a1cdf58eb86e172573f6dc446154961b41f8b0fe
[paraslash.git] / NEWS
1 ---------------------------------------------
2 0.5.0 (to be announced) "invertible validity"
3 ---------------------------------------------
4
5 - The sideband compatibility code has been removed, hence
6 sideband connections (introduced in 0.4.11) are now mandatory.
7 - Addblob commands can produce output.
8 - The stat command no longer sends garbage when para_server was
9 compiled against libgcrypt.
10
11 --------------------------------------
12 0.4.13 (2013-07-29) "spectral gravity"
13 --------------------------------------
14
15 One more 0.4.x release before the API-breaking changes for 0.5.0 go
16 in. The main features of this release are the ogg/opus audio format,
17 and UTF-8 support, but it includes also tons of other improvements
18 and fixes all over the place.
19
20 - New audio format: ogg/opus.
21 - UTF8 support for para_gui and the mp3 audio format handler.
22 - Scheduler improvements and fixes.
23 - The obsolete gettimeofday() function has been replaced
24 by clock_gettime() on systems which support it.
25 - Speed and usability improvements for para_gui.
26 - para_client now restores the fd flags of stdin and stdout
27 on shutdown.
28 - Improved manual pages.
29 - Consistent version strings for all executables.
30 - Reduced dependencies on generated files result in fewer
31 recompilations on changes.
32 - Performance improvements for the compress filter.
33 - Improved downloads web page.
34
35 -----------------------------------------
36 0.4.12 (2012-12-20) "volatile relativity"
37 -----------------------------------------
38 The new command line player, the resample filter, ALSA support for
39 para_fade, and the improved build system are the highlights of this
40 release which probably marks the end of the 0.4.x series.
41
42 - The afh receiver and the para_play executable.
43 - The resample filter: A sample rate converter based on
44 libsamplerate.
45 - The "versions" directory has been removed from the master
46 branch. The tarballs of the old releases are now available
47 in the new "releases" branch.
48 - Overhaul of the build system: All generated files are now
49 written to the "build" directory.
50 - The modular mixer API and the alsa mixer.
51 - Minor fixes for the osx writer.
52
53 --------------------------------------
54 0.4.11 (2012-07-20) "mutual diversity"
55 --------------------------------------
56
57 The major feature in this release is the new sideband API for
58 client-server communication. This API will be used exclusively starting
59 with 0.5.0, which breaks backward compatibility but allows to get rid
60 of quite some compatibility code. Other noteworthy changes include
61 decoder latency improvements and a long-standing bug fix for the
62 ALSA writer.
63
64 - Sideband connections: If both para_server and para_client
65 support this feature, data is sent as a multiplexed stream.
66 - The --no_default_filters option of para_filter has been
67 removed.
68 - Several fixes and latency improvements to various decoders.
69 - The ALSA writer now limits the prebuffer time to 500ms.
70 - Documentation improvements.
71 - Overhaul of the command_util.sh script.
72 - Fixes for some minor problems found by the clang analyzer.
73 - Compiles (almost) without warnings on gcc-3.
74 - Robustness improvements of the buffer tree code.
75
76 ------------------------------------------
77 0.4.10 (2012-03-30) "heterogeneous vacuum"
78 ------------------------------------------
79
80 Nothing earth-shaking in this release, but quite a few usability
81 improvements and the usual mix of cleanups and fixes.
82
83 - The --no_default_filters option of para_filter has been
84 deprecated. It still works but has no effect and will be
85 removed in the next version.
86 - para_gui now prints also the stderr output of the executing
87 command in the bottom window.
88 - Cleanup and consolidation of the various wrappers for
89 write(), writev(), send() and friends.
90 - The obscure error messages on mmap() failures have been
91 replaced by meaningful messages. This affects mainly
92 para_afh.
93 - para_audioc: Cleanups and memory leak fixes.
94 - Test 0004-server no longer fails if para_server is not
95 being built.
96 - New configure options: --with-id3tag-{headers,libs}.
97
98 -------------------------------------
99 0.4.9 (2011-12-06) "hybrid causality"
100 -------------------------------------
101
102 Support for another audio format, interactive mode for para_client
103 and para_audiod and many small improvements/fixes all over the place.
104
105 - Support for flac, the free lossless audio codec.
106 - Fix for an endless loop in the mp3 decoder for certain
107 (corrupt) mp3 files.
108 - When executed without specifying a command, para_client
109 and para_audioc start an interactive shell (requires
110 libreadline being installed). The interactive mode offers
111 full tab completion and command line history.
112 - autogen.sh now detects a distcc setup and adjusts the
113 parameter for the -j option of make accordingly.
114 - Shared memory areas are no longer restricted to 64K. We now
115 detect the maximal size of a shared memory area at runtime.
116 - cleanup of the internal uptime API.
117 - para_server prefaults the mmapped audio file to avoid
118 delays on slow media.
119 - A new test for the test-suite that exercises the
120 communication between para_server and para_audiod.
121 - The alsa writer eats up less CPU cycles when configured to
122 use the DMIX plugin.
123 - Simplified and unified receiver code.
124 - Makefile cleanups.
125 - Commands which print a list of matching audio files now
126 emit a meaningful error message if no audio file matched the
127 given pattern(s).
128
129 --------------------------------------
130 0.4.8 (2011-08-19) "nested assignment"
131 --------------------------------------
132
133 Gcrypt support, the overhauled osx writer and regex format specifiers
134 are the highlights of this release.
