Merge branch 'refs/heads/t/ll'
[paraslash.git] / NEWS.md
1 NEWS
2 ====
3
4 -------------------------------------------
5 0.7.1 (to be announced) "digital spindrift"
6 -------------------------------------------
7
8 - The autogen.sh script now only creates the autoconf specific files
9 but no longer runs configure, make and the test suite.
10 - A stripped down copy of the discontinued libmp4ff library has become
11 part of the paraslash code base. As a result it is no longer necessary
12 to install faad from source to get support for aac/m4a files. The
13 faad decoder package must still be installed.
14 - The log level of the running daemon can now be changed with the
15 new ll command. It is available for para_server and para_audiod.
16 - All calls to select(2) have been replaced by calls to poll(2)
17 to avoid known shortcomings of the select API.
18
19 [tarball](./releases/paraslash-git.tar.xz)
20
21 ----------------------------------
22 0.7.0 (2022-03-12) "seismic orbit"
23 ----------------------------------
24
25 The major incompatible change which requires to bump the major version
26 is the switch from sha1 to sha256, see below for details. However,
27 there are many other improvements, the usual amount of bug fixes and
28 a couple of new features.
29
30 - Starting with paraslash-0.7.0, the sha256 hash value of each known
31 audio file is stored in the database while older versions employed the
32 sha1 hash algorithm which has been considered insecure since 2005
33 and should no longer be used today. The switch from sha1 to sha256
34 requires users to upgrade their database using the new para_upgrade_db
35 program, followed by re-adding all files to recompute the hashes. With
36 this approach all metadata stored in the database (last played date,
37 num played value, moods, playlists, attributes etc.) are kept. An
38 simpler alternative is to start over from scratch by running the
39 "init" command but this will lose these metadata.
40 - Server and client now hash the session keys with sha256 rather
41 than sha1 during the initial handshake. This feature is optional and
42 backwards compatible: old clients can still connect to a new server
43 (using sha1). Also new clients can connect to an old server (again
44 using sha1).
45 - The new "duration" keyword of the mood grammar makes it possible to
46 impose a constraint on the duration of the admissible files.
47 - The long deprecated version 1 mood syntax is no longer supported.
48 - Paraslash writers handle early end-of-file more gracefully.
49 - The alsa writer no longer warns about spurious underruns.
50 - The score formula of the audio file selector has been reworked.
51 - Cleanups of the doubly linked lists code.
52 - New option for configure: --enable-ubsan to detect and report undefined
53 behaviour.
54 - The "tasks" server command has been removed.
55 - The fancy new logo and a minor overhaul of the web pages.
56
57 Downloads:
58 [tarball](./releases/paraslash-0.7.0.tar.xz),
59 [signature](./releases/paraslash-0.7.0.tar.xz.asc)
60
61 --------------------------------------
62 0.6.4 (2021-11-04) "fuzzy calibration"
63 --------------------------------------
64
65 This point release contains a fair number of fixes but no new features.
66 This marks the end of the 0.6 development, although paraslash-0.6 will
67 still be supported for some time and subsequent maintenance releases
68 may follow.
69
70 - The udp sender no longer crashes when empty chunks are encountered.
71 - Fix a double-free bug in the exit path of the server.
72 - The "jmp" command now errors out when given a negative percentage.
73 - A fix for a bug in para_afh which triggered on the attempt to modify
74 the tags of an invalid mp4 file.
75 - A memory leak in para_afh has been fixed.
76 - The udp sender no longer sends multiple EOF packets.
77 - Some gcc warnings have been silenced.
78 - Minor log level adjustments and documentation improvements.
79
80 Downloads:
81 [tarball](./releases/paraslash-0.6.4.tar.xz),
82 [signature](./releases/paraslash-0.6.4.tar.xz.asc)
83
84 ---------------------------------------
85 0.5.9 (2021-11-04) "reversed dimension"
86 ---------------------------------------
87
88 This release contains a few important fixes which have accumulated in
89 the maint branch. The paraslash-0.5.x series has now reached its end
90 of life and will no longer be supported. All users should upgrade to
91 a more recent version at this point.
92
93 - Fix an issue with the bash completion script.
94 - Initialize the random seed also when using libgrypt.
95 - Fix some compiler warnings in the resample filter
96 - Don't return spurious errors from the ff server command.
97
98 Downloads:
99 [tarball](./releases/paraslash-0.5.9.tar.bz2),
100 [signature](./releases/paraslash-0.5.9.tar.bz2.asc)
101
102 ----------------------------------------
103 0.6.3 (2021-02-18) "generalized activity"
104 -----------------------------------------
105
106 It has been a while since the last release, but paraslash-0.6.3
107 finally has arrived. It contains a fair number of improvements,
108 features and fixes all over the place.
109
110 - The ff command now accepts a negative argument to instruct the
111 virtual streaming system to jump backwards in the current audio
112 stream. The old syntax (e.g., "ff 30-") is still supported but it
113 is deprecated and no longer documented. The compatibility code is
114 sheduled for removal after 0.7.0.
