
652 handler/filter (requires libfaad).
653 - each writer has its own command line parser, just like
654 para_recv and para_filter.
655 - para_client and para_audioc use the error subsystem
656 - writers are integrated in para_audiod (currently linux-only)
657 - para_client is integrated in para_audiod
658 - random/playlist selector: improved info strings
659 - new audiod commands: tasks, kill
660 - update to libortp-0.10.1
661 - para_fade: wake time defaults to 8 hours from now
662 - update to autoconf-2.60
663
664 ------------------------------------------
665 0.2.12 (2006-05-12) "oriented abstraction"
666 ------------------------------------------
667
668 Many user-visible changes in this release and lots of new
669 features:
670
671
672 - the new optional dccp sender/receiver. It uses the datagram
673 congestion control protocol. You'll need a fairly new kernel
674 for this.
675 - paraslash works on Mac OS X (thanks to Gerd Becker)
676 - para_play renamed to para_write
677 - modular output plugin design (writers) for para_write
678 - new file_writer output plugin for para_write
679 - compress filter speed improvements
680 - update to libortp-0.9.1
681 - update to gengetopt-2.17rc
682 - para_client no longer depends on libreadline (as the
683 code for the interactive mode was removed).
684 - gcc-2-95 is no longer a supported compiler. It may still
685 work, but it gets no more testing.
686 - the tarball no longer contains the screenshot images which
687 reduces its size quite a bit.
688 - configure: new command line options: --enable-mysql-headers
689 and --enable-mysql-libs
690
691 ------------------------------------
692 0.2.11 (2006-03-11) "atomic duality"
693 ------------------------------------
694
695 Here it is, the first paraslash release developed with git. There
696 are fairly many user-visible changes in this release. As two out of
697 the three "database tools" of paraslash don't use a database at all,
698 they are now called "audio file selectors" instead.
699
700
701 - the cdt command (change database tool) becomes chs (change
702 selector)
703 - no more colon separators: The syntax of some options of
704 para_audiod and para_filter have changed. Use --help for
705 more info (and some examples).
706 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
707 - switch from cvs to git (should've done that earlier)
708 - the new ipc subsystem
709 - new audio file selector: playlist
710 - para_server: the dopey selector is now called "random",
711 and is the default selector. Use the --selector option to
712 choose another selector at startup, or the chs command to
713 change the selector at runtime.
714 - X86_64 fixes (thanks to Steffen Klassert)
715 - para_play fixes
716
717 --------------------------------------
718 0.2.10 (2006-02-17) "cyclic attractor"
719 --------------------------------------
720
721 Huge documentation update, a scrollable window for para_gui, ortp
722 improvements, and of course many small fixes not mentioned here.
723 The diffstat below is rather misleading as most insertions are due
724 to the new source documentation.
725
726 - autoconf cleanup
727 - para_server also uses the new error subsystem
728 - lots of new documentation (UTSL)
729 - gui improvements:
730 - keysyms for cursor keys and for next/previous page keys
731 - scrollable output window
732 - new internal commands: scroll up/down, page up/down
733 - fix color of command output.
734 - ortp: the --chunk_time and --header flags are no longer needed
735 for para_recv/para_audiod as this information is now encoded in
736 each rtp packet sent by para_server.
737
738 -------------------------------------------
739 0.2.9 (2006-01-24) "progressive turbulence"
740 -------------------------------------------
741
742 Internal audiod receivers/filters, the new error subsystem and
743 a lot of small improvements.
744
745 - para_recv and para_filter are integrated into the para_audiod
746 binary, i.e. audiod no longer spawns a new process for
747 each receiver/filter. As para_recv and para_filter might be
748 useful as standalone programs, they still get built (linked
749 against the same object files that are also used for audiod).
750 - further ortp timing improvements should reduce the CPU usage
751 of the ortp receiver.
752 - improved audio grabbing. The 'grab' command of para_audiod
753 has its own set of command line options. Read the output of
754 "para_audioc help grab" for more info.
755 - oggdec: configurable input prebuffer size.
756 - the new error subsystem gives better error diagnostics
757 and reduces code size.
