Merge branch 't/sched_improvements'
[paraslash.git] / NEWS
1 NEWS
2 ====
3
4 -------------------------------------------------
5 0.5.3 (to be released) "symbolic synchronization"
6 -------------------------------------------------
7
8 Not many new features, but lots of fixes and usability improvements.
9
10 - para_gui has been converted to use the paraslash scheduler.
11 - Various alsa-related fixes, mostly for the raspberry pi.
12 - Many scheduler improvements and cleanups.
13 - The test suite has been extended to include sanity checks
14 for the generated man pages.
15 - ao_writer fixes. This writer was in a quite bad shape. Many
16 serious bugs have been fixed.
17 - new audiod command: version.
18 - Minor improvements to the bitstream API.
19 - The cpsi command now prints a meaningful error message if
20 none of the given patterns matched any audio file.
21
22 ----------------------------------------
23 0.5.2 (2014-04-11) "orthogonal interior"
24 ----------------------------------------
25
26 The new sync filter, the AES_CTR128 stream cipher and the overhauled
27 network code are the highlights of this release. It also includes a
28 fair number of smaller fixes and improvements not mentioned here.
29
30 - The new sync filter synchronizes playback between multiple
31 clients.
32 - Connections between para_server and para_client are now
33 encrypted by means of AES rather than RC4 if both sides
34 support it. RC4 is still available as a fallback. This
35 feature is fully transparent, i.e. no command line options
36 are necessary, and a client linked against openssl can
37 speak with a server linked against libgcrypt and vice versa.
38 - Major cleanup of the networking subsystem.
39 - Improvements to para_fade: the new set mode, multi-channel
40 initial volumes, better error logging.
41 - The man pages of para_audiod, para_filter, para_recv, and
42 para_write contain the relevant options for receivers, filters,
43 writers. This broke in 0.5.0.
44 - ogg/vorbis latency improvements.
45 - Improved user manual.
46 - Minor fixes to avoid clang warnings.
47
48 Downloads: ./releases/paraslash-0.5.2.tar.bz2 (tarball),
49 ./releases/paraslash-0.5.2.tar.bz2.asc (signature)
50
51 ------------------------------------------
52 0.5.1 (2013-12-20) "temporary implication"
53 ------------------------------------------
54
55 Lots of fixes and improvements all over the place, and a major overhaul
56 of the build system.
57
58 - Audiod improvements and fixes.
59 - Buffer tree robustness improvements.
60 - Cleanup of the mood subsystem.
61 - Fixes and cleanups for the flac decoder.
62 - Latency improvements for the ogg/opus decoder.
63 - Crypto support is now optional. On systems without
64 openssl/gcrypt, the build succeeds but para_server,
65 para_audiod, para_client won't be built.
66 - The build system now works for cross-compile setups.
67 - The dependency tree has been flattened, which speeds up
68 builds and avoids to recreate the man pages on every change.
69 - The error code helper has been rewritten from perl to C,
70 which further improves build time.
71 - Many small bugs in the build system have been identified
72 and fixed.
73
74 Downloads: ./releases/paraslash-0.5.1.tar.bz2 (tarball),
75 ./releases/paraslash-0.5.1.tar.bz2.asc (signature)
76
77 ----------------------------------------
78 0.5.0 (2013-08-23) "invertible validity"
79 ----------------------------------------
80
81 Some API-breaking changes, one serious bug fix, and a lot of bike-shedding.
82
83 - The sideband compatibility code has been removed, hence
84 sideband connections (introduced in 0.4.11) are now mandatory.
85 - Addblob commands can produce output.
86 - The stat command no longer sends garbage when para_server was
87 compiled against libgcrypt.
88 - Dependencies for gengetopt files are computed automatically.
89 This eliminates a constant source of build bugs.
90 - The setatt command now accepts file name patterns rather than only
91 path names.
92 - overview.pdf is now based on dia, a simple diagram creation program.
93 The new version is much more detailed and contains descriptions of
94 the various programs of the paraslash package.
95 - The separator of all multi-word options has been changed from
96 underscore to dash. For example --log_color becomes --log-color.
97 - Overhauled web pages and the new logo.
98
99 Downloads: ./releases/paraslash-0.5.0.tar.bz2 (tarball),
100 ./releases/paraslash-0.5.0.tar.bz2.asc (signature)
101
102 --------------------------------------
103 0.4.13 (2013-07-29) "spectral gravity"
104 --------------------------------------
105
106 One more 0.4.x release before the API-breaking changes for 0.5.0 go
107 in. The main features of this release are the ogg/opus audio format,
108 and UTF-8 support, but it includes also tons of other improvements
109 and fixes all over the place.
110
111 - New audio format: ogg/opus.
112 - UTF8 support for para_gui and the mp3 audio format handler.
113 - Scheduler improvements and fixes.
114 - The obsolete gettimeofday() function has been replaced
115 by clock_gettime() on systems which support it.
116 - Speed and usability improvements for para_gui.
117 - para_client now restores the fd flags of stdin and stdout
118 on shutdown.
119 - Improved manual pages.
120 - Consistent version strings for all executables.
121 - Reduced dependencies on generated files result in fewer
122 recompilations on changes.
123 - Performance improvements for the compress filter.
124 - Improved downloads web page.
125
126 -----------------------------------------
127 0.4.12 (2012-12-20) "volatile relativity"
128 -----------------------------------------
129 The new command line player, the resample filter, ALSA support for
130 para_fade, and the improved build system are the highlights of this
131 release which probably marks the end of the 0.4.x series.
132
133 - The afh receiver and the para_play executable.
134 - The resample filter: A sample rate converter based on
135 libsamplerate.
136 - The "versions" directory has been removed from the master
137 branch. The tarballs of the old releases are now available
138 in the new "releases" branch.
139 - Overhaul of the build system: All generated files are now
140 written to the "build" directory.
141 - The modular mixer API and the alsa mixer.
142 - Minor fixes for the osx writer.