115 - para_afh: New option: --preserve to reset the modification time to
116 the value of the original file after meta data modification.
117 - Overhaul of the compress filter code. The refined algorithm should
118 reduce clipping. The meaning of --aggressiveness has changed, see the
119 updated and extended documentation of the compress filter for details.
120 - Cleanup of the audio format handler code.
121 - We now build the tree using the .ONESHELL feature of GNU make,
122 which results in a significant speedup.
123 - Two robustness fixes for FreeBSD.
124 - para_client now supports RFC4716 private keys as generated with
125 ssh-keygen -m RFC4716. In fact, this key format has been made the
126 default, and the former PEM keys will be depreciated at some point.
127 - The ogg audio format handlers learned to detect holes and now report
128 the correct duration also if ogg pages are missing in the file. This
129 affects ogg/vorbis ogg/speex and ogg/opus.
130 - Robustness improvements for para_mixer.
131 - A fix for an old bug that could cause the server to crash or report
132 garbage in its status output.
133 - New para_play option: --end-of-playlist
134 - Streaming m4a files over udp has been improved.
135
136 Downloads:
137 [tarball](./releases/paraslash-0.6.3.tar.xz),
138 [signature](./releases/paraslash-0.6.3.tar.xz.asc)
139
140 --------------------------------------
141 0.6.2 (2018-06-30) "elastic diversity"
142 --------------------------------------
143
144 The main new feature of this release is the --listen-address option
145 for para_server and para_audiod. But there are also a large number of
146 small improvements and bug fixes all over the place. The user-visible
147 changes include
148
149 - para_gui no longer waits up to one second to update the screen when
150 the geometry of the terminal changes.
151 - Minor documentation improvements.
152 - Improvements to the crypto subsystem.
153 - The server subcommand "task" has been deprecated. It still works,
154 but prints nothing. It will be removed in the next major release.
155 - Server log output is now serialized, avoiding issues with partial
156 lines.
157 - It is now possible to switch to a different afs database by changing
158 the server configuration and sending SIGHUP to the server process.
159 - New server options: --listen--address, --http-listen-address and
160 --dccp-listen-address. These options restrict the set of listening
161 addresses of the TCP and DCCP sockets of the server process.
162 - The help commands of server, audiod, play now support --long. By default,
163 the short help is shown.
164 - The code which merges the command line options with the config file
165 options has been consolidated.
166 - If the current audio file is renamed, the status items are now updated
167 accordingly.
168 - After the server process received SIGHUP, changes to the current audio
169 file did not trigger an update of the status items. This has been fixed.
170
171 Downloads:
172 [tarball](./releases/paraslash-0.6.2.tar.xz),
173 [signature](./releases/paraslash-0.6.2.tar.xz.asc)
174
175 ----------------------------------------
176 0.6.1 (2017-09-23) "segmented iteration"
177 ----------------------------------------
178
179 The highlight of this release is the version 2 mood parser. But there
180 is a lot more than that, as summarized in the list below. And of
181 course we have many small usability improvements and bug fixes not
182 mentioned here.
183
184 - A more intuitive syntax for moods ("version 2 moods"). The
185 traditional version 1 moods are still supported but are deprecated
186 now. Removal of the version 1 mood parser is scheduled for the next
187 major release.
188 - Flex and bison are now required to build para_server.
189 - New sort order for the ls command: -s=h sorts the ls output by hash
190 value of the audio file.
191 - autogen.sh now runs the test suite after a successful build.
192 - The contents of overview.pdf have been integrated into the user
193 manual.
194 - Fixed sized audio format headers for ogg/opus streams.
195 - The doxygen source browser has been disabled temporarily. The
196 API reference is still online, though.
197 - Overhaul of the source code documentation.
198 - The deprecated --full-path option of the ls command has been
199 removed. It was a no-op since 0.6.0.
200 - The wma decoder has been cleaned up and its bitstream API made
201 more robust.
202 - The image/lyrics ID status items of the current audio file are now
203 updated on changes. This affects para_gui, which used to report the
204 old value until EOF.
205
206 Downloads:
207 [tarball](./releases/paraslash-0.6.1.tar.xz),
208 [signature](./releases/paraslash-0.6.1.tar.xz.asc)
209
210 ---------------------------------------
211 0.5.8 (2017-09-23) "branching parabola"
212 ---------------------------------------
213
214 A couple of fixes for serious old bugs have accummulated in the maint
215 branch since paraslash-0.5.7 was released nine months ago. So here
216 is another 0.5.x release. As usual, all commits in the maint branch
217 have been merged into master, so the upcoming paraslash-0.6.1 also
218 contains them. For details see the git log.
219
220 Downloads:
221 [tarball](./releases/paraslash-0.5.8.tar.bz2),
222 [signature](./releases/paraslash-0.5.8.tar.bz2.asc)
223
224 -------------------------------
225 0.6.0 (2017-04-28) "fuzzy flux"
226 -------------------------------
227
228 The highlights of this release are the conversion of all option parsers
229 to the lopsub library and the improved AAC support. But there are
230 many other user-visible changes all over the place. Also a lot of
231 old cruft has been removed, leading to incompatible changes. Hence
232 the new major version number.