758
759 -----------------------------------------
760 0.2.8 (2006-01-02) "dynamic accumulation"
761 -----------------------------------------
762
763 The new modular filter design and the para_play-hangs bugfix.
764
765 - new executable: para_filter. It combines para_mp3dec,
766 para_oggdec and para_compress. It also adds a further filter
767 type, wav, that just inserts a wave header at the desired point
768 of the filter chain. All 'piping' is done in-memory (i.e. no
769 read/write operations are used).
770 - para_play: fix a stupid bug that caused it to hang under
771 certain circumstances.
772
773 -------------------------------------------
774 0.2.7 (2006-12-27) "transparent invariance"
775 -------------------------------------------
776
777 Not many user-visible changes but a fair amount of internal improvements.
778
779
780 - The http sender buffers data if it can not be sent
781 out immediately (because the socket is not writable). This
782 should prevent para_server from shutting down the connection
783 too early on a loaded network.
784 - para_play also prebuffers data if it is told to start at a
785 future time by the --start_time option.
786 - The return of para_recv: It combines para_ortp_recv and
787 para_http_recv. Use the --receiver option to switch between
788 the two. para_recv builds without libortp, but contains
789 only the http receiver in this case.
790 - update to ortp 0.8.1. As this ortp release contains incompatible
791 changes, para_recv-0.2.7 won't link against older ortp libs.
792 - improved ortp timings.
793 - use of gcc-extensions that #define away for non-gcc and
794 gcc < 3.0.
795
796 -------------------------------------------
797 0.2.6 (2005-10-29) "recursive compensation"
798 -------------------------------------------
799
800 Transparent session encryption (uses openssl's Alleged RC4 cipher),
801 the internal find command and several other improvements and cleanups.
802
803 - Encrypt the session if encryption is requested by the client
804 (default for para_client 0.2.6). This is backwards
805 compatible, so older clients can still connect to para_server
806 0.2.6. Use the new client option --plain to request an
807 uncrypted session (off by default, must be set to on in
808 order to connect to para_server 0.2.x with 0 <= x <= 5).
809 - para_server uses an internal function to locate audio files
810 rather than calling find(1). The server option
811 --mysql_audio_file_dir replaces --mysql_find_cmd.
812 - documentation update
813 - man pages
814 - header file cleanup
815 - para_client code cleanup
816 - para_gui: faster display of output of display commands
817
818 ------------------------------------------
819 0.2.5 (2005-10-13) "aggressive resolution"
820 ------------------------------------------
821
822 This release adds internal senders, i.e. no more external programs are
823 spawned for sending out the audio data. There are two different senders
824 available: The http sender and the ortp sender (former para_send which
825 is no longer needed).
826
827 The new sender code has a plugin-like design so it can be easily
828 extended should there be be any future need for supporting another
829 network streaming protocol. All senders are completely independent of
830 each other. In particular, the http and the ortp sender can operate
831 in parallel.
832
833 - new server command: sender to control senders at runtime.
834 Read the output of "para_server -h" and "para_client help
835 sender" for more information.
836 - para_recv renamed to para_ortp_recv
837 - new executable: para_http_recv, a simple command line
838 http receiver.
839 - major afs/mp3/ogg code simplifications due to internal
840 senders.
841 - ogg timing improvements
842 - fix several minor memory leaks (found by valgrind)
843 - empty stream definitions work again
844 - com_ne(): ignore errors on remove
845 - audiod: fix segfault on server restart
846
847 ---------------------------------------
848 0.2.4 (2005-09-21) "toxic anticipation"
849 ---------------------------------------
850
851 Several small improvements, fixes and the new grab command.
852
853 - audiod:
854 - new command: "grab" to grab the output of the stream reader
855 or any filters. Read the output of "para_audioc help grab"
856 for more information.
857 - fix memory leak
858 - code cleanup
859 - audioc: new command line option: --bufsize to specify a
860 buffer size different from the default size 8192.
861 - improved error diagnostics for para_play.
862 - new configure option: --enable-ssldir so search for openssl in
863 non-standard places
864 - sdl_gui: Make it look nice again for 1024x768
865 - server: report total size of memory allocated with sbrk by malloc,
866 new command line option: --announce_time
867
868 -----------------------------------------
869 0.2.3 (2005-09-01) "hydrophilic movement"
870 -----------------------------------------
871
872 Two new executables and major feature enhancements.
873
874 - audiod filters: It is now possible to specify arbitrary many
875 (including none) filters for each supported audio
876 format. This can be used e.g. for normalizing volume,
877 transforming or grabbing the audio stream, or for using
878 visualizers. Read the output of "para_audiod -h" for the
879 syntax of the new --filter_cmd option.
