com_addatt(): Return negative on errors
[paraslash.git] / NEWS
1 NEWS
2 ====
3
4 -----------------------------------------
5 current master branch "magnetic momentum"
6 -----------------------------------------
7 - On Linux systems, local sockets are now created in the
8 abstract name space by default. This allows to get rid of
9 the socket specials in /var/paraslash.
10 - The --user-allow option of para_audiod now accepts also
11 usernames rather than only user IDs.
12 - New autoconf macros to avoid duplication in configure.ac.
13 - Status items (as shown by para_gui) are updated correctly
14 when the meta information of the current audio changes.
15 - para_server and para_audiod no longer refuse to start in
16 the background if no log file is given. Instead, all log
17 messages go to /dev/null in this case.
18 - Web page cleanup.
19
20
21 Download: ./releases/paraslash-git.tar.bz2 (tarball)
22
23 ------------------------------------------
24 0.5.4 (2015-01-23) "exponential alignment"
25 ------------------------------------------
26
27 Another cleanup and bugfix release.
28
29 - New server command: tasks.
30 - Minor cleanups to daemon.c.
31 - New URLs for home page and git services.
32 - Improved error diagnostics for the mvblob commands.
33 - New sender subcommand: status.
34 - Improved help text for server and afs commands.
35 - audiod memory leak fixes.
36 - Miscellaneous improvements to the build system.
37 - oss_writer improvements.
38 - Improved handling of mp3 files with both id3v1 and id3v2 tags.
39
40 Downloads: ./releases/paraslash-0.5.4.tar.bz2 (tarball)
41 ./releases/paraslash-0.5.4.tar.bz2.asc (signature)
42
43 ---------------------------------------------
44 0.5.3 (2014-08-01) "symbolic synchronization"
45 ---------------------------------------------
46
47 Not many new features, but lots of fixes and usability improvements.
48
49 - para_gui has been converted to use the paraslash scheduler.
50 - Various alsa-related fixes, mostly for the raspberry pi.
51 - Many scheduler improvements and cleanups.
52 - The test suite has been extended to include sanity checks
53 for the generated man pages.
54 - ao_writer fixes. This writer was in a quite bad shape. Many
55 serious bugs have been fixed.
56 - new audiod command: version.
57 - Minor improvements to the bitstream API.
58 - The cpsi command now prints a meaningful error message if
59 none of the given patterns matched any audio file.
60
61 Downloads: ./releases/paraslash-0.5.3.tar.bz2 (tarball),
62 ./releases/paraslash-0.5.3.tar.bz2.asc (signature)
63
64 ----------------------------------------
65 0.5.2 (2014-04-11) "orthogonal interior"
66 ----------------------------------------
67
68 The new sync filter, the AES_CTR128 stream cipher and the overhauled
69 network code are the highlights of this release. It also includes a
70 fair number of smaller fixes and improvements not mentioned here.
71
72 - The new sync filter synchronizes playback between multiple
73 clients.
74 - Connections between para_server and para_client are now
75 encrypted by means of AES rather than RC4 if both sides
76 support it. RC4 is still available as a fallback. This
77 feature is fully transparent, i.e. no command line options
78 are necessary, and a client linked against openssl can
79 speak with a server linked against libgcrypt and vice versa.
80 - Major cleanup of the networking subsystem.
81 - Improvements to para_fade: the new set mode, multi-channel
82 initial volumes, better error logging.
83 - The man pages of para_audiod, para_filter, para_recv, and
84 para_write contain the relevant options for receivers, filters,
85 writers. This broke in 0.5.0.
86 - ogg/vorbis latency improvements.
87 - Improved user manual.
88 - Minor fixes to avoid clang warnings.
89
90 Downloads: ./releases/paraslash-0.5.2.tar.bz2 (tarball),
91 ./releases/paraslash-0.5.2.tar.bz2.asc (signature)
92
93 ------------------------------------------
94 0.5.1 (2013-12-20) "temporary implication"
95 ------------------------------------------
96
97 Lots of fixes and improvements all over the place, and a major overhaul
98 of the build system.
99
100 - Audiod improvements and fixes.
101 - Buffer tree robustness improvements.
102 - Cleanup of the mood subsystem.
103 - Fixes and cleanups for the flac decoder.
104 - Latency improvements for the ogg/opus decoder.
105 - Crypto support is now optional. On systems without
106 openssl/gcrypt, the build succeeds but para_server,
107 para_audiod, para_client won't be built.
108 - The build system now works for cross-compile setups.
109 - The dependency tree has been flattened, which speeds up
110 builds and avoids to recreate the man pages on every change.
111 - The error code helper has been rewritten from perl to C,
112 which further improves build time.
113 - Many small bugs in the build system have been identified
114 and fixed.
115
116 Downloads: ./releases/paraslash-0.5.1.tar.bz2 (tarball),
117 ./releases/paraslash-0.5.1.tar.bz2.asc (signature)
118
119 ----------------------------------------
120 0.5.0 (2013-08-23) "invertible validity"
121 ----------------------------------------
122
123 Some API-breaking changes, one serious bug fix, and a lot of bike-shedding.
124
125 - The sideband compatibility code has been removed, hence
126 sideband connections (introduced in 0.4.11) are now mandatory.
127 - Addblob commands can produce output.
128 - The stat command no longer sends garbage when para_server was
129 compiled against libgcrypt.
130 - Dependencies for gengetopt files are computed automatically.
131 This eliminates a constant source of build bugs.
132 - The setatt command now accepts file name patterns rather than only
133 path names.
134 - overview.pdf is now based on dia, a simple diagram creation program.
135 The new version is much more detailed and contains descriptions of
136 the various programs of the paraslash package.
137 - The separator of all multi-word options has been changed from
138 underscore to dash. For example --log_color becomes --log-color.
139 - Overhauled web pages and the new logo.
140
141 Downloads: ./releases/paraslash-0.5.0.tar.bz2 (tarball),
142 ./releases/paraslash-0.5.0.tar.bz2.asc (signature)
143
144 --------------------------------------
145 0.4.13 (2013-07-29) "spectral gravity"
146 --------------------------------------
147
148 One more 0.4.x release before the API-breaking changes for 0.5.0 go
149 in. The main features of this release are the ogg/opus audio format,
150 and UTF-8 support, but it includes also tons of other improvements
151 and fixes all over the place.