143
144 --------------------------------------
145 0.4.11 (2012-07-20) "mutual diversity"
146 --------------------------------------
147
148 The major feature in this release is the new sideband API for
149 client-server communication. This API will be used exclusively starting
150 with 0.5.0, which breaks backward compatibility but allows to get rid
151 of quite some compatibility code. Other noteworthy changes include
152 decoder latency improvements and a long-standing bug fix for the
153 ALSA writer.
154
155 - Sideband connections: If both para_server and para_client
156 support this feature, data is sent as a multiplexed stream.
157 - The --no_default_filters option of para_filter has been
158 removed.
159 - Several fixes and latency improvements to various decoders.
160 - The ALSA writer now limits the prebuffer time to 500ms.
161 - Documentation improvements.
162 - Overhaul of the command_util.sh script.
163 - Fixes for some minor problems found by the clang analyzer.
164 - Compiles (almost) without warnings on gcc-3.
165 - Robustness improvements of the buffer tree code.
166
167 ------------------------------------------
168 0.4.10 (2012-03-30) "heterogeneous vacuum"
169 ------------------------------------------
170
171 Nothing earth-shaking in this release, but quite a few usability
172 improvements and the usual mix of cleanups and fixes.
173
174 - The --no_default_filters option of para_filter has been
175 deprecated. It still works but has no effect and will be
176 removed in the next version.
177 - para_gui now prints also the stderr output of the executing
178 command in the bottom window.
179 - Cleanup and consolidation of the various wrappers for
180 write(), writev(), send() and friends.
181 - The obscure error messages on mmap() failures have been
182 replaced by meaningful messages. This affects mainly
183 para_afh.
184 - para_audioc: Cleanups and memory leak fixes.
185 - Test 0004-server no longer fails if para_server is not
186 being built.
187 - New configure options: --with-id3tag-{headers,libs}.
188
189 -------------------------------------
190 0.4.9 (2011-12-06) "hybrid causality"
191 -------------------------------------
192
193 Support for another audio format, interactive mode for para_client
194 and para_audiod and many small improvements/fixes all over the place.
195
196 - Support for flac, the free lossless audio codec.
197 - Fix for an endless loop in the mp3 decoder for certain
198 (corrupt) mp3 files.
199 - When executed without specifying a command, para_client
200 and para_audioc start an interactive shell (requires
201 libreadline being installed). The interactive mode offers
202 full tab completion and command line history.
203 - autogen.sh now detects a distcc setup and adjusts the
204 parameter for the -j option of make accordingly.
205 - Shared memory areas are no longer restricted to 64K. We now
206 detect the maximal size of a shared memory area at runtime.
207 - cleanup of the internal uptime API.
208 - para_server prefaults the mmapped audio file to avoid
209 delays on slow media.
210 - A new test for the test-suite that exercises the
211 communication between para_server and para_audiod.
212 - The alsa writer eats up less CPU cycles when configured to
213 use the DMIX plugin.
214 - Simplified and unified receiver code.
215 - Makefile cleanups.
216 - Commands which print a list of matching audio files now
217 emit a meaningful error message if no audio file matched the
218 given pattern(s).
219
220 --------------------------------------
221 0.4.8 (2011-08-19) "nested assignment"
222 --------------------------------------
223
224 Gcrypt support, the overhauled osx writer and regex format specifiers
225 are the highlights of this release.
226
227 - support for libgcrypt as a drop-in replacement for openssl.
228 Run configure --enable-cryptolib=gcrypt to link against
229 libgcrypt. The two crypto implementations are compatible to
230 each other, i.e. a para_client executable linked against
231 gcrypt can connect to para_server linked against libssl
232 and vice versa.
233 - Rewrite of the osx writer (output plugin for Mac OS).
234 - audiod: The format specifier for receivers, filters and
235 writers is now treated as a regular expression. This allows
236 to replace 5 lines in the config file (one for each audio
237 format) by one single line. See the manual for details.
238 - The *.cmdline.[ch] files are no longer contained in the released
239 tarballs. This reduces the size of the tarballs but requires
240 gengetopt to build the tarball.
241 - Compiles cleanly also with llvm/clang.
242 - Corrupt mp3 files are handled more gracefully.
243 - The alsa writer uses poll fds instead of computing timeouts.
244 - Cleanup of the generic writer API.
245 - sched: Optimized zero timeouts.
246 - vss timeout cleanups.
247 - oggdec fixes and improvements.
248
249 --------------------------------------
250 0.4.7 (2011-06-01) "infinite rollback"
251 --------------------------------------
252
253 The new ao writer, support for ssh RSA keys and a couple of other
254 enhancements.
255
256 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
257 platforms through the libao audio library.
258 - Support for RSA keys generated with ssh-keygen.
259 - configure: improved options for ogg/vorbis/speex.
260 - The git version reported by --version always matches HEAD.
261 - The autogen script detects the number of processors and
262 runs a parallel make if possible.
263 - Major cleanup of the crypto API.
264 - Documentation updates.
265
266 ------------------------------------------
267 0.4.6 (2011-03-31) "deterministic entropy"
268 ------------------------------------------
269
270 Lots of ogg/vorbis improvements, the new test suite, enhancements
271 for para_gui and a fair amount of other bug fixes.
272
273 - For DCCP/OGG streams the audio file header is only sent once
274 at the beginning of the stream rather than periodically
275 every five seconds. This reduces network traffic and the
276 FEC group size.
277 - The vorbis comment header is replaced by an empty dummy header
278 before the header is sent over the network. This also results in
279 less network traffic and smaller FEC groups.
280 - The new "test" make target allows to perform some sanity checks prior
281 to installing the package.
282 - ogg timing fixes and performance improvements
283 - Scheduler improvements
284 - Proper exit codes for para_write
285 - para_gui: New option --theme to select a startup theme. Several
286 other improvements and fixes.