135
136 - support for libgcrypt as a drop-in replacement for openssl.
137 Run configure --enable-cryptolib=gcrypt to link against
138 libgcrypt. The two crypto implementations are compatible to
139 each other, i.e. a para_client executable linked against
140 gcrypt can connect to para_server linked against libssl
141 and vice versa.
142 - Rewrite of the osx writer (output plugin for Mac OS).
143 - audiod: The format specifier for receivers, filters and
144 writers is now treated as a regular expression. This allows
145 to replace 5 lines in the config file (one for each audio
146 format) by one single line. See the manual for details.
147 - The *.cmdline.[ch] files are no longer contained in the released
148 tarballs. This reduces the size of the tarballs but requires
149 gengetopt to build the tarball.
150 - Compiles cleanly also with llvm/clang.
151 - Corrupt mp3 files are handled more gracefully.
152 - The alsa writer uses poll fds instead of computing timeouts.
153 - Cleanup of the generic writer API.
154 - sched: Optimized zero timeouts.
155 - vss timeout cleanups.
156 - oggdec fixes and improvements.
157
158 --------------------------------------
159 0.4.7 (2011-06-01) "infinite rollback"
160 --------------------------------------
161
162 The new ao writer, support for ssh RSA keys and a couple of other
163 enhancements.
164
165 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
166 platforms through the libao audio library.
167 - Support for RSA keys generated with ssh-keygen.
168 - configure: improved options for ogg/vorbis/speex.
169 - The git version reported by --version always matches HEAD.
170 - The autogen script detects the number of processors and
171 runs a parallel make if possible.
172 - Major cleanup of the crypto API.
173 - Documentation updates.
174
175 ------------------------------------------
176 0.4.6 (2011-03-31) "deterministic entropy"
177 ------------------------------------------
178
179 Lots of ogg/vorbis improvements, the new test suite, enhancements
180 for para_gui and a fair amount of other bug fixes.
181
182 - For DCCP/OGG streams the audio file header is only sent once
183 at the beginning of the stream rather than periodically
184 every five seconds. This reduces network traffic and the
185 FEC group size.
186 - The vorbis comment header is replaced by an empty dummy header
187 before the header is sent over the network. This also results in
188 less network traffic and smaller FEC groups.
189 - The new "test" make target allows to perform some sanity checks prior
190 to installing the package.
191 - ogg timing fixes and performance improvements
192 - Scheduler improvements
193 - Proper exit codes for para_write
194 - para_gui: New option --theme to select a startup theme. Several
195 other improvements and fixes.
196 - aacdec error message cleanups
197 - simplified color error handling
198
199 --------------------------------------------
200 0.4.5 (2010-12-17) "symmetric randomization"
201 --------------------------------------------
202
203 Bug fixes, internal cleanups and variable-sized FEC slices.
204
205 - Contains a fix for an invalid-free-bug in the ogg audio format
206 handler code.
207 - Switching off the DCCP sender works again.
208 - para_audiod handles crashes of para_server more robustly.
209 - Internal scheduler and writer cleanups.
210 - Reduced latency due to variable-sized FEC slices.
211 - Improved documentation and error diagnostics.
212 - The build of para_server is now optional, allowing the build
213 to succeed in case libosl is not installed.
214
215 ------------------------------------------
216 0.4.4 (2010-08-06) "persistent regularity"
217 ------------------------------------------
218
219 Support for yet another audio format, para_write improvements and
220 bug fixes.
221
222 - Support for the speex codec.
223 - Support for sample formats other than 16 bit little endian.
224 - error2.h is now created by a perl script which speeds up configure
225 considerably.
226 - Fix a bug in the aac decoder which could lead to segfaults in
227 para_filter/para_audiod.
228 - Fixes for autoconf-2.66.
229
230 ----------------------------------------
231 0.4.3 (2010-07-05) "imaginary radiation"
232 ----------------------------------------
233
234 Many improvements for the DCCP and the UDP transport, the new user
235 manual and the usual mix of bug fixes and internal improvements.
236
237 - FEC support for the DCCP sender (Gerrit Renker). The new
238 --dccp_max_slice_size, --dccp_data_slices_per_group and
239 --dccp_slices_per_group options can be used to set the FEC
240 parameters for the DCCP transport.
241 - DNS lookups for UDP targets (Gerrit Renker).
242 - The new user manual replaces the README, README.afs, REQUIREMENTS
243 and INSTALL documents.
244 - Fix an end-of-file detection bug in the oggdec filter.
245 - The new nonblock API.
246 - Both options of the oggdec filter have been removed.
247 - New debug mode for the internal scheduler.
248
249 ------------------------------------------
250 0.4.2 (2010-04-23) "associative expansion"
251 ------------------------------------------
252
253 It's been some time since the last release, but finally here is
254 paraslash-0.4.2. The bulk of the changes comes from the new buffer
255 tree API, but there are changes all over the tree. Mainly performance
256 and usability improvements, but also quite some bug fixes.
257
258 - The new buffer tree API.
259 - DCCP: Support for CCID negotiation (Gerrit Renker).
260 - UDP robustness fixes.
261 - The --bufsize option for mp3dec is gone as it no longer makes sense
262 for the new buffer tree API.
263 - Fix audible buffer underruns for wma streams.
264 - The alsa writer no longer prints meaningless underrun durations.
265 - audiod: Defaults work also for udp streams. If no filter is
266 given for an audio format that is received via upd, fecdec is
267 automatically added as the first filter (along with the decoder).