152
153 - New audio format: ogg/opus.
154 - UTF8 support for para_gui and the mp3 audio format handler.
155 - Scheduler improvements and fixes.
156 - The obsolete gettimeofday() function has been replaced
157 by clock_gettime() on systems which support it.
158 - Speed and usability improvements for para_gui.
159 - para_client now restores the fd flags of stdin and stdout
160 on shutdown.
161 - Improved manual pages.
162 - Consistent version strings for all executables.
163 - Reduced dependencies on generated files result in fewer
164 recompilations on changes.
165 - Performance improvements for the compress filter.
166 - Improved downloads web page.
167
168 -----------------------------------------
169 0.4.12 (2012-12-20) "volatile relativity"
170 -----------------------------------------
171 The new command line player, the resample filter, ALSA support for
172 para_fade, and the improved build system are the highlights of this
173 release which probably marks the end of the 0.4.x series.
174
175 - The afh receiver and the para_play executable.
176 - The resample filter: A sample rate converter based on
177 libsamplerate.
178 - The "versions" directory has been removed from the master
179 branch. The tarballs of the old releases are now available
180 in the new "releases" branch.
181 - Overhaul of the build system: All generated files are now
182 written to the "build" directory.
183 - The modular mixer API and the alsa mixer.
184 - Minor fixes for the osx writer.
185
186 --------------------------------------
187 0.4.11 (2012-07-20) "mutual diversity"
188 --------------------------------------
189
190 The major feature in this release is the new sideband API for
191 client-server communication. This API will be used exclusively starting
192 with 0.5.0, which breaks backward compatibility but allows to get rid
193 of quite some compatibility code. Other noteworthy changes include
194 decoder latency improvements and a long-standing bug fix for the
195 ALSA writer.
196
197 - Sideband connections: If both para_server and para_client
198 support this feature, data is sent as a multiplexed stream.
199 - The --no_default_filters option of para_filter has been
200 removed.
201 - Several fixes and latency improvements to various decoders.
202 - The ALSA writer now limits the prebuffer time to 500ms.
203 - Documentation improvements.
204 - Overhaul of the command_util.sh script.
205 - Fixes for some minor problems found by the clang analyzer.
206 - Compiles (almost) without warnings on gcc-3.
207 - Robustness improvements of the buffer tree code.
208
209 ------------------------------------------
210 0.4.10 (2012-03-30) "heterogeneous vacuum"
211 ------------------------------------------
212
213 Nothing earth-shaking in this release, but quite a few usability
214 improvements and the usual mix of cleanups and fixes.
215
216 - The --no_default_filters option of para_filter has been
217 deprecated. It still works but has no effect and will be
218 removed in the next version.
219 - para_gui now prints also the stderr output of the executing
220 command in the bottom window.
221 - Cleanup and consolidation of the various wrappers for
222 write(), writev(), send() and friends.
223 - The obscure error messages on mmap() failures have been
224 replaced by meaningful messages. This affects mainly
225 para_afh.
226 - para_audioc: Cleanups and memory leak fixes.
227 - Test 0004-server no longer fails if para_server is not
228 being built.
229 - New configure options: --with-id3tag-{headers,libs}.
230
231 -------------------------------------
232 0.4.9 (2011-12-06) "hybrid causality"
233 -------------------------------------
234
235 Support for another audio format, interactive mode for para_client
236 and para_audiod and many small improvements/fixes all over the place.
237
238 - Support for flac, the free lossless audio codec.
239 - Fix for an endless loop in the mp3 decoder for certain
240 (corrupt) mp3 files.
241 - When executed without specifying a command, para_client
242 and para_audioc start an interactive shell (requires
243 libreadline being installed). The interactive mode offers
244 full tab completion and command line history.
245 - autogen.sh now detects a distcc setup and adjusts the
246 parameter for the -j option of make accordingly.
247 - Shared memory areas are no longer restricted to 64K. We now
248 detect the maximal size of a shared memory area at runtime.
249 - cleanup of the internal uptime API.
250 - para_server prefaults the mmapped audio file to avoid
251 delays on slow media.
252 - A new test for the test-suite that exercises the
253 communication between para_server and para_audiod.
254 - The alsa writer eats up less CPU cycles when configured to
255 use the DMIX plugin.
256 - Simplified and unified receiver code.
257 - Makefile cleanups.
258 - Commands which print a list of matching audio files now
259 emit a meaningful error message if no audio file matched the
260 given pattern(s).
261
262 --------------------------------------
263 0.4.8 (2011-08-19) "nested assignment"
264 --------------------------------------
265
266 Gcrypt support, the overhauled osx writer and regex format specifiers
267 are the highlights of this release.
268
269 - support for libgcrypt as a drop-in replacement for openssl.
270 Run configure --enable-cryptolib=gcrypt to link against
271 libgcrypt. The two crypto implementations are compatible to
272 each other, i.e. a para_client executable linked against
273 gcrypt can connect to para_server linked against libssl
274 and vice versa.
275 - Rewrite of the osx writer (output plugin for Mac OS).
276 - audiod: The format specifier for receivers, filters and
277 writers is now treated as a regular expression. This allows
278 to replace 5 lines in the config file (one for each audio
279 format) by one single line. See the manual for details.
280 - The *.cmdline.[ch] files are no longer contained in the released
281 tarballs. This reduces the size of the tarballs but requires
282 gengetopt to build the tarball.
283 - Compiles cleanly also with llvm/clang.
284 - Corrupt mp3 files are handled more gracefully.
285 - The alsa writer uses poll fds instead of computing timeouts.
286 - Cleanup of the generic writer API.
287 - sched: Optimized zero timeouts.
288 - vss timeout cleanups.
289 - oggdec fixes and improvements.
290
291 --------------------------------------
292 0.4.7 (2011-06-01) "infinite rollback"
293 --------------------------------------
294
295 The new ao writer, support for ssh RSA keys and a couple of other
296 enhancements.
297
298 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
299 platforms through the libao audio library.
300 - Support for RSA keys generated with ssh-keygen.