287 - aacdec error message cleanups
288 - simplified color error handling
289
290 --------------------------------------------
291 0.4.5 (2010-12-17) "symmetric randomization"
292 --------------------------------------------
293
294 Bug fixes, internal cleanups and variable-sized FEC slices.
295
296 - Contains a fix for an invalid-free-bug in the ogg audio format
297 handler code.
298 - Switching off the DCCP sender works again.
299 - para_audiod handles crashes of para_server more robustly.
300 - Internal scheduler and writer cleanups.
301 - Reduced latency due to variable-sized FEC slices.
302 - Improved documentation and error diagnostics.
303 - The build of para_server is now optional, allowing the build
304 to succeed in case libosl is not installed.
305
306 ------------------------------------------
307 0.4.4 (2010-08-06) "persistent regularity"
308 ------------------------------------------
309
310 Support for yet another audio format, para_write improvements and
311 bug fixes.
312
313 - Support for the speex codec.
314 - Support for sample formats other than 16 bit little endian.
315 - error2.h is now created by a perl script which speeds up configure
316 considerably.
317 - Fix a bug in the aac decoder which could lead to segfaults in
318 para_filter/para_audiod.
319 - Fixes for autoconf-2.66.
320
321 ----------------------------------------
322 0.4.3 (2010-07-05) "imaginary radiation"
323 ----------------------------------------
324
325 Many improvements for the DCCP and the UDP transport, the new user
326 manual and the usual mix of bug fixes and internal improvements.
327
328 - FEC support for the DCCP sender (Gerrit Renker). The new
329 --dccp_max_slice_size, --dccp_data_slices_per_group and
330 --dccp_slices_per_group options can be used to set the FEC
331 parameters for the DCCP transport.
332 - DNS lookups for UDP targets (Gerrit Renker).
333 - The new user manual replaces the README, README.afs, REQUIREMENTS
334 and INSTALL documents.
335 - Fix an end-of-file detection bug in the oggdec filter.
336 - The new nonblock API.
337 - Both options of the oggdec filter have been removed.
338 - New debug mode for the internal scheduler.
339
340 ------------------------------------------
341 0.4.2 (2010-04-23) "associative expansion"
342 ------------------------------------------
343
344 It's been some time since the last release, but finally here is
345 paraslash-0.4.2. The bulk of the changes comes from the new buffer
346 tree API, but there are changes all over the tree. Mainly performance
347 and usability improvements, but also quite some bug fixes.
348
349 - The new buffer tree API.
350 - DCCP: Support for CCID negotiation (Gerrit Renker).
351 - UDP robustness fixes.
352 - The --bufsize option for mp3dec is gone as it no longer makes sense
353 for the new buffer tree API.
354 - Fix audible buffer underruns for wma streams.
355 - The alsa writer no longer prints meaningless underrun durations.
356 - audiod: Defaults work also for udp streams. If no filter is
357 given for an audio format that is received via upd, fecdec is
358 automatically added as the first filter (along with the decoder).
359
360 ---------------------------------------
361 0.4.1 (2009-12-22) "concurrent horizon"
362 ---------------------------------------
363
364 Support for another audio format, minor feature enhancements and lots of bug
365 fixes. All fixes that have been accumulated in the maint branch (in particular
366 those mentionened in the 0.3.6 release notes) appear in this release as well.
367
368 - wma support.
369 - new afh option: --human to activate human-readable output.
370 - new server/audiod option: --log-timing to print timing information.
371 - build system improvements.
372 - source code documentation updates.
373
374 -------------------------------------
375 0.3.6 (2009-12-07) "cubic continuity"
376 -------------------------------------
377
378 Quite a few bugs have been found and fixed since 0.3.5, so here's
379 another 0.3.x release. No new features.
380
381 - Always check return value of malloc().
382 - ogg vorbis/FEC: Do not write garbage after the audio file header.
383 - exit if root privileges could not be dropped.
384 - FEC: Fix computation of extra slices.
385 - oss: Fix check for empty input buffer.
386 - Avoid buffer underruns due to filter chain output buffer constraints.
387 - server: Fix assignment of afs_pid.
388 - Don't panic if the afs database contains unknown audio formats.
389 - http/dccp: Do not send the audio file header twice.
390 - FEC: Timing improvements.
391
392 ----------------------------------------------
393 0.4.0 (2009-11-10) "simultaneous independence"
394 ----------------------------------------------
395
396 Two significant changes which require the new version number: The
397 improved authentication dialog and the fact that the database code
398 has been moved to a library, libosl. To use the new version, you have
399 to generate new RSA keys, see INSTALL for details. A shell script is
400 provided for conversion of the 0.3 database to the new 0.4 format.
401
402 - stronger crypto for client authentication
403 - the database code has been moved to a library
404 - improved status item handling
405 - cleanup of the build system
406 - The "-V" option now also prints the git version
407 - the new parser-friendly listing mode for the ls and stat commands
408 - mandatory rc4 encryption
409 - major audio format handler cleanups
410 - (id3,...) tags are no longer stored as a combined string in the database
411 - new mood methods: artist_matches, title_matches, comment_matches,
412 album_matches, year_maches, year.
413
414 --------------------------------------------
415 0.3.5 (2009-09-21) "symplectic separability"
416 --------------------------------------------
417
418 Full client support for *BSD Unixes, complete re-write of the ogg
419 vorbis audio format handler, various improvements all over the place
420 and the usual mix of bugfixes. This release marks the end of the 0.3
421 series if no serious problems show up.
422
423 - the new oss writer (supported on *BSD and Linux)
424 - rewrite of the ogg vorbis audio format handler. It's
425 recommended to replace the chunk tables of existing ogg
426 vorbis files in the afs database by re-adding these files
427 with "add -f".
