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