301 - configure: improved options for ogg/vorbis/speex.
302 - The git version reported by --version always matches HEAD.
303 - The autogen script detects the number of processors and
304 runs a parallel make if possible.
305 - Major cleanup of the crypto API.
306 - Documentation updates.
307
308 ------------------------------------------
309 0.4.6 (2011-03-31) "deterministic entropy"
310 ------------------------------------------
311
312 Lots of ogg/vorbis improvements, the new test suite, enhancements
313 for para_gui and a fair amount of other bug fixes.
314
315 - For DCCP/OGG streams the audio file header is only sent once
316 at the beginning of the stream rather than periodically
317 every five seconds. This reduces network traffic and the
318 FEC group size.
319 - The vorbis comment header is replaced by an empty dummy header
320 before the header is sent over the network. This also results in
321 less network traffic and smaller FEC groups.
322 - The new "test" make target allows to perform some sanity checks prior
323 to installing the package.
324 - ogg timing fixes and performance improvements
325 - Scheduler improvements
326 - Proper exit codes for para_write
327 - para_gui: New option --theme to select a startup theme. Several
328 other improvements and fixes.
329 - aacdec error message cleanups
330 - simplified color error handling
331
332 --------------------------------------------
333 0.4.5 (2010-12-17) "symmetric randomization"
334 --------------------------------------------
335
336 Bug fixes, internal cleanups and variable-sized FEC slices.
337
338 - Contains a fix for an invalid-free-bug in the ogg audio format
339 handler code.
340 - Switching off the DCCP sender works again.
341 - para_audiod handles crashes of para_server more robustly.
342 - Internal scheduler and writer cleanups.
343 - Reduced latency due to variable-sized FEC slices.
344 - Improved documentation and error diagnostics.
345 - The build of para_server is now optional, allowing the build
346 to succeed in case libosl is not installed.
347
348 ------------------------------------------
349 0.4.4 (2010-08-06) "persistent regularity"
350 ------------------------------------------
351
352 Support for yet another audio format, para_write improvements and
353 bug fixes.
354
355 - Support for the speex codec.
356 - Support for sample formats other than 16 bit little endian.
357 - error2.h is now created by a perl script which speeds up configure
358 considerably.
359 - Fix a bug in the aac decoder which could lead to segfaults in
360 para_filter/para_audiod.
361 - Fixes for autoconf-2.66.
362
363 ----------------------------------------
364 0.4.3 (2010-07-05) "imaginary radiation"
365 ----------------------------------------
366
367 Many improvements for the DCCP and the UDP transport, the new user
368 manual and the usual mix of bug fixes and internal improvements.
369
370 - FEC support for the DCCP sender (Gerrit Renker). The new
371 --dccp_max_slice_size, --dccp_data_slices_per_group and
372 --dccp_slices_per_group options can be used to set the FEC
373 parameters for the DCCP transport.
374 - DNS lookups for UDP targets (Gerrit Renker).
375 - The new user manual replaces the README, README.afs, REQUIREMENTS
376 and INSTALL documents.
377 - Fix an end-of-file detection bug in the oggdec filter.
378 - The new nonblock API.
379 - Both options of the oggdec filter have been removed.
380 - New debug mode for the internal scheduler.
381
382 ------------------------------------------
383 0.4.2 (2010-04-23) "associative expansion"
384 ------------------------------------------
385
386 It's been some time since the last release, but finally here is
387 paraslash-0.4.2. The bulk of the changes comes from the new buffer
388 tree API, but there are changes all over the tree. Mainly performance
389 and usability improvements, but also quite some bug fixes.
390
391 - The new buffer tree API.
392 - DCCP: Support for CCID negotiation (Gerrit Renker).
393 - UDP robustness fixes.
394 - The --bufsize option for mp3dec is gone as it no longer makes sense
395 for the new buffer tree API.
396 - Fix audible buffer underruns for wma streams.
397 - The alsa writer no longer prints meaningless underrun durations.
398 - audiod: Defaults work also for udp streams. If no filter is
399 given for an audio format that is received via upd, fecdec is
400 automatically added as the first filter (along with the decoder).
401
402 ---------------------------------------
403 0.4.1 (2009-12-22) "concurrent horizon"
404 ---------------------------------------
405
406 Support for another audio format, minor feature enhancements and lots of bug
407 fixes. All fixes that have been accumulated in the maint branch (in particular
408 those mentionened in the 0.3.6 release notes) appear in this release as well.
409
410 - wma support.
411 - new afh option: --human to activate human-readable output.
412 - new server/audiod option: --log-timing to print timing information.
413 - build system improvements.
414 - source code documentation updates.
415
416 -------------------------------------
417 0.3.6 (2009-12-07) "cubic continuity"
418 -------------------------------------
419
420 Quite a few bugs have been found and fixed since 0.3.5, so here's
421 another 0.3.x release. No new features.
422
423 - Always check return value of malloc().
424 - ogg vorbis/FEC: Do not write garbage after the audio file header.
425 - exit if root privileges could not be dropped.
426 - FEC: Fix computation of extra slices.
427 - oss: Fix check for empty input buffer.
428 - Avoid buffer underruns due to filter chain output buffer constraints.
429 - server: Fix assignment of afs_pid.
430 - Don't panic if the afs database contains unknown audio formats.
431 - http/dccp: Do not send the audio file header twice.
432 - FEC: Timing improvements.
433
434 ----------------------------------------------
435 0.4.0 (2009-11-10) "simultaneous independence"
436 ----------------------------------------------
437
438 Two significant changes which require the new version number: The
439 improved authentication dialog and the fact that the database code
440 has been moved to a library, libosl. To use the new version, you have
441 to generate new RSA keys, see INSTALL for details. A shell script is
442 provided for conversion of the 0.3 database to the new 0.4 format.
443
444 - stronger crypto for client authentication
445 - the database code has been moved to a library
446 - improved status item handling
447 - cleanup of the build system
448 - The "-V" option now also prints the git version
449 - the new parser-friendly listing mode for the ls and stat commands
450 - mandatory rc4 encryption
451 - major audio format handler cleanups
452 - (id3,...) tags are no longer stored as a combined string in the database
453 - new mood methods: artist_matches, title_matches, comment_matches,
454 album_matches, year_maches, year.
