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