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