233
234 - Support for Mac OS X has been removed.
235 - On Linux systems, glibc-2.17 or newer is required to build the
236 source tree.
237 - Support for RSA public keys in ASN format (as generated by openssl
238 genrsa) has been removed. These keys have been deprecated since
239 2011, so users should have long switched to keys generated with
240 ssh-keygen(1).
241 - If libgcrypt is used as the crypto library, we now require version
242 1.5.0 (released in 2011) or later.
243 - The insecure RC4 stream cipher has been removed. It was superseded
244 by aes_ctr128 three years ago but the RC4 code had been kept for
245 backwards compatibility.
246 - On Linux, abstract unix domain sockets are used unconditionally.
247 - The "install" target no longer strips executables, the new
248 install-strip target can be used to get the old behaviour.
249 - The clean targets have been renamed: clean2 is gone, and the new
250 mostlyclean removes only the executables and object files.
251 - New target: check (identical to test).
252 - The DESTDIR make variable is honored to prepend a path to the
253 installation directory. This feature is orthogonal to the --prefix
254 option to configure.
255 - Minor WMA cleanups.
256 - The aac audio format handler has been rewritten to use the mp4ff library.
257 See the manual for how to install the library on your system.
258 - New status item: max_chunk_size. The value is stored in a previously
259 unused field of the afhi object of the aft table. Although backwards
260 compatible, users are encouraged to re-add m4a files to populate
261 the new field.
262 - No more chunk tables for aac. Chunk boundaries are determined
263 dynamically at stream time.
264 - Release and master branch tarballs are now compressed with xz rather
265 than bzip2.
266 - The lopsub package is required to install the paraslash package.
267 Gengetopt is no longer needed.
268 - make dep is gone. Dependencies have been created automatically for
269 a long time and it was never necessary to run make dep manually.
270 - para_gui lost its --timeout option.
271 - Most manual pages have been extended to include an overall
272 description of the command.
273 - The --stream-delay option of para_audiod has been removed. It had
274 been a no-op for many years.
275 - The deprecated --path option of the server ls command has been
276 removed. The command now prints full paths by default, making
277 --full-path a no-op. Hence --full-path has been depreacted and is
278 scheduled for removal in v0.6.1.
279 - It is now possible to use 'CFLAGS' to override the default compiler
280 options.
281 - para_fade has been renamed to para_mixer. The four modes of operation
282 (set, fade, snooze, sleep) are implemented as subcommands of the
283 new program.
284 - The sleep subcommand of para_mixer (former sleep mode of para_fade)
285 lost its --wake-hour and --wake-min options in favor of the new
286 --wake-time option which also accepts relative wakeup times like
287 "+9:30".
288 - The new --fade-exponent option of para_mixer allows for non-linear
289 channel fading.
290 - The new logo.
291
292 Downloads:
293 [tarball](./releases/paraslash-0.6.0.tar.xz),
294 [signature](./releases/paraslash-0.6.0.tar.xz.asc)
295
296 -------------------------------------
297 0.5.7 (2016-12-31) "semantic density"
298 -------------------------------------
299
300 Mostly a bug fix release, and a bunch of internal improvements.
301 The only user-visible changes are the sanity checks for the touch
302 command and the new options to the ls command.
303
304 - Speedup of the base64 decoder.
305 - One of the two source browsers has been removed from the web pages.
306 The doxygen API reference still contains an HTML version of each
307 source file.
308 - Two race conditions in para_server have been fixed.
309 - ls -p is now deprecated in favor of -F or -b. See the help text of
310 the ls command for details.
311 - The openssl code has been adjusted to work also with openssl-1.1.
312 - The wma decoder and audio format handler now correctly decodes
313 files with unusual block sizes.
314 - We now compile with -Wformat-signedness if possible.
315 - The touch command now refuses to set an invalid image or lyrics ID.
316 - New section on contributing for the user manual.
317 - Major simplification of the error subsystem.
318
319 Downloads:
320 [tarball](./releases/paraslash-0.5.7.tar.bz2),
321 [signature](./releases/paraslash-0.5.7.tar.bz2.asc)
322
323 -------------------------------------------
324 0.4.14 (2016-12-31) "branching oscillation"
325 -------------------------------------------
326
327 Many fixes for nasty bugs have accumulated over the years since 0.4.13
328 was released in 2013. This release contains all of them. It is the
329 last v0.4.x release, so v0.4.x is now end of life. The maint branch
330 of the git repository has been forwarded to v0.5.7 and will start to
331 receive important fixes for 0.5.x, which is now in maintenance mode.
332
333 - Fix for the "double encryption bug" with gcrypt, which caused
334 garbage to be sent to stat clients.
335 - The "prev" command of para_play now works as expected.
336 - Invalid parameters for the begin and end chunk no longer cause
337 the afh receiver of para_recv to segfault.
338 - A fix for an assertion that triggered in the FEC slice setup code.