428 - support for netmask subsets (Gerrit Renker)
429 - the new prebuffer filter
430 - improved signal handling
431 - variable fec output buffer size
432 - improved FEC timing fixes audible buffer underruns in UDP mode
433 - --log_color actually works
434 - new ls option: -d (print dates as seconds after the epoch)
435 - update to gengetopt 2.22.2
436 - support for RSA keys of size > 512 bits
437 - new option "mixer_channel" for para_fade
438
439 -----------------------------------------
440 0.3.4 (2009-05-07) "elliptic inheritance"
441 -----------------------------------------
442
443 The new udp sender, forward error correction, colored logs and various
444 other improvements. As the udp sender does not depend on any special
445 libraries, it is built unconditionally.
446
447 - The udp sender replaces the ortp sender. The new code uses forward
448 error correction to protect against packet losses. Many thanks to
449 Gerrit Renker for providing ipv6 support.
450 - The default port for udp streaming now defaults to 8000, like
451 for the http and the dccp senders/receivers.
452 - Loglevels are now specified as symbolic names, e.g.
453 "--loglevel info".
454 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
455 - para_server/para_audiod: Color support for log messages.
456 - new options for mp3dec: --ignore-crc, --bufsize
457 - new audiod option: --config-file.
458 - gengetopt cleanups.
459 - Improved help/man pages: The documentation of para_audiod,
460 para_recv, para_filter and para_write now also contains
461 all options of the available receivers/filters/writers. The
462 man page of para_fade contains a description of the different
463 modes of operation.
464 - More source code documentation.
465 - vss timing fixes.
466
467 --------------------------------------------
468 0.3.3 (2008-12-01) "axiomatic perspectivity"
469 --------------------------------------------
470
471 Internal code cleanups, bug fixes, improved tag handling and the new
472 amplification filter.
473
474 - para_server uses the generic scheduling code.
475 - overhaul of the virtual streaming system.
476 - mp3: id3 version 2 support via libid3tag (optional)
477 - ogg: vorbis comment support.
478 - aac meta info support.
479 - mp3 audio format handler cleanups.
480 - new filter: "amp" to amplify the amplitude of the audio stream
481 - new status item/database entry: amplification. It is
482 used by the amp filter to pre-amplify the audio stream.
483 - fix a close-without-open bug in para_write.
484 - fix a bug in com_init() which was introduced in 0.3.2.
485 - better error diagnostics for para_client.
486
487 -----------------------------------------
488 0.3.2 (2008-04-11) "probabilistic parity"
489 -----------------------------------------
490
491 The new para_afh executable, scheduling and documentation improvements.
492
493 - new ls option: -lc (list chunk table)
494 - new executable: para_afh, the stand-alone audio file handler tool
495 - afs commands can send output more than SHMMAX (32MB on Linux). This
496 also reduces the memory usage of commands that produce large amounts
497 of output.
498 - major scheduler and audiod cleanups.
499 - more detailed and much nicer man pages.
500
501 ---------------------------------------
502 0.3.1 (2008-02-23) "liquid interaction"
503 ---------------------------------------
504
505 A mix of cleanups, bug fixes, improvements, and some new features. No
506 significant changes to the new database (osl) code, which is generally
507 a good sign.
508
509 - Share some similar/duplicate code between the http and the
510 dccp sender.
511 - Generic access control lists for paraslash senders.
512 - dccp sender: Access control lists, connection limiting and support
513 for the allow,deny,on,off,help sender commands.
514 - The default dccp port changed from 5001 to 8000 (suggested by
515 Gerrit Renker).
516 - para_server starts even if not all public keys could be loaded.
517 - Audiod performance improvements.
518 - fix a bug in the "off" command of the http sender.
519 - fix some fd and memory leaks.
520 - Update to gengetopt-2.22.
521
522 -------------------------------------
523 0.3.0 (2008-01-12) "solar saturation"
524 -------------------------------------
525
526 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
527
528
529 Usually one might expect lots of new features AND a big increase in size
530 for a major release like this.
531
532 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
533 ever. The decrease in size is mostly due to the removal of some
534 graphical tools (which were only quick hacks anyway). But also the
535 fact that the mysql code is gone cuts down the size a bit.
536
537 Being independent of mysql comes at a cost: The fact that paraslash
538 now contains its own database (the object storage layer, osl) increases
539 the (stripped) binary size of para_server by ~50K on i386.
540
541 - no more restrictions on unique basenames.
542 - independent of mysql: The new self-contained object
543 storage layer (osl) replaces the mysql database.
544 - New executable para_fsck: Check integrity of osl tables.
545 - Lyrics support.
546 - Reliable audio file move/rename detection.
547 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
548 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
549 Solaris (sparc64).
550 - the new osl-based audio file selector (afs) replaces the random,
551 the playlist and the mysql selector of paraslash-0.2.x.
552 - IPv6 support (thanks to Gerrit Renker).
553 - paraslash-0.2.x streams are now called "moods". Writing
554 0.3.x-mood definitions should be both easier and more
555 powerful than writing 0.2.x-stream definitions.
556 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
557 been removed. The world is a better place without them. However,
558 para_gui is still there.
559 - afs tracks audio file selection also in playlist mode.
560 - few easy-to-use afs commands replace the many not-so-easy-to-use
561 mysql commands (and are available also in playlist mode).
562 - Improved error subsystem.
563 - The earth-shaking new logo.
564
565 -----------------------------------------
566 0.2.17 (2007-11-20) "isotropic threshold"
567 -----------------------------------------
568
569 Mainly bugfixes and cleanups in this version which marks the end of
570 the 0.2.x series if no serious bugs show up after the release.
571
572 - mysql_selector: fix a locking bug.
573 - universal chunk queueing.
574 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
575 to Gerrit Renker).
576 - be more carful wrt. signed vs. unsigned argument passing.
577 - cleanup error.h and fix some references to invalid error
578 codes.
579 - update to gengetopt-2.21.
580 - update to ortp-0.13.1.
581 - autoconf: extend checks for headers, library functions and
582 compiler characteristics.
583 - Fix streaming of large mp3 files.
584 - Fix an off-by-one bug in playlist handling.
