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