339 - Closing the terminal with a running para_gui causes the process to
340 die, rather than using 100% of CPU time.
341 - para_gui no longer shows stale information about the previous
342 audio file.
343 - A fix for a bug in the fec decoder which caused aborts if the buffer
344 pool filled up.
345 - The assertion in the buffer tree code no longer triggers.
346 - The "afs_event: table moods, event 0: key already exists in rbtree"
347 error no longer occurs on attribute changes.
348 - A NULL pointer dereference in com_sender() has been fixed.
349 - The -i option of lsatt works as advertised.
350 - Various fixes for the build system.
351
352 Downloads:
353 [tarball](./releases/paraslash-0.4.14.tar.bz2),
354 [signature](./releases/paraslash-0.4.14.tar.bz2.asc)
355
356 ---------------------------------------
357 0.5.6 (2016-07-10) "cascading gradient"
358 ---------------------------------------
359
360 The highlight of this release is the new -m flag for para_afh which
361 lets it modify the meta tags of the given audio file(s). This feature
362 is supported for all audio formats. Many small cleanups and bug fixes
363 not mentioned here have accumulated and are also part of the release.
364
365 - para_afh learned to modify meta tags of mp3 wma ogg spx
366 opus flac aac files.
367 - afs commands propagate error codes to the client.
368 - The check command now also checks the attribute table for
369 inconsistencies.
370 - New -v flag for the version command (print verbose version string)
371 - New option --priority for para_server and para_audiod.
372 - New mood methods: image_id and lyrics_id.
373 - The manual and this NEWS file have been converted to markdown.
374 - Support for the compile-time loglevel feature has been removed.
375 - Cleanup of the wma decoder and bitstream code.
376 - Improved wide-character support and fixes related to signal handling.
377 - para_gui no longer reports 100% playing time at the stream start.
378 - Opus cleanups.
379
380 Downloads:
381 [tarball](./releases/paraslash-0.5.6.tar.bz2),
382 [signature](./releases/paraslash-0.5.6.tar.bz2.asc)
383
384 --------------------------------------
385 0.5.5 (2015-09-20) "magnetic momentum"
386 --------------------------------------
387
388 Many new features and a lot of other improvements.
389
390 - On Linux systems, local sockets are now created in the
391 abstract name space by default. This allows to get rid of
392 the socket specials in /var/paraslash.
393 - The --user-allow option of para_audiod now accepts also
394 usernames rather than only user IDs.
395 - New autoconf macros to avoid duplication in configure.ac.
396 - Status items (as shown by para_gui) are updated correctly
397 when the meta information of the current audio changes.
398 - para_server and para_audiod no longer refuse to start in
399 the background if no log file is given. Instead, all log
400 messages go to /dev/null in this case.
401 - Web page cleanup.
402 - New syntax for the -l and -s options of the ls command.
403 These options should now be specified as -l=v rather than
404 -lv, for example. The old syntax still works, but support
405 will be dropped in v0.6.0.
406
407 Downloads:
408 [tarball](./releases/paraslash-0.5.5.tar.bz2),
409 [signature](./releases/paraslash-0.5.5.tar.bz2.asc)
410
411 ------------------------------------------
412 0.5.4 (2015-01-23) "exponential alignment"
413 ------------------------------------------
414
415 Another cleanup and bugfix release.
416
417 - New server command: tasks.
418 - Minor cleanups to daemon.c.
419 - New URLs for home page and git services.
420 - Improved error diagnostics for the mvblob commands.
421 - New sender subcommand: status.
422 - Improved help text for server and afs commands.
423 - audiod memory leak fixes.
424 - Miscellaneous improvements to the build system.
425 - oss_writer improvements.
426 - Improved handling of mp3 files with both id3v1 and id3v2 tags.
427
428 Downloads:
429 [tarball](./releases/paraslash-0.5.4.tar.bz2),
430 [signature](./releases/paraslash-0.5.4.tar.bz2.asc)
431
432 ---------------------------------------------
433 0.5.3 (2014-08-01) "symbolic synchronization"
434 ---------------------------------------------
435
436 Not many new features, but lots of fixes and usability improvements.
437
438 - para_gui has been converted to use the paraslash scheduler.
439 - Various alsa-related fixes, mostly for the raspberry pi.
440 - Many scheduler improvements and cleanups.
441 - The test suite has been extended to include sanity checks
442 for the generated man pages.
443 - ao_writer fixes. This writer was in a quite bad shape. Many
444 serious bugs have been fixed.
445 - new audiod command: version.
446 - Minor improvements to the bitstream API.
447 - The cpsi command now prints a meaningful error message if
448 none of the given patterns matched any audio file.
449
450 Downloads:
451 [tarball](./releases/paraslash-0.5.3.tar.bz2),
452 [signature](./releases/paraslash-0.5.3.tar.bz2.asc)
453
454 ----------------------------------------
455 0.5.2 (2014-04-11) "orthogonal interior"
456 ----------------------------------------
457
458 The new sync filter, the AES_CTR128 stream cipher and the overhauled
459 network code are the highlights of this release. It also includes a
460 fair number of smaller fixes and improvements not mentioned here.