880 - new executable: para_play, a tiny alsa player. It
881 can play wave files or raw pcm (16 bit little endian)
882 from stdin.
883 - new executable: para_compress, a dynamic range compressor
884 intended to keep audio output at a consistent volume. Derived
885 from AudioCompress, http://trikuare.cx/code/AudioCompress.html.
886 - audiod: New option: --stream_delay. This can be used in
887 a local network to syncronize the audio output of all
888 clients that play the same stream.
889
890 ------------------------------------------
891 0.2.2 (2005-08-19) "tangential excitation"
892 ------------------------------------------
893
894 Mostly internal changes in this release, but also some new commands
895 for the mysql database tool.
896
897 - cleanup exec.c, fix para_exec bug
898 - compile time loglevel (log messages below the given level
899 won't be compiled in, which reduces the size of the
900 resulting binaries)
901 - new log macros that shorten the size of the source code.
902 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
903 to send only zeros. With this workaround, para_audioc works
904 again.
905 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
906 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
907 (new entry), "snp" (set numplayed). Read the manual for more
908 information.
909
910 ---------------------------------------
911 0.2.1 (2005-08-15) "surreal experience"
912 ---------------------------------------
913
914 Here comes paraslash-0.2.1. It contains a couple of new features and,
915 surprise, only minor bug fixes.
916
917 - kill noisy mp3 debug message
918 - cleanup of the build system
919 - para_server and para_client directly use the crypto routines
920 of the openssl library rather than invoking the openssl command
921 line utitlity
922 - server/audiod: new option --user to switch to the given user
923 when invoked as root. Read the output of "para_server -h" for
924 more information.
925 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
926 status. Default: "para_audioc stat"
927 - sdl_gui: new option --pic_cmd to be used to download the picture.
928 Default: "para_client pic"
929 - audiod: 5 slots ought to be enough for everybody
930 - audiod: new status item: Uptime, kill hup command
931
932 ------------------------------------------
933 0.2.0 (2005-08-06) "distributed diffusion"
934 ------------------------------------------
935
936 After several month of increased development activity, paraslash-0.2.0
937 has arrived. It contains many new features and is much more
938 self-contained than the old 0.1.x series. Enjoy!
939
940
941 - para_server: fix hang on song change and crash on sighup.
942 Speed up mysql queries. The DIR_LIKE macro is gone.
943 - new executables: para_audiod, the local audio daemon that
944 starts playback (uses SCM_CREDENTIALS socket magic) and
945 para_audioc, the corresponding client.
946 - new executables: para_mp3dec/para_oggdec, two really teensy
947 decoders. para_mp3dec is based on libmad, para_oggdec requires
948 libvorbisfile.
949 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
950 they are now called para_send and para_recv respectively.
951 - documentation updates
952 - para_gui is themable. For now there is the default theme that
953 looks as before and the simple theme: blue and easy.
954 - gui: audio streaming is now handled by audiod. Time display shows
955 playback time rather than streaming time
956 - slider: update to libzmw-0.2.0
957 - para_krell: fix crash on server shutdown
958 - switch from gzip to bzip2
959
960 ----------------------------------------
961 0.1.7 (2005-04-18) "melting penetration"
962 ----------------------------------------
963
964 The main change in this release is clearly the oggvorbis rewrite,
965 but there are also lots of smaller changes. If you intend to use both
966 the mp3 and the ogg plugin, it is recommended to use software mixing,
967 e.g. the dmix plugin which is provided by ALSA.
968
969 - new executables: para_ovsend and para_ovrecv for sending/receiving
970 oggvorbis files via rtp. Requires the open rtp library. Get it at
971 http://www.linphone.org/ortp/
972 - rewrite of the ogg_vorbis core code
973 - configure detects libzmw and, if detected, includes
974 para_slider to the list of binaries to be built by make
975 - server stream writers read from their associated fifo rather
976 than from stdin
977 - slider: two new sliders, lastplayed and numplayed
978 - fix nasty double free bug which caused random segfaults in case of
979 mp3 files with invalid header information
980 - gui: new command line option: --stream_timeout=seconds to
981 deactivate a slot if it is idle for that many seconds (default=`5')
982 - diffstats
983
984 ---------------------------------------
985 0.1.6 (2005-03-05) "asymptotic balance"
986 ---------------------------------------
987
988 Only little user-visible changes in this release. Mainly bugfixes and
989 core code cleanup. This is probably the most stable version ever if you
990 stick to mp3...