455
456 --------------------------------------------
457 0.3.5 (2009-09-21) "symplectic separability"
458 --------------------------------------------
459
460 Full client support for *BSD Unixes, complete re-write of the ogg
461 vorbis audio format handler, various improvements all over the place
462 and the usual mix of bugfixes. This release marks the end of the 0.3
463 series if no serious problems show up.
464
465 - the new oss writer (supported on *BSD and Linux)
466 - rewrite of the ogg vorbis audio format handler. It's
467 recommended to replace the chunk tables of existing ogg
468 vorbis files in the afs database by re-adding these files
469 with "add -f".
470 - support for netmask subsets (Gerrit Renker)
471 - the new prebuffer filter
472 - improved signal handling
473 - variable fec output buffer size
474 - improved FEC timing fixes audible buffer underruns in UDP mode
475 - --log_color actually works
476 - new ls option: -d (print dates as seconds after the epoch)
477 - update to gengetopt 2.22.2
478 - support for RSA keys of size > 512 bits
479 - new option "mixer_channel" for para_fade
480
481 -----------------------------------------
482 0.3.4 (2009-05-07) "elliptic inheritance"
483 -----------------------------------------
484
485 The new udp sender, forward error correction, colored logs and various
486 other improvements. As the udp sender does not depend on any special
487 libraries, it is built unconditionally.
488
489 - The udp sender replaces the ortp sender. The new code uses forward
490 error correction to protect against packet losses. Many thanks to
491 Gerrit Renker for providing ipv6 support.
492 - The default port for udp streaming now defaults to 8000, like
493 for the http and the dccp senders/receivers.
494 - Loglevels are now specified as symbolic names, e.g.
495 "--loglevel info".
496 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
497 - para_server/para_audiod: Color support for log messages.
498 - new options for mp3dec: --ignore-crc, --bufsize
499 - new audiod option: --config-file.
500 - gengetopt cleanups.
501 - Improved help/man pages: The documentation of para_audiod,
502 para_recv, para_filter and para_write now also contains
503 all options of the available receivers/filters/writers. The
504 man page of para_fade contains a description of the different
505 modes of operation.
506 - More source code documentation.
507 - vss timing fixes.
508
509 --------------------------------------------
510 0.3.3 (2008-12-01) "axiomatic perspectivity"
511 --------------------------------------------
512
513 Internal code cleanups, bug fixes, improved tag handling and the new
514 amplification filter.
515
516 - para_server uses the generic scheduling code.
517 - overhaul of the virtual streaming system.
518 - mp3: id3 version 2 support via libid3tag (optional)
519 - ogg: vorbis comment support.
520 - aac meta info support.
521 - mp3 audio format handler cleanups.
522 - new filter: "amp" to amplify the amplitude of the audio stream
523 - new status item/database entry: amplification. It is
524 used by the amp filter to pre-amplify the audio stream.
525 - fix a close-without-open bug in para_write.
526 - fix a bug in com_init() which was introduced in 0.3.2.
527 - better error diagnostics for para_client.
528
529 -----------------------------------------
530 0.3.2 (2008-04-11) "probabilistic parity"
531 -----------------------------------------
532
533 The new para_afh executable, scheduling and documentation improvements.
534
535 - new ls option: -lc (list chunk table)
536 - new executable: para_afh, the stand-alone audio file handler tool
537 - afs commands can send output more than SHMMAX (32MB on Linux). This
538 also reduces the memory usage of commands that produce large amounts
539 of output.
540 - major scheduler and audiod cleanups.
541 - more detailed and much nicer man pages.
542
543 ---------------------------------------
544 0.3.1 (2008-02-23) "liquid interaction"
545 ---------------------------------------
546
547 A mix of cleanups, bug fixes, improvements, and some new features. No
548 significant changes to the new database (osl) code, which is generally
549 a good sign.
550
551 - Share some similar/duplicate code between the http and the
552 dccp sender.
553 - Generic access control lists for paraslash senders.
554 - dccp sender: Access control lists, connection limiting and support
555 for the allow,deny,on,off,help sender commands.
556 - The default dccp port changed from 5001 to 8000 (suggested by
557 Gerrit Renker).
558 - para_server starts even if not all public keys could be loaded.
559 - Audiod performance improvements.
560 - fix a bug in the "off" command of the http sender.
561 - fix some fd and memory leaks.
562 - Update to gengetopt-2.22.
563
564 -------------------------------------
565 0.3.0 (2008-01-12) "solar saturation"
566 -------------------------------------
567
568 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
569
570
571 Usually one might expect lots of new features AND a big increase in size
572 for a major release like this.
573
574 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
575 ever. The decrease in size is mostly due to the removal of some
576 graphical tools (which were only quick hacks anyway). But also the
577 fact that the mysql code is gone cuts down the size a bit.
578
579 Being independent of mysql comes at a cost: The fact that paraslash
580 now contains its own database (the object storage layer, osl) increases
581 the (stripped) binary size of para_server by ~50K on i386.
582
583 - no more restrictions on unique basenames.
584 - independent of mysql: The new self-contained object
585 storage layer (osl) replaces the mysql database.
586 - New executable para_fsck: Check integrity of osl tables.
587 - Lyrics support.
588 - Reliable audio file move/rename detection.
589 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
590 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
591 Solaris (sparc64).
592 - the new osl-based audio file selector (afs) replaces the random,
593 the playlist and the mysql selector of paraslash-0.2.x.
594 - IPv6 support (thanks to Gerrit Renker).
595 - paraslash-0.2.x streams are now called "moods". Writing
596 0.3.x-mood definitions should be both easier and more
597 powerful than writing 0.2.x-stream definitions.
598 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
599 been removed. The world is a better place without them. However,
600 para_gui is still there.
601 - afs tracks audio file selection also in playlist mode.
602 - few easy-to-use afs commands replace the many not-so-easy-to-use
603 mysql commands (and are available also in playlist mode).
604 - Improved error subsystem.
605 - The earth-shaking new logo.
