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