585
586 --------------------------------------
587 0.2.16 (2007-04-05) "neural discharge"
588 --------------------------------------
589
590 The main change in this release is the major audio format handler
591 cleanup which removes some similar/duplicate code and makes it easier
592 to implement plugins for other audio formats. Of course, the usual mix
593 of other improvements/changes/bugfixes also made it into the release.
594
595 - simplified audio format handlers (most of the handling functions
596 were moved one layer up to the virtual streaming system).
597 - para_server uses mmap to read audio files
598 - repositioning of mp3 streams is much faster, in particular for
599 jumping near the end of large mp3 files.
600 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
601 and AFS_WRITE.
602 - fix a bug in para_filter that caused decoding of aac files
603 to start only after a few seconds.
604 - fix osx_writer hangs
605 - simplified dccp code (thanks to Gerrit Renker)
606 - the compress filter works also on big endian systems (ppc)
607
608 -----------------------------------------
609 0.2.15 (2007-02-16) "inductive resonance"
610 -----------------------------------------
611
612 Minor improvements, more documentation and a bunch of bug fixes.
613
614 - para_server: The server.users file is only read once on server
615 startup rather than for each connection
616 - mp3dec: Fix decoding of corrupt mp3 files
617 - afs (audio file sender) is now called vss (virtual streaming
618 system). Consequently, the permission flags specified in
619 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
620 become VSS_READ and VSS_WRITE respectively.
621 - para_audiod/para_filter: Fix a bug that caused the last chunk
622 of audio data not being written under certain circumstances
623 - audiod: compute the difference of server time and local time
624 correctly
625 - para_server/para_audiod: Fix some memory leaks
626 - documentation improvements
627 - configure.ac: fix checks for para_krell
628 - new man pages
629
630 -------------------------------------------
631 0.2.14 (2006-10-15) "transient singularity"
632 -------------------------------------------
633
634 The only major enhancement of this version is the osx writer which completes
635 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
636 until now for reasons beyond the scope of this changelog entry.
637
638 - new output plugin for Mac Os: the osx writer
639 - rename configure command line options from --enable-xxx-headers to
640 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
641 - configure: new command line options: --with-mad-headers,
642 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
643 - some robustness fixes
644 - dymamic audio format recognition for audiod
645 - para_server: new command line option: --autoplay_delay
646 - para_audiod: new command line option: --clock_diff_count
647
648 ---------------------------------------
649 0.2.13 (2006-07-14) "sonic convolution"
650 ---------------------------------------
651
652 A bunch of new features and core changes.
653
654 - the new paraslash scheduler, short and sweet.
655 - Support for m4a/mp4 files via the new aac audio format
656 handler/filter (requires libfaad).
657 - each writer has its own command line parser, just like
658 para_recv and para_filter.
659 - para_client and para_audioc use the error subsystem
660 - writers are integrated in para_audiod (currently linux-only)
661 - para_client is integrated in para_audiod
662 - random/playlist selector: improved info strings
663 - new audiod commands: tasks, kill
664 - update to libortp-0.10.1
665 - para_fade: wake time defaults to 8 hours from now
666 - update to autoconf-2.60
667
668 ------------------------------------------
669 0.2.12 (2006-05-12) "oriented abstraction"
670 ------------------------------------------
671
672 Many user-visible changes in this release and lots of new
673 features:
674
675
676 - the new optional dccp sender/receiver. It uses the datagram
677 congestion control protocol. You'll need a fairly new kernel
678 for this.
679 - paraslash works on Mac OS X (thanks to Gerd Becker)
680 - para_play renamed to para_write
681 - modular output plugin design (writers) for para_write
682 - new file_writer output plugin for para_write
683 - compress filter speed improvements
684 - update to libortp-0.9.1
685 - update to gengetopt-2.17rc
686 - para_client no longer depends on libreadline (as the
687 code for the interactive mode was removed).
688 - gcc-2-95 is no longer a supported compiler. It may still
689 work, but it gets no more testing.
690 - the tarball no longer contains the screenshot images which
691 reduces its size quite a bit.
692 - configure: new command line options: --enable-mysql-headers
693 and --enable-mysql-libs
694
695 ------------------------------------
696 0.2.11 (2006-03-11) "atomic duality"
697 ------------------------------------
698
699 Here it is, the first paraslash release developed with git. There
700 are fairly many user-visible changes in this release. As two out of
701 the three "database tools" of paraslash don't use a database at all,
702 they are now called "audio file selectors" instead.
703
704
705 - the cdt command (change database tool) becomes chs (change
706 selector)
707 - no more colon separators: The syntax of some options of
708 para_audiod and para_filter have changed. Use --help for
709 more info (and some examples).
710 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
711 - switch from cvs to git (should've done that earlier)
712 - the new ipc subsystem
713 - new audio file selector: playlist
714 - para_server: the dopey selector is now called "random",
715 and is the default selector. Use the --selector option to
716 choose another selector at startup, or the chs command to
717 change the selector at runtime.
718 - X86_64 fixes (thanks to Steffen Klassert)
719 - para_play fixes
720
721 --------------------------------------
722 0.2.10 (2006-02-17) "cyclic attractor"
723 --------------------------------------
724
725 Huge documentation update, a scrollable window for para_gui, ortp
726 improvements, and of course many small fixes not mentioned here.
727 The diffstat below is rather misleading as most insertions are due
728 to the new source documentation.
729
730 - autoconf cleanup
731 - para_server also uses the new error subsystem
732 - lots of new documentation (UTSL)
733 - gui improvements:
734 - keysyms for cursor keys and for next/previous page keys
735 - scrollable output window
736 - new internal commands: scroll up/down, page up/down
737 - fix color of command output.
738 - ortp: the --chunk_time and --header flags are no longer needed
739 for para_recv/para_audiod as this information is now encoded in
740 each rtp packet sent by para_server.
741
742 -------------------------------------------
743 0.2.9 (2006-01-24) "progressive turbulence"
744 -------------------------------------------
745
746 Internal audiod receivers/filters, the new error subsystem and
747 a lot of small improvements.