991
992 - fix several memory leaks
993 - rename default name of mysql database from "music" to "paraslash".
994 Use para_server's --mysql_database option if you do not want to
995 rename your old database.
996 - rework ogg vorbis code
997 - make update command work on mysql servers with LOCAL_INFILE
998 disabled
999 - gui: improved stream I/O (slots)
1000 - simplified audio format API
1001 - para_pob_ogg is gone
1002
1003 ------------------------------------
1004 0.1.5 (2004-12-31) "opaque eternity"
1005 ------------------------------------
1006
1007 Let's slide gently into the new year.
1008
1009 - new: para_slider (not built automatically, type "make
1010 para_slider" to build). A toy for those who always felt that
1011 creating stream definitions is difficult. See screenshots,
1012 README and FEATURES for more info.
1013 - improved signal handling. Fixes server segfault on SIGHUP
1014 for linux kernels newer than Aug 24 2004 and makes para_gui
1015 race-free.
1016 - reload database tool on SIGHUP
1017 - improved help message for sl
1018 - do not log "broken pipe" messages as errors. They are
1019 perfectly ok.
1020 - fix wrong error message on permission errors
1021
1022 -----------------------------------------
1023 0.1.4 (2004-12-19) "tunneling transition"
1024 -----------------------------------------
1025
1026 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
1027 Hopefully, most of them got wiped out with this release. Some
1028 enhancements went also in.
1029
1030 - improved error diagnostics for all commands
1031 - stradd/picadd: overwrite previous contents if entry already
1032 exists, rather than returning errors
1033 - stradd: use current stream if invoked without args
1034 - faster (and hopefully more stable) ogg-vorbis handling
1035 - para_krell: reap children to avoid zombie-flooding in case
1036 no server is running
1037 - si: report also server pid
1038 - server: don't busy-loop if dbtool reports only invalid files.
1039 - gui: CTRL+C works again, fix stream_read command line option
1040 - fix pic_add, hist
1041 - fix mysql dbtool startup in case no database exists
1042 - many small fixes and cleanups
1043
1044 ---------------------------------------
1045 0.1.3: (2004-12-10) "vanishing inertia"
1046 ---------------------------------------
1047
1048 Starting from this release, the database tools are integrated in the
1049 server binary. This decreases server startup time, reduces code size
1050 and speeds up database commands. However, the layout of the underlying
1051 mysql database changed only slightly and 0.1.3 should be backwards
1052 compatible in that respect.
1053
1054 Visible changes:
1055
1056 - If mysql is not detected at compile time, or fails to init
1057 at runtime, fall back to the dopey database tool which should
1058 always work.
1059 - para_dbtool and dbtool.conf are gone. All mysql specific
1060 options are read from server.conf and are prefixed by 'mysql_'.
1061 - new command: cdt (change database tool)
1062 - new command line option: dbtool (choose startup database tool)
1063 - The name of current stream is now stored in the database,
1064 so paraslash remembers its current stream when restarted.
1065 - new command: csp (change stream and play)
1066 - para_gui also reports current database tool and server uptime
1067
1068 -------------------------------------------
1069 0.1.2: (2004-11-28) "spherical fluctuation"
1070 -------------------------------------------
1071
1072 Point release before the big dbtool changes go in.
1073
1074 - dbtool: rename ca to cam (copy all meta data). It now also
1075 copies numplayed and lastplayed time as well as the picture
1076 id.
1077 - fix endless-loop-bug caused by mp3 files with invalid header
1078
1079 -----------------------------------------
1080 0.1.1: (2004-11-05) "floating atmosphere"
1081 -----------------------------------------
1082
1083 - gkrellm plugin
1084 - new dbtool command: mbox. Browse your sound-file collection
1085 with your favorite mail reader.