461
462 - The new sync filter synchronizes playback between multiple
463 clients.
464 - Connections between para_server and para_client are now
465 encrypted by means of AES rather than RC4 if both sides
466 support it. RC4 is still available as a fallback. This
467 feature is fully transparent, i.e. no command line options
468 are necessary, and a client linked against openssl can
469 speak with a server linked against libgcrypt and vice versa.
470 - Major cleanup of the networking subsystem.
471 - Improvements to para_fade: the new set mode, multi-channel
472 initial volumes, better error logging.
473 - The man pages of para_audiod, para_filter, para_recv, and
474 para_write contain the relevant options for receivers, filters,
475 writers. This broke in 0.5.0.
476 - ogg/vorbis latency improvements.
477 - Improved user manual.
478 - Minor fixes to avoid clang warnings.
479
480 Downloads:
481 [tarball](./releases/paraslash-0.5.2.tar.bz2),
482 [signature](./releases/paraslash-0.5.2.tar.bz2.asc)
483
484 ------------------------------------------
485 0.5.1 (2013-12-20) "temporary implication"
486 ------------------------------------------
487
488 Lots of fixes and improvements all over the place, and a major overhaul
489 of the build system.
490
491 - Audiod improvements and fixes.
492 - Buffer tree robustness improvements.
493 - Cleanup of the mood subsystem.
494 - Fixes and cleanups for the flac decoder.
495 - Latency improvements for the ogg/opus decoder.
496 - Crypto support is now optional. On systems without
497 openssl/gcrypt, the build succeeds but para_server,
498 para_audiod, para_client won't be built.
499 - The build system now works for cross-compile setups.
500 - The dependency tree has been flattened, which speeds up
501 builds and avoids to recreate the man pages on every change.
502 - The error code helper has been rewritten from perl to C,
503 which further improves build time.
504 - Many small bugs in the build system have been identified
505 and fixed.
506
507 Downloads:
508 [tarball](./releases/paraslash-0.5.1.tar.bz2),
509 [signature](./releases/paraslash-0.5.1.tar.bz2.asc)
510
511 ----------------------------------------
512 0.5.0 (2013-08-23) "invertible validity"
513 ----------------------------------------
514
515 Some API-breaking changes, one serious bug fix, and a lot of bike-shedding.
516
517 - The sideband compatibility code has been removed, hence
518 sideband connections (introduced in 0.4.11) are now mandatory.
519 - Addblob commands can produce output.
520 - The stat command no longer sends garbage when para_server was
521 compiled against libgcrypt.
522 - Dependencies for gengetopt files are computed automatically.
523 This eliminates a constant source of build bugs.
524 - The setatt command now accepts file name patterns rather than only
525 path names.
526 - overview.pdf is now based on dia, a simple diagram creation program.
527 The new version is much more detailed and contains descriptions of
528 the various programs of the paraslash package.
529 - The separator of all multi-word options has been changed from
530 underscore to dash. For example --log_color becomes --log-color.
531 - Overhauled web pages and the new logo.
532
533 Downloads:
534 [tarball](./releases/paraslash-0.5.0.tar.bz2),
535 [signature](./releases/paraslash-0.5.0.tar.bz2.asc)
536
537 --------------------------------------
538 0.4.13 (2013-07-29) "spectral gravity"
539 --------------------------------------
540
541 One more 0.4.x release before the API-breaking changes for 0.5.0 go
542 in. The main features of this release are the ogg/opus audio format,
543 and UTF-8 support, but it includes also tons of other improvements
544 and fixes all over the place.
545
546 - New audio format: ogg/opus.
547 - UTF8 support for para_gui and the mp3 audio format handler.
548 - Scheduler improvements and fixes.
549 - The obsolete gettimeofday() function has been replaced
550 by clock_gettime() on systems which support it.
551 - Speed and usability improvements for para_gui.
552 - para_client now restores the fd flags of stdin and stdout
553 on shutdown.
554 - Improved manual pages.
555 - Consistent version strings for all executables.
556 - Reduced dependencies on generated files result in fewer
557 recompilations on changes.
558 - Performance improvements for the compress filter.
559 - Improved downloads web page.
560
561 -----------------------------------------
562 0.4.12 (2012-12-20) "volatile relativity"
563 -----------------------------------------
564 The new command line player, the resample filter, ALSA support for
565 para_fade, and the improved build system are the highlights of this
566 release which probably marks the end of the 0.4.x series.
567
568 - The afh receiver and the para_play executable.
569 - The resample filter: A sample rate converter based on
570 libsamplerate.
571 - The "versions" directory has been removed from the master
572 branch. The tarballs of the old releases are now available
573 in the new "releases" branch.
574 - Overhaul of the build system: All generated files are now
575 written to the "build" directory.
576 - The modular mixer API and the alsa mixer.
577 - Minor fixes for the osx writer.