606
607 -----------------------------------------
608 0.2.17 (2007-11-20) "isotropic threshold"
609 -----------------------------------------
610
611 Mainly bugfixes and cleanups in this version which marks the end of
612 the 0.2.x series if no serious bugs show up after the release.
613
614 - mysql_selector: fix a locking bug.
615 - universal chunk queueing.
616 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
617 to Gerrit Renker).
618 - be more carful wrt. signed vs. unsigned argument passing.
619 - cleanup error.h and fix some references to invalid error
620 codes.
621 - update to gengetopt-2.21.
622 - update to ortp-0.13.1.
623 - autoconf: extend checks for headers, library functions and
624 compiler characteristics.
625 - Fix streaming of large mp3 files.
626 - Fix an off-by-one bug in playlist handling.
627
628 --------------------------------------
629 0.2.16 (2007-04-05) "neural discharge"
630 --------------------------------------
631
632 The main change in this release is the major audio format handler
633 cleanup which removes some similar/duplicate code and makes it easier
634 to implement plugins for other audio formats. Of course, the usual mix
635 of other improvements/changes/bugfixes also made it into the release.
636
637 - simplified audio format handlers (most of the handling functions
638 were moved one layer up to the virtual streaming system).
639 - para_server uses mmap to read audio files
640 - repositioning of mp3 streams is much faster, in particular for
641 jumping near the end of large mp3 files.
642 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
643 and AFS_WRITE.
644 - fix a bug in para_filter that caused decoding of aac files
645 to start only after a few seconds.
646 - fix osx_writer hangs
647 - simplified dccp code (thanks to Gerrit Renker)
648 - the compress filter works also on big endian systems (ppc)
649
650 -----------------------------------------
651 0.2.15 (2007-02-16) "inductive resonance"
652 -----------------------------------------
653
654 Minor improvements, more documentation and a bunch of bug fixes.
655
656 - para_server: The server.users file is only read once on server
657 startup rather than for each connection
658 - mp3dec: Fix decoding of corrupt mp3 files
659 - afs (audio file sender) is now called vss (virtual streaming
660 system). Consequently, the permission flags specified in
661 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
662 become VSS_READ and VSS_WRITE respectively.
663 - para_audiod/para_filter: Fix a bug that caused the last chunk
664 of audio data not being written under certain circumstances
665 - audiod: compute the difference of server time and local time
666 correctly
667 - para_server/para_audiod: Fix some memory leaks
668 - documentation improvements
669 - configure.ac: fix checks for para_krell
670 - new man pages
671
672 -------------------------------------------
673 0.2.14 (2006-10-15) "transient singularity"
674 -------------------------------------------
675
676 The only major enhancement of this version is the osx writer which completes
677 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
678 until now for reasons beyond the scope of this changelog entry.
679
680 - new output plugin for Mac Os: the osx writer
681 - rename configure command line options from --enable-xxx-headers to
682 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
683 - configure: new command line options: --with-mad-headers,
684 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
685 - some robustness fixes
686 - dymamic audio format recognition for audiod
687 - para_server: new command line option: --autoplay_delay
688 - para_audiod: new command line option: --clock_diff_count
689
690 ---------------------------------------
691 0.2.13 (2006-07-14) "sonic convolution"
692 ---------------------------------------
693
694 A bunch of new features and core changes.
695
696 - the new paraslash scheduler, short and sweet.
697 - Support for m4a/mp4 files via the new aac audio format
698 handler/filter (requires libfaad).
699 - each writer has its own command line parser, just like
700 para_recv and para_filter.
701 - para_client and para_audioc use the error subsystem
702 - writers are integrated in para_audiod (currently linux-only)
703 - para_client is integrated in para_audiod
704 - random/playlist selector: improved info strings
705 - new audiod commands: tasks, kill
706 - update to libortp-0.10.1
707 - para_fade: wake time defaults to 8 hours from now
708 - update to autoconf-2.60
709
710 ------------------------------------------
711 0.2.12 (2006-05-12) "oriented abstraction"
712 ------------------------------------------
713
714 Many user-visible changes in this release and lots of new
715 features:
716
717
718 - the new optional dccp sender/receiver. It uses the datagram
719 congestion control protocol. You'll need a fairly new kernel
720 for this.
721 - paraslash works on Mac OS X (thanks to Gerd Becker)
722 - para_play renamed to para_write
723 - modular output plugin design (writers) for para_write
724 - new file_writer output plugin for para_write
725 - compress filter speed improvements
726 - update to libortp-0.9.1
727 - update to gengetopt-2.17rc
728 - para_client no longer depends on libreadline (as the
729 code for the interactive mode was removed).
730 - gcc-2-95 is no longer a supported compiler. It may still
731 work, but it gets no more testing.
732 - the tarball no longer contains the screenshot images which
733 reduces its size quite a bit.
734 - configure: new command line options: --enable-mysql-headers
735 and --enable-mysql-libs
736
737 ------------------------------------
738 0.2.11 (2006-03-11) "atomic duality"
739 ------------------------------------
740
741 Here it is, the first paraslash release developed with git. There
742 are fairly many user-visible changes in this release. As two out of
743 the three "database tools" of paraslash don't use a database at all,
744 they are now called "audio file selectors" instead.
745
746
747 - the cdt command (change database tool) becomes chs (change
748 selector)
749 - no more colon separators: The syntax of some options of
750 para_audiod and para_filter have changed. Use --help for
751 more info (and some examples).
752 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
753 - switch from cvs to git (should've done that earlier)
754 - the new ipc subsystem
755 - new audio file selector: playlist
756 - para_server: the dopey selector is now called "random",
757 and is the default selector. Use the --selector option to
758 choose another selector at startup, or the chs command to
759 change the selector at runtime.
760 - X86_64 fixes (thanks to Steffen Klassert)
761 - para_play fixes
762
763 --------------------------------------
764 0.2.10 (2006-02-17) "cyclic attractor"
765 --------------------------------------
766
767 Huge documentation update, a scrollable window for para_gui, ortp
768 improvements, and of course many small fixes not mentioned here.
769 The diffstat below is rather misleading as most insertions are due
770 to the new source documentation.