268
269 ---------------------------------------
270 0.4.1 (2009-12-22) "concurrent horizon"
271 ---------------------------------------
272
273 Support for another audio format, minor feature enhancements and lots of bug
274 fixes. All fixes that have been accumulated in the maint branch (in particular
275 those mentionened in the 0.3.6 release notes) appear in this release as well.
276
277 - wma support.
278 - new afh option: --human to activate human-readable output.
279 - new server/audiod option: --log-timing to print timing information.
280 - build system improvements.
281 - source code documentation updates.
282
283 -------------------------------------
284 0.3.6 (2009-12-07) "cubic continuity"
285 -------------------------------------
286
287 Quite a few bugs have been found and fixed since 0.3.5, so here's
288 another 0.3.x release. No new features.
289
290 - Always check return value of malloc().
291 - ogg vorbis/FEC: Do not write garbage after the audio file header.
292 - exit if root privileges could not be dropped.
293 - FEC: Fix computation of extra slices.
294 - oss: Fix check for empty input buffer.
295 - Avoid buffer underruns due to filter chain output buffer constraints.
296 - server: Fix assignment of afs_pid.
297 - Don't panic if the afs database contains unknown audio formats.
298 - http/dccp: Do not send the audio file header twice.
299 - FEC: Timing improvements.
300
301 ----------------------------------------------
302 0.4.0 (2009-11-10) "simultaneous independence"
303 ----------------------------------------------
304
305 Two significant changes which require the new version number: The
306 improved authentication dialog and the fact that the database code
307 has been moved to a library, libosl. To use the new version, you have
308 to generate new RSA keys, see INSTALL for details. A shell script is
309 provided for conversion of the 0.3 database to the new 0.4 format.
310
311 - stronger crypto for client authentication
312 - the database code has been moved to a library
313 - improved status item handling
314 - cleanup of the build system
315 - The "-V" option now also prints the git version
316 - the new parser-friendly listing mode for the ls and stat commands
317 - mandatory rc4 encryption
318 - major audio format handler cleanups
319 - (id3,...) tags are no longer stored as a combined string in the database
320 - new mood methods: artist_matches, title_matches, comment_matches,
321 album_matches, year_maches, year.
322
323 --------------------------------------------
324 0.3.5 (2009-09-21) "symplectic separability"
325 --------------------------------------------
326
327 Full client support for *BSD Unixes, complete re-write of the ogg
328 vorbis audio format handler, various improvements all over the place
329 and the usual mix of bugfixes. This release marks the end of the 0.3
330 series if no serious problems show up.
331
332 - the new oss writer (supported on *BSD and Linux)
333 - rewrite of the ogg vorbis audio format handler. It's
334 recommended to replace the chunk tables of existing ogg
335 vorbis files in the afs database by re-adding these files
336 with "add -f".
337 - support for netmask subsets (Gerrit Renker)
338 - the new prebuffer filter
339 - improved signal handling
340 - variable fec output buffer size
341 - improved FEC timing fixes audible buffer underruns in UDP mode
342 - --log_color actually works
343 - new ls option: -d (print dates as seconds after the epoch)
344 - update to gengetopt 2.22.2
345 - support for RSA keys of size > 512 bits
346 - new option "mixer_channel" for para_fade
347
348 -----------------------------------------
349 0.3.4 (2009-05-07) "elliptic inheritance"
350 -----------------------------------------
351
352 The new udp sender, forward error correction, colored logs and various
353 other improvements. As the udp sender does not depend on any special
354 libraries, it is built unconditionally.
355
356 - The udp sender replaces the ortp sender. The new code uses forward
357 error correction to protect against packet losses. Many thanks to
358 Gerrit Renker for providing ipv6 support.
359 - The default port for udp streaming now defaults to 8000, like
360 for the http and the dccp senders/receivers.
361 - Loglevels are now specified as symbolic names, e.g.
362 "--loglevel info".
363 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
364 - para_server/para_audiod: Color support for log messages.
365 - new options for mp3dec: --ignore-crc, --bufsize
366 - new audiod option: --config-file.
367 - gengetopt cleanups.
368 - Improved help/man pages: The documentation of para_audiod,
369 para_recv, para_filter and para_write now also contains
370 all options of the available receivers/filters/writers. The
371 man page of para_fade contains a description of the different
372 modes of operation.
373 - More source code documentation.
374 - vss timing fixes.
375
376 --------------------------------------------
377 0.3.3 (2008-12-01) "axiomatic perspectivity"
378 --------------------------------------------
379
380 Internal code cleanups, bug fixes, improved tag handling and the new
381 amplification filter.
382
383 - para_server uses the generic scheduling code.
384 - overhaul of the virtual streaming system.
385 - mp3: id3 version 2 support via libid3tag (optional)
386 - ogg: vorbis comment support.
387 - aac meta info support.
388 - mp3 audio format handler cleanups.
389 - new filter: "amp" to amplify the amplitude of the audio stream
390 - new status item/database entry: amplification. It is
391 used by the amp filter to pre-amplify the audio stream.
392 - fix a close-without-open bug in para_write.
393 - fix a bug in com_init() which was introduced in 0.3.2.
394 - better error diagnostics for para_client.
395
396 -----------------------------------------
397 0.3.2 (2008-04-11) "probabilistic parity"
398 -----------------------------------------
399
400 The new para_afh executable, scheduling and documentation improvements.
401
402 - new ls option: -lc (list chunk table)
403 - new executable: para_afh, the stand-alone audio file handler tool
404 - afs commands can send output more than SHMMAX (32MB on Linux). This
405 also reduces the memory usage of commands that produce large amounts
406 of output.