748
749 - para_recv and para_filter are integrated into the para_audiod
750 binary, i.e. audiod no longer spawns a new process for
751 each receiver/filter. As para_recv and para_filter might be
752 useful as standalone programs, they still get built (linked
753 against the same object files that are also used for audiod).
754 - further ortp timing improvements should reduce the CPU usage
755 of the ortp receiver.
756 - improved audio grabbing. The 'grab' command of para_audiod
757 has its own set of command line options. Read the output of
758 "para_audioc help grab" for more info.
759 - oggdec: configurable input prebuffer size.
760 - the new error subsystem gives better error diagnostics
761 and reduces code size.
762
763 -----------------------------------------
764 0.2.8 (2006-01-02) "dynamic accumulation"
765 -----------------------------------------
766
767 The new modular filter design and the para_play-hangs bugfix.
768
769 - new executable: para_filter. It combines para_mp3dec,
770 para_oggdec and para_compress. It also adds a further filter
771 type, wav, that just inserts a wave header at the desired point
772 of the filter chain. All 'piping' is done in-memory (i.e. no
773 read/write operations are used).
774 - para_play: fix a stupid bug that caused it to hang under
775 certain circumstances.
776
777 -------------------------------------------
778 0.2.7 (2006-12-27) "transparent invariance"
779 -------------------------------------------
780
781 Not many user-visible changes but a fair amount of internal improvements.
782
783
784 - The http sender buffers data if it can not be sent
785 out immediately (because the socket is not writable). This
786 should prevent para_server from shutting down the connection
787 too early on a loaded network.
788 - para_play also prebuffers data if it is told to start at a
789 future time by the --start_time option.
790 - The return of para_recv: It combines para_ortp_recv and
791 para_http_recv. Use the --receiver option to switch between
792 the two. para_recv builds without libortp, but contains
793 only the http receiver in this case.
794 - update to ortp 0.8.1. As this ortp release contains incompatible
795 changes, para_recv-0.2.7 won't link against older ortp libs.
796 - improved ortp timings.
797 - use of gcc-extensions that #define away for non-gcc and
798 gcc < 3.0.
799
800 -------------------------------------------
801 0.2.6 (2005-10-29) "recursive compensation"
802 -------------------------------------------
803
804 Transparent session encryption (uses openssl's Alleged RC4 cipher),
805 the internal find command and several other improvements and cleanups.
806
807 - Encrypt the session if encryption is requested by the client
808 (default for para_client 0.2.6). This is backwards
809 compatible, so older clients can still connect to para_server
810 0.2.6. Use the new client option --plain to request an
811 uncrypted session (off by default, must be set to on in
812 order to connect to para_server 0.2.x with 0 <= x <= 5).
813 - para_server uses an internal function to locate audio files
814 rather than calling find(1). The server option
815 --mysql_audio_file_dir replaces --mysql_find_cmd.
816 - documentation update
817 - man pages
818 - header file cleanup
819 - para_client code cleanup
820 - para_gui: faster display of output of display commands
821
822 ------------------------------------------
823 0.2.5 (2005-10-13) "aggressive resolution"
824 ------------------------------------------
825
826 This release adds internal senders, i.e. no more external programs are
827 spawned for sending out the audio data. There are two different senders
828 available: The http sender and the ortp sender (former para_send which
829 is no longer needed).
830
831 The new sender code has a plugin-like design so it can be easily
832 extended should there be be any future need for supporting another
833 network streaming protocol. All senders are completely independent of
834 each other. In particular, the http and the ortp sender can operate
835 in parallel.
836
837 - new server command: sender to control senders at runtime.
838 Read the output of "para_server -h" and "para_client help
839 sender" for more information.
840 - para_recv renamed to para_ortp_recv
841 - new executable: para_http_recv, a simple command line
842 http receiver.
843 - major afs/mp3/ogg code simplifications due to internal
844 senders.
845 - ogg timing improvements
846 - fix several minor memory leaks (found by valgrind)
847 - empty stream definitions work again
848 - com_ne(): ignore errors on remove
849 - audiod: fix segfault on server restart
850
851 ---------------------------------------
852 0.2.4 (2005-09-21) "toxic anticipation"
853 ---------------------------------------
854
855 Several small improvements, fixes and the new grab command.
856
857 - audiod:
858 - new command: "grab" to grab the output of the stream reader
859 or any filters. Read the output of "para_audioc help grab"
860 for more information.
861 - fix memory leak
862 - code cleanup
863 - audioc: new command line option: --bufsize to specify a
864 buffer size different from the default size 8192.
865 - improved error diagnostics for para_play.
866 - new configure option: --enable-ssldir so search for openssl in
867 non-standard places
868 - sdl_gui: Make it look nice again for 1024x768
869 - server: report total size of memory allocated with sbrk by malloc,
870 new command line option: --announce_time
871
872 -----------------------------------------
873 0.2.3 (2005-09-01) "hydrophilic movement"
874 -----------------------------------------
875
876 Two new executables and major feature enhancements.
877
878 - audiod filters: It is now possible to specify arbitrary many
879 (including none) filters for each supported audio
880 format. This can be used e.g. for normalizing volume,
881 transforming or grabbing the audio stream, or for using
882 visualizers. Read the output of "para_audiod -h" for the
883 syntax of the new --filter_cmd option.
884 - new executable: para_play, a tiny alsa player. It
885 can play wave files or raw pcm (16 bit little endian)
886 from stdin.
887 - new executable: para_compress, a dynamic range compressor
888 intended to keep audio output at a consistent volume. Derived
889 from AudioCompress, http://trikuare.cx/code/AudioCompress.html.
890 - audiod: New option: --stream_delay. This can be used in
891 a local network to syncronize the audio output of all
892 clients that play the same stream.
893
894 ------------------------------------------
895 0.2.2 (2005-08-19) "tangential excitation"
896 ------------------------------------------
897
898 Mostly internal changes in this release, but also some new commands
899 for the mysql database tool.