771
772 - autoconf cleanup
773 - para_server also uses the new error subsystem
774 - lots of new documentation (UTSL)
775 - gui improvements:
776 - keysyms for cursor keys and for next/previous page keys
777 - scrollable output window
778 - new internal commands: scroll up/down, page up/down
779 - fix color of command output.
780 - ortp: the --chunk_time and --header flags are no longer needed
781 for para_recv/para_audiod as this information is now encoded in
782 each rtp packet sent by para_server.
783
784 -------------------------------------------
785 0.2.9 (2006-01-24) "progressive turbulence"
786 -------------------------------------------
787
788 Internal audiod receivers/filters, the new error subsystem and
789 a lot of small improvements.
790
791 - para_recv and para_filter are integrated into the para_audiod
792 binary, i.e. audiod no longer spawns a new process for
793 each receiver/filter. As para_recv and para_filter might be
794 useful as standalone programs, they still get built (linked
795 against the same object files that are also used for audiod).
796 - further ortp timing improvements should reduce the CPU usage
797 of the ortp receiver.
798 - improved audio grabbing. The 'grab' command of para_audiod
799 has its own set of command line options. Read the output of
800 "para_audioc help grab" for more info.
801 - oggdec: configurable input prebuffer size.
802 - the new error subsystem gives better error diagnostics
803 and reduces code size.
804
805 -----------------------------------------
806 0.2.8 (2006-01-02) "dynamic accumulation"
807 -----------------------------------------
808
809 The new modular filter design and the para_play-hangs bugfix.
810
811 - new executable: para_filter. It combines para_mp3dec,
812 para_oggdec and para_compress. It also adds a further filter
813 type, wav, that just inserts a wave header at the desired point
814 of the filter chain. All 'piping' is done in-memory (i.e. no
815 read/write operations are used).
816 - para_play: fix a stupid bug that caused it to hang under
817 certain circumstances.
818
819 -------------------------------------------
820 0.2.7 (2006-12-27) "transparent invariance"
821 -------------------------------------------
822
823 Not many user-visible changes but a fair amount of internal improvements.
824
825
826 - The http sender buffers data if it can not be sent
827 out immediately (because the socket is not writable). This
828 should prevent para_server from shutting down the connection
829 too early on a loaded network.
830 - para_play also prebuffers data if it is told to start at a
831 future time by the --start_time option.
832 - The return of para_recv: It combines para_ortp_recv and
833 para_http_recv. Use the --receiver option to switch between
834 the two. para_recv builds without libortp, but contains
835 only the http receiver in this case.
836 - update to ortp 0.8.1. As this ortp release contains incompatible
837 changes, para_recv-0.2.7 won't link against older ortp libs.
838 - improved ortp timings.
839 - use of gcc-extensions that #define away for non-gcc and
840 gcc < 3.0.
841
842 -------------------------------------------
843 0.2.6 (2005-10-29) "recursive compensation"
844 -------------------------------------------
845
846 Transparent session encryption (uses openssl's Alleged RC4 cipher),
847 the internal find command and several other improvements and cleanups.
848
849 - Encrypt the session if encryption is requested by the client
850 (default for para_client 0.2.6). This is backwards
851 compatible, so older clients can still connect to para_server
852 0.2.6. Use the new client option --plain to request an
853 uncrypted session (off by default, must be set to on in
854 order to connect to para_server 0.2.x with 0 <= x <= 5).
855 - para_server uses an internal function to locate audio files
856 rather than calling find(1). The server option
857 --mysql_audio_file_dir replaces --mysql_find_cmd.
858 - documentation update
859 - man pages
860 - header file cleanup
861 - para_client code cleanup
862 - para_gui: faster display of output of display commands
863
864 ------------------------------------------
865 0.2.5 (2005-10-13) "aggressive resolution"
866 ------------------------------------------
867
868 This release adds internal senders, i.e. no more external programs are
869 spawned for sending out the audio data. There are two different senders
870 available: The http sender and the ortp sender (former para_send which
871 is no longer needed).
872
873 The new sender code has a plugin-like design so it can be easily
874 extended should there be be any future need for supporting another
875 network streaming protocol. All senders are completely independent of
876 each other. In particular, the http and the ortp sender can operate
877 in parallel.
878
879 - new server command: sender to control senders at runtime.
880 Read the output of "para_server -h" and "para_client help
881 sender" for more information.
882 - para_recv renamed to para_ortp_recv
883 - new executable: para_http_recv, a simple command line
884 http receiver.
885 - major afs/mp3/ogg code simplifications due to internal
886 senders.
887 - ogg timing improvements
888 - fix several minor memory leaks (found by valgrind)
889 - empty stream definitions work again
890 - com_ne(): ignore errors on remove
891 - audiod: fix segfault on server restart
892
893 ---------------------------------------
894 0.2.4 (2005-09-21) "toxic anticipation"
895 ---------------------------------------
896
897 Several small improvements, fixes and the new grab command.
898
899 - audiod:
900 - new command: "grab" to grab the output of the stream reader
901 or any filters. Read the output of "para_audioc help grab"
902 for more information.
903 - fix memory leak
904 - code cleanup
905 - audioc: new command line option: --bufsize to specify a
906 buffer size different from the default size 8192.
907 - improved error diagnostics for para_play.
908 - new configure option: --enable-ssldir so search for openssl in
909 non-standard places
910 - sdl_gui: Make it look nice again for 1024x768
911 - server: report total size of memory allocated with sbrk by malloc,
912 new command line option: --announce_time
913
914 -----------------------------------------
915 0.2.3 (2005-09-01) "hydrophilic movement"
916 -----------------------------------------
917
918 Two new executables and major feature enhancements.
919
920 - audiod filters: It is now possible to specify arbitrary many
921 (including none) filters for each supported audio
922 format. This can be used e.g. for normalizing volume,
923 transforming or grabbing the audio stream, or for using
924 visualizers. Read the output of "para_audiod -h" for the
925 syntax of the new --filter_cmd option.
926 - new executable: para_play, a tiny alsa player. It
927 can play wave files or raw pcm (16 bit little endian)
928 from stdin.
929 - new executable: para_compress, a dynamic range compressor
930 intended to keep audio output at a consistent volume. Derived
931 from AudioCompress, http://trikuare.cx/code/AudioCompress.html.