407 - major scheduler and audiod cleanups.
408 - more detailed and much nicer man pages.
409
410 ---------------------------------------
411 0.3.1 (2008-02-23) "liquid interaction"
412 ---------------------------------------
413
414 A mix of cleanups, bug fixes, improvements, and some new features. No
415 significant changes to the new database (osl) code, which is generally
416 a good sign.
417
418 - Share some similar/duplicate code between the http and the
419 dccp sender.
420 - Generic access control lists for paraslash senders.
421 - dccp sender: Access control lists, connection limiting and support
422 for the allow,deny,on,off,help sender commands.
423 - The default dccp port changed from 5001 to 8000 (suggested by
424 Gerrit Renker).
425 - para_server starts even if not all public keys could be loaded.
426 - Audiod performance improvements.
427 - fix a bug in the "off" command of the http sender.
428 - fix some fd and memory leaks.
429 - Update to gengetopt-2.22.
430
431 -------------------------------------
432 0.3.0 (2008-01-12) "solar saturation"
433 -------------------------------------
434
435 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
436
437
438 Usually one might expect lots of new features AND a big increase in size
439 for a major release like this.
440
441 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
442 ever. The decrease in size is mostly due to the removal of some
443 graphical tools (which were only quick hacks anyway). But also the
444 fact that the mysql code is gone cuts down the size a bit.
445
446 Being independent of mysql comes at a cost: The fact that paraslash
447 now contains its own database (the object storage layer, osl) increases
448 the (stripped) binary size of para_server by ~50K on i386.
449
450 - no more restrictions on unique basenames.
451 - independent of mysql: The new self-contained object
452 storage layer (osl) replaces the mysql database.
453 - New executable para_fsck: Check integrity of osl tables.
454 - Lyrics support.
455 - Reliable audio file move/rename detection.
456 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
457 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
458 Solaris (sparc64).
459 - the new osl-based audio file selector (afs) replaces the random,
460 the playlist and the mysql selector of paraslash-0.2.x.
461 - IPv6 support (thanks to Gerrit Renker).
462 - paraslash-0.2.x streams are now called "moods". Writing
463 0.3.x-mood definitions should be both easier and more
464 powerful than writing 0.2.x-stream definitions.
465 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
466 been removed. The world is a better place without them. However,
467 para_gui is still there.
468 - afs tracks audio file selection also in playlist mode.
469 - few easy-to-use afs commands replace the many not-so-easy-to-use
470 mysql commands (and are available also in playlist mode).
471 - Improved error subsystem.
472 - The earth-shaking new logo.
473
474 -----------------------------------------
475 0.2.17 (2007-11-20) "isotropic threshold"
476 -----------------------------------------
477
478 Mainly bugfixes and cleanups in this version which marks the end of
479 the 0.2.x series if no serious bugs show up after the release.
480
481 - mysql_selector: fix a locking bug.
482 - universal chunk queueing.
483 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
484 to Gerrit Renker).
485 - be more carful wrt. signed vs. unsigned argument passing.
486 - cleanup error.h and fix some references to invalid error
487 codes.
488 - update to gengetopt-2.21.
489 - update to ortp-0.13.1.
490 - autoconf: extend checks for headers, library functions and
491 compiler characteristics.
492 - Fix streaming of large mp3 files.
493 - Fix an off-by-one bug in playlist handling.
494
495 --------------------------------------
496 0.2.16 (2007-04-05) "neural discharge"
497 --------------------------------------
498
499 The main change in this release is the major audio format handler
500 cleanup which removes some similar/duplicate code and makes it easier
501 to implement plugins for other audio formats. Of course, the usual mix
502 of other improvements/changes/bugfixes also made it into the release.
503
504 - simplified audio format handlers (most of the handling functions
505 were moved one layer up to the virtual streaming system).
506 - para_server uses mmap to read audio files
507 - repositioning of mp3 streams is much faster, in particular for
508 jumping near the end of large mp3 files.
509 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
510 and AFS_WRITE.
511 - fix a bug in para_filter that caused decoding of aac files
512 to start only after a few seconds.
513 - fix osx_writer hangs
514 - simplified dccp code (thanks to Gerrit Renker)
515 - the compress filter works also on big endian systems (ppc)
516
517 -----------------------------------------
518 0.2.15 (2007-02-16) "inductive resonance"
519 -----------------------------------------
520
521 Minor improvements, more documentation and a bunch of bug fixes.
522
523 - para_server: The server.users file is only read once on server
524 startup rather than for each connection
525 - mp3dec: Fix decoding of corrupt mp3 files
526 - afs (audio file sender) is now called vss (virtual streaming
527 system). Consequently, the permission flags specified in
528 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
529 become VSS_READ and VSS_WRITE respectively.
530 - para_audiod/para_filter: Fix a bug that caused the last chunk
531 of audio data not being written under certain circumstances
532 - audiod: compute the difference of server time and local time
533 correctly
534 - para_server/para_audiod: Fix some memory leaks
535 - documentation improvements
536 - configure.ac: fix checks for para_krell
537 - new man pages
538
539 -------------------------------------------
540 0.2.14 (2006-10-15) "transient singularity"
541 -------------------------------------------
542
543 The only major enhancement of this version is the osx writer which completes
544 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
545 until now for reasons beyond the scope of this changelog entry.
