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