900
901 - cleanup exec.c, fix para_exec bug
902 - compile time loglevel (log messages below the given level
903 won't be compiled in, which reduces the size of the
904 resulting binaries)
905 - new log macros that shorten the size of the source code.
906 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
907 to send only zeros. With this workaround, para_audioc works
908 again.
909 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
910 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
911 (new entry), "snp" (set numplayed). Read the manual for more
912 information.
913
914 ---------------------------------------
915 0.2.1 (2005-08-15) "surreal experience"
916 ---------------------------------------
917
918 Here comes paraslash-0.2.1. It contains a couple of new features and,
919 surprise, only minor bug fixes.
920
921 - kill noisy mp3 debug message
922 - cleanup of the build system
923 - para_server and para_client directly use the crypto routines
924 of the openssl library rather than invoking the openssl command
925 line utitlity
926 - server/audiod: new option --user to switch to the given user
927 when invoked as root. Read the output of "para_server -h" for
928 more information.
929 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
930 status. Default: "para_audioc stat"
931 - sdl_gui: new option --pic_cmd to be used to download the picture.
932 Default: "para_client pic"
933 - audiod: 5 slots ought to be enough for everybody
934 - audiod: new status item: Uptime, kill hup command
935
936 ------------------------------------------
937 0.2.0 (2005-08-06) "distributed diffusion"
938 ------------------------------------------
939
940 After several month of increased development activity, paraslash-0.2.0
941 has arrived. It contains many new features and is much more
942 self-contained than the old 0.1.x series. Enjoy!
943
944
945 - para_server: fix hang on song change and crash on sighup.
946 Speed up mysql queries. The DIR_LIKE macro is gone.
947 - new executables: para_audiod, the local audio daemon that
948 starts playback (uses SCM_CREDENTIALS socket magic) and
949 para_audioc, the corresponding client.
950 - new executables: para_mp3dec/para_oggdec, two really teensy
951 decoders. para_mp3dec is based on libmad, para_oggdec requires
952 libvorbisfile.
953 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
954 they are now called para_send and para_recv respectively.
955 - documentation updates
956 - para_gui is themable. For now there is the default theme that
957 looks as before and the simple theme: blue and easy.
958 - gui: audio streaming is now handled by audiod. Time display shows
959 playback time rather than streaming time
960 - slider: update to libzmw-0.2.0
961 - para_krell: fix crash on server shutdown
962 - switch from gzip to bzip2
963
964 ----------------------------------------
965 0.1.7 (2005-04-18) "melting penetration"
966 ----------------------------------------
967
968 The main change in this release is clearly the oggvorbis rewrite,
969 but there are also lots of smaller changes. If you intend to use both
970 the mp3 and the ogg plugin, it is recommended to use software mixing,
971 e.g. the dmix plugin which is provided by ALSA.
972
973 - new executables: para_ovsend and para_ovrecv for sending/receiving
974 oggvorbis files via rtp. Requires the open rtp library. Get it at
975 http://www.linphone.org/ortp/
976 - rewrite of the ogg_vorbis core code
977 - configure detects libzmw and, if detected, includes
978 para_slider to the list of binaries to be built by make
979 - server stream writers read from their associated fifo rather
980 than from stdin
981 - slider: two new sliders, lastplayed and numplayed
982 - fix nasty double free bug which caused random segfaults in case of
983 mp3 files with invalid header information
984 - gui: new command line option: --stream_timeout=seconds to
985 deactivate a slot if it is idle for that many seconds (default=`5')
986 - diffstats
987
988 ---------------------------------------
989 0.1.6 (2005-03-05) "asymptotic balance"
990 ---------------------------------------
991
992 Only little user-visible changes in this release. Mainly bugfixes and
993 core code cleanup. This is probably the most stable version ever if you
994 stick to mp3...
995
996 - fix several memory leaks
997 - rename default name of mysql database from "music" to "paraslash".
998 Use para_server's --mysql_database option if you do not want to
999 rename your old database.
1000 - rework ogg vorbis code
1001 - make update command work on mysql servers with LOCAL_INFILE
1002 disabled
1003 - gui: improved stream I/O (slots)
1004 - simplified audio format API
1005 - para_pob_ogg is gone
1006
1007 ------------------------------------
1008 0.1.5 (2004-12-31) "opaque eternity"
1009 ------------------------------------
1010
1011 Let's slide gently into the new year.
1012
1013 - new: para_slider (not built automatically, type "make
1014 para_slider" to build). A toy for those who always felt that
1015 creating stream definitions is difficult. See screenshots,
1016 README and FEATURES for more info.
1017 - improved signal handling. Fixes server segfault on SIGHUP
1018 for linux kernels newer than Aug 24 2004 and makes para_gui
1019 race-free.
1020 - reload database tool on SIGHUP
1021 - improved help message for sl
1022 - do not log "broken pipe" messages as errors. They are
1023 perfectly ok.
1024 - fix wrong error message on permission errors
1025
1026 -----------------------------------------
1027 0.1.4 (2004-12-19) "tunneling transition"
1028 -----------------------------------------
1029
1030 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
1031 Hopefully, most of them got wiped out with this release. Some
1032 enhancements went also in.
1033
1034 - improved error diagnostics for all commands
1035 - stradd/picadd: overwrite previous contents if entry already
1036 exists, rather than returning errors
1037 - stradd: use current stream if invoked without args
1038 - faster (and hopefully more stable) ogg-vorbis handling
1039 - para_krell: reap children to avoid zombie-flooding in case
1040 no server is running
1041 - si: report also server pid
1042 - server: don't busy-loop if dbtool reports only invalid files.
1043 - gui: CTRL+C works again, fix stream_read command line option
1044 - fix pic_add, hist
1045 - fix mysql dbtool startup in case no database exists
1046 - many small fixes and cleanups
1047
1048 ---------------------------------------
1049 0.1.3: (2004-12-10) "vanishing inertia"
1050 ---------------------------------------
1051
1052 Starting from this release, the database tools are integrated in the
1053 server binary. This decreases server startup time, reduces code size
1054 and speeds up database commands. However, the layout of the underlying
1055 mysql database changed only slightly and 0.1.3 should be backwards
1056 compatible in that respect.