546
547 - new output plugin for Mac Os: the osx writer
548 - rename configure command line options from --enable-xxx-headers to
549 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
550 - configure: new command line options: --with-mad-headers,
551 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
552 - some robustness fixes
553 - dymamic audio format recognition for audiod
554 - para_server: new command line option: --autoplay_delay
555 - para_audiod: new command line option: --clock_diff_count
556
557 ---------------------------------------
558 0.2.13 (2006-07-14) "sonic convolution"
559 ---------------------------------------
560
561 A bunch of new features and core changes.
562
563 - the new paraslash scheduler, short and sweet.
564 - Support for m4a/mp4 files via the new aac audio format
565 handler/filter (requires libfaad).
566 - each writer has its own command line parser, just like
567 para_recv and para_filter.
568 - para_client and para_audioc use the error subsystem
569 - writers are integrated in para_audiod (currently linux-only)
570 - para_client is integrated in para_audiod
571 - random/playlist selector: improved info strings
572 - new audiod commands: tasks, kill
573 - update to libortp-0.10.1
574 - para_fade: wake time defaults to 8 hours from now
575 - update to autoconf-2.60
576
577 ------------------------------------------
578 0.2.12 (2006-05-12) "oriented abstraction"
579 ------------------------------------------
580
581 Many user-visible changes in this release and lots of new
582 features:
583
584
585 - the new optional dccp sender/receiver. It uses the datagram
586 congestion control protocol. You'll need a fairly new kernel
587 for this.
588 - paraslash works on Mac OS X (thanks to Gerd Becker)
589 - para_play renamed to para_write
590 - modular output plugin design (writers) for para_write
591 - new file_writer output plugin for para_write
592 - compress filter speed improvements
593 - update to libortp-0.9.1
594 - update to gengetopt-2.17rc
595 - para_client no longer depends on libreadline (as the
596 code for the interactive mode was removed).
597 - gcc-2-95 is no longer a supported compiler. It may still
598 work, but it gets no more testing.
599 - the tarball no longer contains the screenshot images which
600 reduces its size quite a bit.
601 - configure: new command line options: --enable-mysql-headers
602 and --enable-mysql-libs
603
604 ------------------------------------
605 0.2.11 (2006-03-11) "atomic duality"
606 ------------------------------------
607
608 Here it is, the first paraslash release developed with git. There
609 are fairly many user-visible changes in this release. As two out of
610 the three "database tools" of paraslash don't use a database at all,
611 they are now called "audio file selectors" instead.
612
613
614 - the cdt command (change database tool) becomes chs (change
615 selector)
616 - no more colon separators: The syntax of some options of
617 para_audiod and para_filter have changed. Use --help for
618 more info (and some examples).
619 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
620 - switch from cvs to git (should've done that earlier)
621 - the new ipc subsystem
622 - new audio file selector: playlist
623 - para_server: the dopey selector is now called "random",
624 and is the default selector. Use the --selector option to
625 choose another selector at startup, or the chs command to
626 change the selector at runtime.
627 - X86_64 fixes (thanks to Steffen Klassert)
628 - para_play fixes
629
630 --------------------------------------
631 0.2.10 (2006-02-17) "cyclic attractor"
632 --------------------------------------
633
634 Huge documentation update, a scrollable window for para_gui, ortp
635 improvements, and of course many small fixes not mentioned here.
636 The diffstat below is rather misleading as most insertions are due
637 to the new source documentation.
638
639 - autoconf cleanup
640 - para_server also uses the new error subsystem
641 - lots of new documentation (UTSL)
642 - gui improvements:
643 - keysyms for cursor keys and for next/previous page keys
644 - scrollable output window
645 - new internal commands: scroll up/down, page up/down
646 - fix color of command output.
647 - ortp: the --chunk_time and --header flags are no longer needed
648 for para_recv/para_audiod as this information is now encoded in
649 each rtp packet sent by para_server.
650
651 -------------------------------------------
652 0.2.9 (2006-01-24) "progressive turbulence"
653 -------------------------------------------
654
655 Internal audiod receivers/filters, the new error subsystem and
656 a lot of small improvements.
657
658 - para_recv and para_filter are integrated into the para_audiod
659 binary, i.e. audiod no longer spawns a new process for
660 each receiver/filter. As para_recv and para_filter might be
661 useful as standalone programs, they still get built (linked
662 against the same object files that are also used for audiod).
663 - further ortp timing improvements should reduce the CPU usage
664 of the ortp receiver.
665 - improved audio grabbing. The 'grab' command of para_audiod
666 has its own set of command line options. Read the output of
667 "para_audioc help grab" for more info.
668 - oggdec: configurable input prebuffer size.
669 - the new error subsystem gives better error diagnostics
670 and reduces code size.
671
672 -----------------------------------------
673 0.2.8 (2006-01-02) "dynamic accumulation"
674 -----------------------------------------
675
676 The new modular filter design and the para_play-hangs bugfix.
677
678 - new executable: para_filter. It combines para_mp3dec,
679 para_oggdec and para_compress. It also adds a further filter
680 type, wav, that just inserts a wave header at the desired point
681 of the filter chain. All 'piping' is done in-memory (i.e. no
682 read/write operations are used).
683 - para_play: fix a stupid bug that caused it to hang under
684 certain circumstances.