932 - audiod: New option: --stream_delay. This can be used in
933 a local network to syncronize the audio output of all
934 clients that play the same stream.
935
936 ------------------------------------------
937 0.2.2 (2005-08-19) "tangential excitation"
938 ------------------------------------------
939
940 Mostly internal changes in this release, but also some new commands
941 for the mysql database tool.
942
943 - cleanup exec.c, fix para_exec bug
944 - compile time loglevel (log messages below the given level
945 won't be compiled in, which reduces the size of the
946 resulting binaries)
947 - new log macros that shorten the size of the source code.
948 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
949 to send only zeros. With this workaround, para_audioc works
950 again.
951 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
952 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
953 (new entry), "snp" (set numplayed). Read the manual for more
954 information.
955
956 ---------------------------------------
957 0.2.1 (2005-08-15) "surreal experience"
958 ---------------------------------------
959
960 Here comes paraslash-0.2.1. It contains a couple of new features and,
961 surprise, only minor bug fixes.
962
963 - kill noisy mp3 debug message
964 - cleanup of the build system
965 - para_server and para_client directly use the crypto routines
966 of the openssl library rather than invoking the openssl command
967 line utitlity
968 - server/audiod: new option --user to switch to the given user
969 when invoked as root. Read the output of "para_server -h" for
970 more information.
971 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
972 status. Default: "para_audioc stat"
973 - sdl_gui: new option --pic_cmd to be used to download the picture.
974 Default: "para_client pic"
975 - audiod: 5 slots ought to be enough for everybody
976 - audiod: new status item: Uptime, kill hup command
977
978 ------------------------------------------
979 0.2.0 (2005-08-06) "distributed diffusion"
980 ------------------------------------------
981
982 After several month of increased development activity, paraslash-0.2.0
983 has arrived. It contains many new features and is much more
984 self-contained than the old 0.1.x series. Enjoy!
985
986
987 - para_server: fix hang on song change and crash on sighup.
988 Speed up mysql queries. The DIR_LIKE macro is gone.
989 - new executables: para_audiod, the local audio daemon that
990 starts playback (uses SCM_CREDENTIALS socket magic) and
991 para_audioc, the corresponding client.
992 - new executables: para_mp3dec/para_oggdec, two really teensy
993 decoders. para_mp3dec is based on libmad, para_oggdec requires
994 libvorbisfile.
995 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
996 they are now called para_send and para_recv respectively.
997 - documentation updates
998 - para_gui is themable. For now there is the default theme that
999 looks as before and the simple theme: blue and easy.
1000 - gui: audio streaming is now handled by audiod. Time display shows
1001 playback time rather than streaming time
1002 - slider: update to libzmw-0.2.0
1003 - para_krell: fix crash on server shutdown
1004 - switch from gzip to bzip2
1005
1006 ----------------------------------------
1007 0.1.7 (2005-04-18) "melting penetration"
1008 ----------------------------------------
1009
1010 The main change in this release is clearly the oggvorbis rewrite,
1011 but there are also lots of smaller changes. If you intend to use both
1012 the mp3 and the ogg plugin, it is recommended to use software mixing,
1013 e.g. the dmix plugin which is provided by ALSA.
1014
1015 - new executables: para_ovsend and para_ovrecv for sending/receiving
1016 oggvorbis files via rtp. Requires the open rtp library. Get it at
1017 http://www.linphone.org/ortp/
1018 - rewrite of the ogg_vorbis core code
1019 - configure detects libzmw and, if detected, includes
1020 para_slider to the list of binaries to be built by make
1021 - server stream writers read from their associated fifo rather
1022 than from stdin
1023 - slider: two new sliders, lastplayed and numplayed
1024 - fix nasty double free bug which caused random segfaults in case of
1025 mp3 files with invalid header information
1026 - gui: new command line option: --stream_timeout=seconds to
1027 deactivate a slot if it is idle for that many seconds (default=`5')
1028 - diffstats
1029
1030 ---------------------------------------
1031 0.1.6 (2005-03-05) "asymptotic balance"
1032 ---------------------------------------
1033
1034 Only little user-visible changes in this release. Mainly bugfixes and
1035 core code cleanup. This is probably the most stable version ever if you
1036 stick to mp3...
1037
1038 - fix several memory leaks
1039 - rename default name of mysql database from "music" to "paraslash".
1040 Use para_server's --mysql_database option if you do not want to
1041 rename your old database.
1042 - rework ogg vorbis code
1043 - make update command work on mysql servers with LOCAL_INFILE
1044 disabled
1045 - gui: improved stream I/O (slots)
1046 - simplified audio format API
1047 - para_pob_ogg is gone
1048
1049 ------------------------------------
1050 0.1.5 (2004-12-31) "opaque eternity"
1051 ------------------------------------
1052
1053 Let's slide gently into the new year.
1054
1055 - new: para_slider (not built automatically, type "make
1056 para_slider" to build). A toy for those who always felt that
1057 creating stream definitions is difficult. See screenshots,
1058 README and FEATURES for more info.
1059 - improved signal handling. Fixes server segfault on SIGHUP
1060 for linux kernels newer than Aug 24 2004 and makes para_gui
1061 race-free.
1062 - reload database tool on SIGHUP
1063 - improved help message for sl
1064 - do not log "broken pipe" messages as errors. They are
1065 perfectly ok.
1066 - fix wrong error message on permission errors
1067
1068 -----------------------------------------
1069 0.1.4 (2004-12-19) "tunneling transition"
1070 -----------------------------------------
1071
1072 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
1073 Hopefully, most of them got wiped out with this release. Some
1074 enhancements went also in.
1075
1076 - improved error diagnostics for all commands
1077 - stradd/picadd: overwrite previous contents if entry already
1078 exists, rather than returning errors
1079 - stradd: use current stream if invoked without args
1080 - faster (and hopefully more stable) ogg-vorbis handling
1081 - para_krell: reap children to avoid zombie-flooding in case
1082 no server is running
1083 - si: report also server pid
1084 - server: don't busy-loop if dbtool reports only invalid files.