578
579 --------------------------------------
580 0.4.11 (2012-07-20) "mutual diversity"
581 --------------------------------------
582
583 The major feature in this release is the new sideband API for
584 client-server communication. This API will be used exclusively starting
585 with 0.5.0, which breaks backward compatibility but allows to get rid
586 of quite some compatibility code. Other noteworthy changes include
587 decoder latency improvements and a long-standing bug fix for the
588 ALSA writer.
589
590 - Sideband connections: If both para_server and para_client
591 support this feature, data is sent as a multiplexed stream.
592 - The --no_default_filters option of para_filter has been
593 removed.
594 - Several fixes and latency improvements to various decoders.
595 - The ALSA writer now limits the prebuffer time to 500ms.
596 - Documentation improvements.
597 - Overhaul of the command_util.sh script.
598 - Fixes for some minor problems found by the clang analyzer.
599 - Compiles (almost) without warnings on gcc-3.
600 - Robustness improvements of the buffer tree code.
601
602 ------------------------------------------
603 0.4.10 (2012-03-30) "heterogeneous vacuum"
604 ------------------------------------------
605
606 Nothing earth-shaking in this release, but quite a few usability
607 improvements and the usual mix of cleanups and fixes.
608
609 - The --no_default_filters option of para_filter has been
610 deprecated. It still works but has no effect and will be
611 removed in the next version.
612 - para_gui now prints also the stderr output of the executing
613 command in the bottom window.
614 - Cleanup and consolidation of the various wrappers for
615 write(), writev(), send() and friends.
616 - The obscure error messages on mmap() failures have been
617 replaced by meaningful messages. This affects mainly
618 para_afh.
619 - para_audioc: Cleanups and memory leak fixes.
620 - Test 0004-server no longer fails if para_server is not
621 being built.
622 - New configure options: --with-id3tag-{headers,libs}.
623
624 -------------------------------------
625 0.4.9 (2011-12-06) "hybrid causality"
626 -------------------------------------
627
628 Support for another audio format, interactive mode for para_client
629 and para_audiod and many small improvements/fixes all over the place.
630
631 - Support for flac, the free lossless audio codec.
632 - Fix for an endless loop in the mp3 decoder for certain
633 (corrupt) mp3 files.
634 - When executed without specifying a command, para_client
635 and para_audioc start an interactive shell (requires
636 libreadline being installed). The interactive mode offers
637 full tab completion and command line history.
638 - autogen.sh now detects a distcc setup and adjusts the
639 parameter for the -j option of make accordingly.
640 - Shared memory areas are no longer restricted to 64K. We now
641 detect the maximal size of a shared memory area at runtime.
642 - cleanup of the internal uptime API.
643 - para_server prefaults the mmapped audio file to avoid
644 delays on slow media.
645 - A new test for the test-suite that exercises the
646 communication between para_server and para_audiod.
647 - The alsa writer eats up less CPU cycles when configured to
648 use the DMIX plugin.
649 - Simplified and unified receiver code.
650 - Makefile cleanups.
651 - Commands which print a list of matching audio files now
652 emit a meaningful error message if no audio file matched the
653 given pattern(s).
654
655 --------------------------------------
656 0.4.8 (2011-08-19) "nested assignment"
657 --------------------------------------
658
659 Gcrypt support, the overhauled osx writer and regex format specifiers
660 are the highlights of this release.
661
662 - support for libgcrypt as a drop-in replacement for openssl.
663 Run configure --enable-cryptolib=gcrypt to link against
664 libgcrypt. The two crypto implementations are compatible to
665 each other, i.e. a para_client executable linked against
666 gcrypt can connect to para_server linked against libssl
667 and vice versa.
668 - Rewrite of the osx writer (output plugin for Mac OS).
669 - audiod: The format specifier for receivers, filters and
670 writers is now treated as a regular expression. This allows
671 to replace 5 lines in the config file (one for each audio
672 format) by one single line. See the manual for details.
673 - The \*.cmdline.[ch] files are no longer contained in the released
674 tarballs. This reduces the size of the tarballs but requires
675 gengetopt to build the tarball.
676 - Compiles cleanly also with llvm/clang.
677 - Corrupt mp3 files are handled more gracefully.
678 - The alsa writer uses poll fds instead of computing timeouts.
679 - Cleanup of the generic writer API.
680 - sched: Optimized zero timeouts.
681 - vss timeout cleanups.
682 - oggdec fixes and improvements.
683
684 --------------------------------------
685 0.4.7 (2011-06-01) "infinite rollback"
686 --------------------------------------
687
688 The new ao writer, support for ssh RSA keys and a couple of other
689 enhancements.
690
691 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
692 platforms through the libao audio library.
693 - Support for RSA keys generated with ssh-keygen.
694 - configure: improved options for ogg/vorbis/speex.
695 - The git version reported by --version always matches HEAD.
696 - The autogen script detects the number of processors and
697 runs a parallel make if possible.
698 - Major cleanup of the crypto API.
699 - Documentation updates.
700
701 ------------------------------------------
702 0.4.6 (2011-03-31) "deterministic entropy"
703 ------------------------------------------
704
705 Lots of ogg/vorbis improvements, the new test suite, enhancements
706 for para_gui and a fair amount of other bug fixes.