685
686 -------------------------------------------
687 0.2.7 (2006-12-27) "transparent invariance"
688 -------------------------------------------
689
690 Not many user-visible changes but a fair amount of internal improvements.
691
692
693 - The http sender buffers data if it can not be sent
694 out immediately (because the socket is not writable). This
695 should prevent para_server from shutting down the connection
696 too early on a loaded network.
697 - para_play also prebuffers data if it is told to start at a
698 future time by the --start_time option.
699 - The return of para_recv: It combines para_ortp_recv and
700 para_http_recv. Use the --receiver option to switch between
701 the two. para_recv builds without libortp, but contains
702 only the http receiver in this case.
703 - update to ortp 0.8.1. As this ortp release contains incompatible
704 changes, para_recv-0.2.7 won't link against older ortp libs.
705 - improved ortp timings.
706 - use of gcc-extensions that #define away for non-gcc and
707 gcc < 3.0.
708
709 -------------------------------------------
710 0.2.6 (2005-10-29) "recursive compensation"
711 -------------------------------------------
712
713 Transparent session encryption (uses openssl's Alleged RC4 cipher),
714 the internal find command and several other improvements and cleanups.
715
716 - Encrypt the session if encryption is requested by the client
717 (default for para_client 0.2.6). This is backwards
718 compatible, so older clients can still connect to para_server
719 0.2.6. Use the new client option --plain to request an
720 uncrypted session (off by default, must be set to on in
721 order to connect to para_server 0.2.x with 0 <= x <= 5).
722 - para_server uses an internal function to locate audio files
723 rather than calling find(1). The server option
724 --mysql_audio_file_dir replaces --mysql_find_cmd.
725 - documentation update
726 - man pages
727 - header file cleanup
728 - para_client code cleanup
729 - para_gui: faster display of output of display commands
730
731 ------------------------------------------
732 0.2.5 (2005-10-13) "aggressive resolution"
733 ------------------------------------------
734
735 This release adds internal senders, i.e. no more external programs are
736 spawned for sending out the audio data. There are two different senders
737 available: The http sender and the ortp sender (former para_send which
738 is no longer needed).
739
740 The new sender code has a plugin-like design so it can be easily
741 extended should there be be any future need for supporting another
742 network streaming protocol. All senders are completely independent of
743 each other. In particular, the http and the ortp sender can operate
744 in parallel.
745
746 - new server command: sender to control senders at runtime.
747 Read the output of "para_server -h" and "para_client help
748 sender" for more information.
749 - para_recv renamed to para_ortp_recv
750 - new executable: para_http_recv, a simple command line
751 http receiver.
752 - major afs/mp3/ogg code simplifications due to internal
753 senders.
754 - ogg timing improvements
755 - fix several minor memory leaks (found by valgrind)
756 - empty stream definitions work again
757 - com_ne(): ignore errors on remove
758 - audiod: fix segfault on server restart
759
760 ---------------------------------------
761 0.2.4 (2005-09-21) "toxic anticipation"
762 ---------------------------------------
763
764 Several small improvements, fixes and the new grab command.
765
766 - audiod:
767 - new command: "grab" to grab the output of the stream reader
768 or any filters. Read the output of "para_audioc help grab"
769 for more information.
770 - fix memory leak
771 - code cleanup
772 - audioc: new command line option: --bufsize to specify a
773 buffer size different from the default size 8192.
774 - improved error diagnostics for para_play.
775 - new configure option: --enable-ssldir so search for openssl in
776 non-standard places
777 - sdl_gui: Make it look nice again for 1024x768
778 - server: report total size of memory allocated with sbrk by malloc,
779 new command line option: --announce_time
780
781 -----------------------------------------
782 0.2.3 (2005-09-01) "hydrophilic movement"
783 -----------------------------------------
784
785 Two new executables and major feature enhancements.
786
787 - audiod filters: It is now possible to specify arbitrary many
788 (including none) filters for each supported audio
789 format. This can be used e.g. for normalizing volume,
790 transforming or grabbing the audio stream, or for using
791 visualizers. Read the output of "para_audiod -h" for the
792 syntax of the new --filter_cmd option.
793 - new executable: para_play, a tiny alsa player. It
794 can play wave files or raw pcm (16 bit little endian)
795 from stdin.
796 - new executable: para_compress, a dynamic range compressor
797 intended to keep audio output at a consistent volume. Derived
798 from AudioCompress, http://trikuare.cx/code/AudioCompress.html.
799 - audiod: New option: --stream_delay. This can be used in
800 a local network to syncronize the audio output of all
801 clients that play the same stream.
802
803 ------------------------------------------
804 0.2.2 (2005-08-19) "tangential excitation"
805 ------------------------------------------
806
807 Mostly internal changes in this release, but also some new commands
808 for the mysql database tool.
809
810 - cleanup exec.c, fix para_exec bug
811 - compile time loglevel (log messages below the given level
812 won't be compiled in, which reduces the size of the
813 resulting binaries)
814 - new log macros that shorten the size of the source code.
815 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
816 to send only zeros. With this workaround, para_audioc works
817 again.
818 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
819 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
820 (new entry), "snp" (set numplayed). Read the manual for more
821 information.
822
823 ---------------------------------------
824 0.2.1 (2005-08-15) "surreal experience"
825 ---------------------------------------
826
827 Here comes paraslash-0.2.1. It contains a couple of new features and,
828 surprise, only minor bug fixes.
