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