1085 - gui: CTRL+C works again, fix stream_read command line option
1086 - fix pic_add, hist
1087 - fix mysql dbtool startup in case no database exists
1088 - many small fixes and cleanups
1089
1090 ---------------------------------------
1091 0.1.3: (2004-12-10) "vanishing inertia"
1092 ---------------------------------------
1093
1094 Starting from this release, the database tools are integrated in the
1095 server binary. This decreases server startup time, reduces code size
1096 and speeds up database commands. However, the layout of the underlying
1097 mysql database changed only slightly and 0.1.3 should be backwards
1098 compatible in that respect.
1099
1100 Visible changes:
1101
1102 - If mysql is not detected at compile time, or fails to init
1103 at runtime, fall back to the dopey database tool which should
1104 always work.
1105 - para_dbtool and dbtool.conf are gone. All mysql specific
1106 options are read from server.conf and are prefixed by 'mysql_'.
1107 - new command: cdt (change database tool)
1108 - new command line option: dbtool (choose startup database tool)
1109 - The name of current stream is now stored in the database,
1110 so paraslash remembers its current stream when restarted.
1111 - new command: csp (change stream and play)
1112 - para_gui also reports current database tool and server uptime
1113
1114 -------------------------------------------
1115 0.1.2: (2004-11-28) "spherical fluctuation"
1116 -------------------------------------------
1117
1118 Point release before the big dbtool changes go in.
1119
1120 - dbtool: rename ca to cam (copy all meta data). It now also
1121 copies numplayed and lastplayed time as well as the picture
1122 id.
1123 - fix endless-loop-bug caused by mp3 files with invalid header
1124
1125 -----------------------------------------
1126 0.1.1: (2004-11-05) "floating atmosphere"
1127 -----------------------------------------
1128
1129 - gkrellm plugin
1130 - new dbtool command: mbox. Browse your sound-file collection
1131 with your favorite mail reader.
1132 - several small fixes
1133
1134 -------------------------------------
1135 0.1.0: (2004-10-22) "rotating cortex"
1136 -------------------------------------
1137
1138 - fix logging bug for loglevel > VERBOSE
1139 - fix skip command
1140 - correct timings for vbr mp3s
1141 - modular audio format support
1142 - ogg-vorbis support (experimental)
1143 - new server option: autoplay
1144
1145 -----------------------------------------
1146 0.0.99: (2004-07-25) "harmonic deviation"
1147 -----------------------------------------
1148
1149 - rename projectname from icc to paraslash (play, archive, rate
1150 and stream large audio sets happily)
1151 - paraslash is no longer restricted to one particular audio
1152 streaming software
1153 - new dbtool commands (stradd, strq, strdel) for easy stream
1154 managment w/o configuration file. That obsoletes stream_defs
1155 file/config option for dbtool.
1156 - picadd accepts jpeg data from stdin
1157 - new server commands: ps (select previous stream), sc (song change)
1158 - new default pictures for sdl_gui
1159 - gui: new key_map option for binding commands and internal
1160 functions to arbitrary keys, nice help screen, rip out
1161 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1162 show silly logo on startup
1163 - new executables: para_fade for fading volume, para_dbadm for
1164 manipulating attributes
1165 - cdb adds _all_ tables to mysql database
1166 - revised and beautified documentation
1167 - sample dbtool rewritten in C
1168 - autoconf
1169
1170 ---------------------------------------------
1171 0.0.98: (2003-12-26) "incremental smoothness"
1172 ---------------------------------------------
1173
1174 - kick icecast in favour of poc. That removes some races and reduces
1175 core code considerably.
1176 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1177 give finer info on afs' status.
1178 - gui can start decoder (see config options). Further new gui
1179 commands: refresh (^L), jmp (F1-F10)
1180 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1181 terminates gui. This fixes dead instances consuming memory
1182 continuously.
1183 - new dbtool command: verb for sending verbatim sql queries.
1184 - fix pid_list races (by removing pid_list)
1185 - codename funnies
1186
1187 --------------------
1188 0.0.97: (2003-10-26)
1189 --------------------
1190
1191 - installation prefix now defaults to /usr/local
1192 - new commands for gui: snozze, sleep and reread config
1193 - config file for gui and sdl_gui
1194 - fix problems with filenames containing funny characters
1195 (reported by Thomas Forell)
1196 - improved signal handling for gui, now it rereads conf on SIGHUP
1197 - new dbtool command: cdb (create database)
1198 - switch from argtable to gengetopt
1199 - major code cleanup and speed improvements
1200 - fix several potential buffer overflows
1201 - many small fixes and cleanups
1202
1203 -------------------
1204 0.0.96 (2003-08-30)
1205 -------------------
1206
1207 - easy stream_defs syntax
1208 - sdl_gui can display images associated to the file being played
1209 - Major feature enhancements for icc_gui including dynamic text
1210 placement and the top/bottom window design
1211 - vrfy/clean now also checks for NULL values in attributes as
1212 well as for invalid picture pointers
1213 - fix long outstanding case sensitivity bug
1214 - many small fixes and cleanups
1215
1216 -------------------
1217 0.0.95 (2003-06-29)
1218 -------------------
1219
1220 - sdl gui runs much faster
1221 - new dbtool command: ca (copy attributes)
1222 - count and display number of times the song has been played
1223 - new feature: scoring
1224 - command line options for sdl_gui
1225 - simpler syntax of streams file
1226 - decrease network traffic of stat
1227 - fix zombie bug
1228 - many small fixes and cleanups
1229
1230 -------------------
1231 0.0.94 (2003-05-04)
1232 -------------------
1233
1234 - new server command: ns (next stream)
1235 - new icc_gui command: c (change stream)
1236 - internal mp3info
1237 - stat shows also id3 tag info
1238 - new sdl based gui
1239 - log flodding bug fixed
1240 - many small fixes and cleanups
1241
1242 -------------------
1243 0.0.93 (2003-03-28)
1244 -------------------
1245
1246 - colors for icc_gui
1247 - icc_gui sets volume directly (linux only)
1248 - proper locking that fixes some races
1249 - fix security bug that caused commands to be executed even
1250 with unsufficient permissions
1251 - new command: hup to make all servers reread their configuration file
1252 - icecast meta data streaming
1253 - many small fixes and cleanups