829
830 - kill noisy mp3 debug message
831 - cleanup of the build system
832 - para_server and para_client directly use the crypto routines
833 of the openssl library rather than invoking the openssl command
834 line utitlity
835 - server/audiod: new option --user to switch to the given user
836 when invoked as root. Read the output of "para_server -h" for
837 more information.
838 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
839 status. Default: "para_audioc stat"
840 - sdl_gui: new option --pic_cmd to be used to download the picture.
841 Default: "para_client pic"
842 - audiod: 5 slots ought to be enough for everybody
843 - audiod: new status item: Uptime, kill hup command
844
845 ------------------------------------------
846 0.2.0 (2005-08-06) "distributed diffusion"
847 ------------------------------------------
848
849 After several month of increased development activity, paraslash-0.2.0
850 has arrived. It contains many new features and is much more
851 self-contained than the old 0.1.x series. Enjoy!
852
853
854 - para_server: fix hang on song change and crash on sighup.
855 Speed up mysql queries. The DIR_LIKE macro is gone.
856 - new executables: para_audiod, the local audio daemon that
857 starts playback (uses SCM_CREDENTIALS socket magic) and
858 para_audioc, the corresponding client.
859 - new executables: para_mp3dec/para_oggdec, two really teensy
860 decoders. para_mp3dec is based on libmad, para_oggdec requires
861 libvorbisfile.
862 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
863 they are now called para_send and para_recv respectively.
864 - documentation updates
865 - para_gui is themable. For now there is the default theme that
866 looks as before and the simple theme: blue and easy.
867 - gui: audio streaming is now handled by audiod. Time display shows
868 playback time rather than streaming time
869 - slider: update to libzmw-0.2.0
870 - para_krell: fix crash on server shutdown
871 - switch from gzip to bzip2
872
873 ----------------------------------------
874 0.1.7 (2005-04-18) "melting penetration"
875 ----------------------------------------
876
877 The main change in this release is clearly the oggvorbis rewrite,
878 but there are also lots of smaller changes. If you intend to use both
879 the mp3 and the ogg plugin, it is recommended to use software mixing,
880 e.g. the dmix plugin which is provided by ALSA.
881
882 - new executables: para_ovsend and para_ovrecv for sending/receiving
883 oggvorbis files via rtp. Requires the open rtp library. Get it at
884 http://www.linphone.org/ortp/
885 - rewrite of the ogg_vorbis core code
886 - configure detects libzmw and, if detected, includes
887 para_slider to the list of binaries to be built by make
888 - server stream writers read from their associated fifo rather
889 than from stdin
890 - slider: two new sliders, lastplayed and numplayed
891 - fix nasty double free bug which caused random segfaults in case of
892 mp3 files with invalid header information
893 - gui: new command line option: --stream_timeout=seconds to
894 deactivate a slot if it is idle for that many seconds (default=`5')
895 - diffstats
896
897 ---------------------------------------
898 0.1.6 (2005-03-05) "asymptotic balance"
899 ---------------------------------------
900
901 Only little user-visible changes in this release. Mainly bugfixes and
902 core code cleanup. This is probably the most stable version ever if you
903 stick to mp3...
904
905 - fix several memory leaks
906 - rename default name of mysql database from "music" to "paraslash".
907 Use para_server's --mysql_database option if you do not want to
908 rename your old database.
909 - rework ogg vorbis code
910 - make update command work on mysql servers with LOCAL_INFILE
911 disabled
912 - gui: improved stream I/O (slots)
913 - simplified audio format API
914 - para_pob_ogg is gone
915
916 ------------------------------------
917 0.1.5 (2004-12-31) "opaque eternity"
918 ------------------------------------
919
920 Let's slide gently into the new year.
921
922 - new: para_slider (not built automatically, type "make
923 para_slider" to build). A toy for those who always felt that
924 creating stream definitions is difficult. See screenshots,
925 README and FEATURES for more info.
926 - improved signal handling. Fixes server segfault on SIGHUP
927 for linux kernels newer than Aug 24 2004 and makes para_gui
928 race-free.
929 - reload database tool on SIGHUP
930 - improved help message for sl
931 - do not log "broken pipe" messages as errors. They are
932 perfectly ok.
933 - fix wrong error message on permission errors
934
935 -----------------------------------------
936 0.1.4 (2004-12-19) "tunneling transition"
937 -----------------------------------------
938
939 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
940 Hopefully, most of them got wiped out with this release. Some
941 enhancements went also in.
942
943 - improved error diagnostics for all commands
944 - stradd/picadd: overwrite previous contents if entry already
945 exists, rather than returning errors
946 - stradd: use current stream if invoked without args
947 - faster (and hopefully more stable) ogg-vorbis handling
948 - para_krell: reap children to avoid zombie-flooding in case
949 no server is running
950 - si: report also server pid
951 - server: don't busy-loop if dbtool reports only invalid files.
952 - gui: CTRL+C works again, fix stream_read command line option
953 - fix pic_add, hist
954 - fix mysql dbtool startup in case no database exists
955 - many small fixes and cleanups
956
957 ---------------------------------------
958 0.1.3: (2004-12-10) "vanishing inertia"
959 ---------------------------------------
960
961 Starting from this release, the database tools are integrated in the
962 server binary. This decreases server startup time, reduces code size
963 and speeds up database commands. However, the layout of the underlying
964 mysql database changed only slightly and 0.1.3 should be backwards
965 compatible in that respect.
966
967 Visible changes:
968
969 - If mysql is not detected at compile time, or fails to init
970 at runtime, fall back to the dopey database tool which should
971 always work.