707
708 - For DCCP/OGG streams the audio file header is only sent once
709 at the beginning of the stream rather than periodically
710 every five seconds. This reduces network traffic and the
711 FEC group size.
712 - The vorbis comment header is replaced by an empty dummy header
713 before the header is sent over the network. This also results in
714 less network traffic and smaller FEC groups.
715 - The new "test" make target allows to perform some sanity checks prior
716 to installing the package.
717 - ogg timing fixes and performance improvements
718 - Scheduler improvements
719 - Proper exit codes for para_write
720 - para_gui: New option --theme to select a startup theme. Several
721 other improvements and fixes.
722 - aacdec error message cleanups
723 - simplified color error handling
724
725 --------------------------------------------
726 0.4.5 (2010-12-17) "symmetric randomization"
727 --------------------------------------------
728
729 Bug fixes, internal cleanups and variable-sized FEC slices.
730
731 - Contains a fix for an invalid-free-bug in the ogg audio format
732 handler code.
733 - Switching off the DCCP sender works again.
734 - para_audiod handles crashes of para_server more robustly.
735 - Internal scheduler and writer cleanups.
736 - Reduced latency due to variable-sized FEC slices.
737 - Improved documentation and error diagnostics.
738 - The build of para_server is now optional, allowing the build
739 to succeed in case libosl is not installed.
740
741 ------------------------------------------
742 0.4.4 (2010-08-06) "persistent regularity"
743 ------------------------------------------
744
745 Support for yet another audio format, para_write improvements and
746 bug fixes.
747
748 - Support for the speex codec.
749 - Support for sample formats other than 16 bit little endian.
750 - error2.h is now created by a perl script which speeds up configure
751 considerably.
752 - Fix a bug in the aac decoder which could lead to segfaults in
753 para_filter/para_audiod.
754 - Fixes for autoconf-2.66.
755
756 ----------------------------------------
757 0.4.3 (2010-07-05) "imaginary radiation"
758 ----------------------------------------
759
760 Many improvements for the DCCP and the UDP transport, the new user
761 manual and the usual mix of bug fixes and internal improvements.
762
763 - FEC support for the DCCP sender (Gerrit Renker). The new
764 --dccp_max_slice_size, --dccp_data_slices_per_group and
765 --dccp_slices_per_group options can be used to set the FEC
766 parameters for the DCCP transport.
767 - DNS lookups for UDP targets (Gerrit Renker).
768 - The new user manual replaces the README, README.afs, REQUIREMENTS
769 and INSTALL documents.
770 - Fix an end-of-file detection bug in the oggdec filter.
771 - The new nonblock API.
772 - Both options of the oggdec filter have been removed.
773 - New debug mode for the internal scheduler.
774
775 ------------------------------------------
776 0.4.2 (2010-04-23) "associative expansion"
777 ------------------------------------------
778
779 It's been some time since the last release, but finally here is
780 paraslash-0.4.2. The bulk of the changes comes from the new buffer
781 tree API, but there are changes all over the tree. Mainly performance
782 and usability improvements, but also quite some bug fixes.
783
784 - The new buffer tree API.
785 - DCCP: Support for CCID negotiation (Gerrit Renker).
786 - UDP robustness fixes.
787 - The --bufsize option for mp3dec is gone as it no longer makes sense
788 for the new buffer tree API.
789 - Fix audible buffer underruns for wma streams.
790 - The alsa writer no longer prints meaningless underrun durations.
791 - audiod: Defaults work also for udp streams. If no filter is
792 given for an audio format that is received via upd, fecdec is
793 automatically added as the first filter (along with the decoder).
794
795 ---------------------------------------
796 0.4.1 (2009-12-22) "concurrent horizon"
797 ---------------------------------------
798
799 Support for another audio format, minor feature enhancements and lots of bug
800 fixes. All fixes that have been accumulated in the maint branch (in particular
801 those mentionened in the 0.3.6 release notes) appear in this release as well.
802
803 - wma support.
804 - new afh option: --human to activate human-readable output.
805 - new server/audiod option: --log-timing to print timing information.
806 - build system improvements.
807 - source code documentation updates.
808
809 -------------------------------------
810 0.3.6 (2009-12-07) "cubic continuity"
811 -------------------------------------
812
813 Quite a few bugs have been found and fixed since 0.3.5, so here's
814 another 0.3.x release. No new features.
815
816 - Always check return value of malloc().
817 - ogg vorbis/FEC: Do not write garbage after the audio file header.
818 - exit if root privileges could not be dropped.
819 - FEC: Fix computation of extra slices.
820 - oss: Fix check for empty input buffer.
821 - Avoid buffer underruns due to filter chain output buffer constraints.
822 - server: Fix assignment of afs_pid.
823 - Don't panic if the afs database contains unknown audio formats.
824 - http/dccp: Do not send the audio file header twice.
825 - FEC: Timing improvements.