1086 - several small fixes
1087
1088 -------------------------------------
1089 0.1.0: (2004-10-22) "rotating cortex"
1090 -------------------------------------
1091
1092 - fix logging bug for loglevel > VERBOSE
1093 - fix skip command
1094 - correct timings for vbr mp3s
1095 - modular audio format support
1096 - ogg-vorbis support (experimental)
1097 - new server option: autoplay
1098
1099 -----------------------------------------
1100 0.0.99: (2004-07-25) "harmonic deviation"
1101 -----------------------------------------
1102
1103 - rename projectname from icc to paraslash (play, archive, rate
1104 and stream large audio sets happily)
1105 - paraslash is no longer restricted to one particular audio
1106 streaming software
1107 - new dbtool commands (stradd, strq, strdel) for easy stream
1108 managment w/o configuration file. That obsoletes stream_defs
1109 file/config option for dbtool.
1110 - picadd accepts jpeg data from stdin
1111 - new server commands: ps (select previous stream), sc (song change)
1112 - new default pictures for sdl_gui
1113 - gui: new key_map option for binding commands and internal
1114 functions to arbitrary keys, nice help screen, rip out
1115 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1116 show silly logo on startup
1117 - new executables: para_fade for fading volume, para_dbadm for
1118 manipulating attributes
1119 - cdb adds _all_ tables to mysql database
1120 - revised and beautified documentation
1121 - sample dbtool rewritten in C
1122 - autoconf
1123
1124 ---------------------------------------------
1125 0.0.98: (2003-12-26) "incremental smoothness"
1126 ---------------------------------------------
1127
1128 - kick icecast in favour of poc. That removes some races and reduces
1129 core code considerably.
1130 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1131 give finer info on afs' status.
1132 - gui can start decoder (see config options). Further new gui
1133 commands: refresh (^L), jmp (F1-F10)
1134 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1135 terminates gui. This fixes dead instances consuming memory
1136 continuously.
1137 - new dbtool command: verb for sending verbatim sql queries.
1138 - fix pid_list races (by removing pid_list)
1139 - codename funnies
1140
1141 --------------------
1142 0.0.97: (2003-10-26)
1143 --------------------
1144
1145 - installation prefix now defaults to /usr/local
1146 - new commands for gui: snozze, sleep and reread config
1147 - config file for gui and sdl_gui
1148 - fix problems with filenames containing funny characters
1149 (reported by Thomas Forell)
1150 - improved signal handling for gui, now it rereads conf on SIGHUP
1151 - new dbtool command: cdb (create database)
1152 - switch from argtable to gengetopt
1153 - major code cleanup and speed improvements
1154 - fix several potential buffer overflows
1155 - many small fixes and cleanups
1156
1157 -------------------
1158 0.0.96 (2003-08-30)
1159 -------------------
1160
1161 - easy stream_defs syntax
1162 - sdl_gui can display images associated to the file being played
1163 - Major feature enhancements for icc_gui including dynamic text
1164 placement and the top/bottom window design
1165 - vrfy/clean now also checks for NULL values in attributes as
1166 well as for invalid picture pointers
1167 - fix long outstanding case sensitivity bug
1168 - many small fixes and cleanups
1169
1170 -------------------
1171 0.0.95 (2003-06-29)
1172 -------------------
1173
1174 - sdl gui runs much faster
1175 - new dbtool command: ca (copy attributes)
1176 - count and display number of times the song has been played
1177 - new feature: scoring
1178 - command line options for sdl_gui
1179 - simpler syntax of streams file
1180 - decrease network traffic of stat
1181 - fix zombie bug
1182 - many small fixes and cleanups
1183
1184 -------------------
1185 0.0.94 (2003-05-04)
1186 -------------------
1187
1188 - new server command: ns (next stream)
1189 - new icc_gui command: c (change stream)
1190 - internal mp3info
1191 - stat shows also id3 tag info
1192 - new sdl based gui
1193 - log flodding bug fixed
1194 - many small fixes and cleanups
1195
1196 -------------------
1197 0.0.93 (2003-03-28)
1198 -------------------
1199
1200 - colors for icc_gui
1201 - icc_gui sets volume directly (linux only)
1202 - proper locking that fixes some races
1203 - fix security bug that caused commands to be executed even
1204 with unsufficient permissions
1205 - new command: hup to make all servers reread their configuration file
1206 - icecast meta data streaming
1207 - many small fixes and cleanups