972 - para_dbtool and dbtool.conf are gone. All mysql specific
973 options are read from server.conf and are prefixed by 'mysql_'.
974 - new command: cdt (change database tool)
975 - new command line option: dbtool (choose startup database tool)
976 - The name of current stream is now stored in the database,
977 so paraslash remembers its current stream when restarted.
978 - new command: csp (change stream and play)
979 - para_gui also reports current database tool and server uptime
980
981 -------------------------------------------
982 0.1.2: (2004-11-28) "spherical fluctuation"
983 -------------------------------------------
984
985 Point release before the big dbtool changes go in.
986
987 - dbtool: rename ca to cam (copy all meta data). It now also
988 copies numplayed and lastplayed time as well as the picture
989 id.
990 - fix endless-loop-bug caused by mp3 files with invalid header
991
992 -----------------------------------------
993 0.1.1: (2004-11-05) "floating atmosphere"
994 -----------------------------------------
995
996 - gkrellm plugin
997 - new dbtool command: mbox. Browse your sound-file collection
998 with your favorite mail reader.
999 - several small fixes
1000
1001 -------------------------------------
1002 0.1.0: (2004-10-22) "rotating cortex"
1003 -------------------------------------
1004
1005 - fix logging bug for loglevel > VERBOSE
1006 - fix skip command
1007 - correct timings for vbr mp3s
1008 - modular audio format support
1009 - ogg-vorbis support (experimental)
1010 - new server option: autoplay
1011
1012 -----------------------------------------
1013 0.0.99: (2004-07-25) "harmonic deviation"
1014 -----------------------------------------
1015
1016 - rename projectname from icc to paraslash (play, archive, rate
1017 and stream large audio sets happily)
1018 - paraslash is no longer restricted to one particular audio
1019 streaming software
1020 - new dbtool commands (stradd, strq, strdel) for easy stream
1021 managment w/o configuration file. That obsoletes stream_defs
1022 file/config option for dbtool.
1023 - picadd accepts jpeg data from stdin
1024 - new server commands: ps (select previous stream), sc (song change)
1025 - new default pictures for sdl_gui
1026 - gui: new key_map option for binding commands and internal
1027 functions to arbitrary keys, nice help screen, rip out
1028 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1029 show silly logo on startup
1030 - new executables: para_fade for fading volume, para_dbadm for
1031 manipulating attributes
1032 - cdb adds _all_ tables to mysql database
1033 - revised and beautified documentation
1034 - sample dbtool rewritten in C
1035 - autoconf
1036
1037 ---------------------------------------------
1038 0.0.98: (2003-12-26) "incremental smoothness"
1039 ---------------------------------------------
1040
1041 - kick icecast in favour of poc. That removes some races and reduces
1042 core code considerably.
1043 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1044 give finer info on afs' status.
1045 - gui can start decoder (see config options). Further new gui
1046 commands: refresh (^L), jmp (F1-F10)
1047 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1048 terminates gui. This fixes dead instances consuming memory
1049 continuously.
1050 - new dbtool command: verb for sending verbatim sql queries.
1051 - fix pid_list races (by removing pid_list)
1052 - codename funnies
1053
1054 --------------------
1055 0.0.97: (2003-10-26)
1056 --------------------
1057
1058 - installation prefix now defaults to /usr/local
1059 - new commands for gui: snozze, sleep and reread config
1060 - config file for gui and sdl_gui
1061 - fix problems with filenames containing funny characters
1062 (reported by Thomas Forell)
1063 - improved signal handling for gui, now it rereads conf on SIGHUP
1064 - new dbtool command: cdb (create database)
1065 - switch from argtable to gengetopt
1066 - major code cleanup and speed improvements
1067 - fix several potential buffer overflows
1068 - many small fixes and cleanups
1069
1070 -------------------
1071 0.0.96 (2003-08-30)
1072 -------------------
1073
1074 - easy stream_defs syntax
1075 - sdl_gui can display images associated to the file being played
1076 - Major feature enhancements for icc_gui including dynamic text
1077 placement and the top/bottom window design
1078 - vrfy/clean now also checks for NULL values in attributes as
1079 well as for invalid picture pointers
1080 - fix long outstanding case sensitivity bug
1081 - many small fixes and cleanups
1082
1083 -------------------
1084 0.0.95 (2003-06-29)
1085 -------------------
1086
1087 - sdl gui runs much faster
1088 - new dbtool command: ca (copy attributes)
1089 - count and display number of times the song has been played
1090 - new feature: scoring
1091 - command line options for sdl_gui
1092 - simpler syntax of streams file
1093 - decrease network traffic of stat
1094 - fix zombie bug
1095 - many small fixes and cleanups
1096
1097 -------------------
1098 0.0.94 (2003-05-04)
1099 -------------------
1100
1101 - new server command: ns (next stream)
1102 - new icc_gui command: c (change stream)
1103 - internal mp3info
1104 - stat shows also id3 tag info
1105 - new sdl based gui
1106 - log flodding bug fixed
1107 - many small fixes and cleanups
1108
1109 -------------------
1110 0.0.93 (2003-03-28)
1111 -------------------
1112
1113 - colors for icc_gui
1114 - icc_gui sets volume directly (linux only)
1115 - proper locking that fixes some races
1116 - fix security bug that caused commands to be executed even
1117 with unsufficient permissions
1118 - new command: hup to make all servers reread their configuration file
1119 - icecast meta data streaming
1120 - many small fixes and cleanups