1057
1058 Visible changes:
1059
1060 - If mysql is not detected at compile time, or fails to init
1061 at runtime, fall back to the dopey database tool which should
1062 always work.
1063 - para_dbtool and dbtool.conf are gone. All mysql specific
1064 options are read from server.conf and are prefixed by 'mysql_'.
1065 - new command: cdt (change database tool)
1066 - new command line option: dbtool (choose startup database tool)
1067 - The name of current stream is now stored in the database,
1068 so paraslash remembers its current stream when restarted.
1069 - new command: csp (change stream and play)
1070 - para_gui also reports current database tool and server uptime
1071
1072 -------------------------------------------
1073 0.1.2: (2004-11-28) "spherical fluctuation"
1074 -------------------------------------------
1075
1076 Point release before the big dbtool changes go in.
1077
1078 - dbtool: rename ca to cam (copy all meta data). It now also
1079 copies numplayed and lastplayed time as well as the picture
1080 id.
1081 - fix endless-loop-bug caused by mp3 files with invalid header
1082
1083 -----------------------------------------
1084 0.1.1: (2004-11-05) "floating atmosphere"
1085 -----------------------------------------
1086
1087 - gkrellm plugin
1088 - new dbtool command: mbox. Browse your sound-file collection
1089 with your favorite mail reader.
1090 - several small fixes
1091
1092 -------------------------------------
1093 0.1.0: (2004-10-22) "rotating cortex"
1094 -------------------------------------
1095
1096 - fix logging bug for loglevel > VERBOSE
1097 - fix skip command
1098 - correct timings for vbr mp3s
1099 - modular audio format support
1100 - ogg-vorbis support (experimental)
1101 - new server option: autoplay
1102
1103 -----------------------------------------
1104 0.0.99: (2004-07-25) "harmonic deviation"
1105 -----------------------------------------
1106
1107 - rename projectname from icc to paraslash (play, archive, rate
1108 and stream large audio sets happily)
1109 - paraslash is no longer restricted to one particular audio
1110 streaming software
1111 - new dbtool commands (stradd, strq, strdel) for easy stream
1112 managment w/o configuration file. That obsoletes stream_defs
1113 file/config option for dbtool.
1114 - picadd accepts jpeg data from stdin
1115 - new server commands: ps (select previous stream), sc (song change)
1116 - new default pictures for sdl_gui
1117 - gui: new key_map option for binding commands and internal
1118 functions to arbitrary keys, nice help screen, rip out
1119 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1120 show silly logo on startup
1121 - new executables: para_fade for fading volume, para_dbadm for
1122 manipulating attributes
1123 - cdb adds _all_ tables to mysql database
1124 - revised and beautified documentation
1125 - sample dbtool rewritten in C
1126 - autoconf
1127
1128 ---------------------------------------------
1129 0.0.98: (2003-12-26) "incremental smoothness"
1130 ---------------------------------------------
1131
1132 - kick icecast in favour of poc. That removes some races and reduces
1133 core code considerably.
1134 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1135 give finer info on afs' status.
1136 - gui can start decoder (see config options). Further new gui
1137 commands: refresh (^L), jmp (F1-F10)
1138 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1139 terminates gui. This fixes dead instances consuming memory
1140 continuously.
1141 - new dbtool command: verb for sending verbatim sql queries.
1142 - fix pid_list races (by removing pid_list)
1143 - codename funnies
1144
1145 --------------------
1146 0.0.97: (2003-10-26)
1147 --------------------
1148
1149 - installation prefix now defaults to /usr/local
1150 - new commands for gui: snozze, sleep and reread config
1151 - config file for gui and sdl_gui
1152 - fix problems with filenames containing funny characters
1153 (reported by Thomas Forell)
1154 - improved signal handling for gui, now it rereads conf on SIGHUP
1155 - new dbtool command: cdb (create database)
1156 - switch from argtable to gengetopt
1157 - major code cleanup and speed improvements
1158 - fix several potential buffer overflows
1159 - many small fixes and cleanups
1160
1161 -------------------
1162 0.0.96 (2003-08-30)
1163 -------------------
1164
1165 - easy stream_defs syntax
1166 - sdl_gui can display images associated to the file being played
1167 - Major feature enhancements for icc_gui including dynamic text
1168 placement and the top/bottom window design
1169 - vrfy/clean now also checks for NULL values in attributes as
1170 well as for invalid picture pointers
1171 - fix long outstanding case sensitivity bug
1172 - many small fixes and cleanups
1173
1174 -------------------
1175 0.0.95 (2003-06-29)
1176 -------------------
1177
1178 - sdl gui runs much faster
1179 - new dbtool command: ca (copy attributes)
1180 - count and display number of times the song has been played
1181 - new feature: scoring
1182 - command line options for sdl_gui
1183 - simpler syntax of streams file
1184 - decrease network traffic of stat
1185 - fix zombie bug
1186 - many small fixes and cleanups
1187
1188 -------------------
1189 0.0.94 (2003-05-04)
1190 -------------------
1191
1192 - new server command: ns (next stream)
1193 - new icc_gui command: c (change stream)
1194 - internal mp3info
1195 - stat shows also id3 tag info
1196 - new sdl based gui
1197 - log flodding bug fixed
1198 - many small fixes and cleanups
1199
1200 -------------------
1201 0.0.93 (2003-03-28)
1202 -------------------
1203
1204 - colors for icc_gui
1205 - icc_gui sets volume directly (linux only)
1206 - proper locking that fixes some races
1207 - fix security bug that caused commands to be executed even
1208 with unsufficient permissions
1209 - new command: hup to make all servers reread their configuration file
1210 - icecast meta data streaming
1211 - many small fixes and cleanups