826
827 ----------------------------------------------
828 0.4.0 (2009-11-10) "simultaneous independence"
829 ----------------------------------------------
830
831 Two significant changes which require the new version number: The
832 improved authentication dialog and the fact that the database code
833 has been moved to a library, libosl. To use the new version, you have
834 to generate new RSA keys, see INSTALL for details. A shell script is
835 provided for conversion of the 0.3 database to the new 0.4 format.
836
837 - stronger crypto for client authentication
838 - the database code has been moved to a library
839 - improved status item handling
840 - cleanup of the build system
841 - The "-V" option now also prints the git version
842 - the new parser-friendly listing mode for the ls and stat commands
843 - mandatory rc4 encryption
844 - major audio format handler cleanups
845 - (id3,...) tags are no longer stored as a combined string in the database
846 - new mood methods: artist_matches, title_matches, comment_matches,
847 album_matches, year_maches, year.
848
849 --------------------------------------------
850 0.3.5 (2009-09-21) "symplectic separability"
851 --------------------------------------------
852
853 Full client support for \*BSD Unixes, complete re-write of the ogg
854 vorbis audio format handler, various improvements all over the place
855 and the usual mix of bugfixes. This release marks the end of the 0.3
856 series if no serious problems show up.
857
858 - the new oss writer (supported on \*BSD and Linux)
859 - rewrite of the ogg vorbis audio format handler. It's
860 recommended to replace the chunk tables of existing ogg
861 vorbis files in the afs database by re-adding these files
862 with "add -f".
863 - support for netmask subsets (Gerrit Renker)
864 - the new prebuffer filter
865 - improved signal handling
866 - variable fec output buffer size
867 - improved FEC timing fixes audible buffer underruns in UDP mode
868 - --log_color actually works
869 - new ls option: -d (print dates as seconds after the epoch)
870 - update to gengetopt 2.22.2
871 - support for RSA keys of size > 512 bits
872 - new option "mixer_channel" for para_fade
873
874 -----------------------------------------
875 0.3.4 (2009-05-07) "elliptic inheritance"
876 -----------------------------------------
877
878 The new udp sender, forward error correction, colored logs and various
879 other improvements. As the udp sender does not depend on any special
880 libraries, it is built unconditionally.
881
882 - The udp sender replaces the ortp sender. The new code uses forward
883 error correction to protect against packet losses. Many thanks to
884 Gerrit Renker for providing ipv6 support.
885 - The default port for udp streaming now defaults to 8000, like
886 for the http and the dccp senders/receivers.
887 - Loglevels are now specified as symbolic names, e.g.
888 "--loglevel info".
889 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
890 - para_server/para_audiod: Color support for log messages.
891 - new options for mp3dec: --ignore-crc, --bufsize
892 - new audiod option: --config-file.
893 - gengetopt cleanups.
894 - Improved help/man pages: The documentation of para_audiod,
895 para_recv, para_filter and para_write now also contains
896 all options of the available receivers/filters/writers. The
897 man page of para_fade contains a description of the different
898 modes of operation.
899 - More source code documentation.
900 - vss timing fixes.
901
902 --------------------------------------------
903 0.3.3 (2008-12-01) "axiomatic perspectivity"
904 --------------------------------------------
905
906 Internal code cleanups, bug fixes, improved tag handling and the new
907 amplification filter.
908
909 - para_server uses the generic scheduling code.
910 - overhaul of the virtual streaming system.
911 - mp3: id3 version 2 support via libid3tag (optional)
912 - ogg: vorbis comment support.
913 - aac meta info support.
914 - mp3 audio format handler cleanups.
915 - new filter: "amp" to amplify the amplitude of the audio stream
916 - new status item/database entry: amplification. It is
917 used by the amp filter to pre-amplify the audio stream.
918 - fix a close-without-open bug in para_write.
919 - fix a bug in com_init() which was introduced in 0.3.2.
920 - better error diagnostics for para_client.
921
922 -----------------------------------------
923 0.3.2 (2008-04-11) "probabilistic parity"
924 -----------------------------------------
925
926 The new para_afh executable, scheduling and documentation improvements.
927
928 - new ls option: -lc (list chunk table)
929 - new executable: para_afh, the stand-alone audio file handler tool
930 - afs commands can send output more than SHMMAX (32MB on Linux). This
931 also reduces the memory usage of commands that produce large amounts
932 of output.
933 - major scheduler and audiod cleanups.
934 - more detailed and much nicer man pages.
935
936 ---------------------------------------
937 0.3.1 (2008-02-23) "liquid interaction"
938 ---------------------------------------
939
940 A mix of cleanups, bug fixes, improvements, and some new features. No
941 significant changes to the new database (osl) code, which is generally
942 a good sign.
943
944 - Share some similar/duplicate code between the http and the
945 dccp sender.
946 - Generic access control lists for paraslash senders.
947 - dccp sender: Access control lists, connection limiting and support
948 for the allow,deny,on,off,help sender commands.
949 - The default dccp port changed from 5001 to 8000 (suggested by
950 Gerrit Renker).
951 - para_server starts even if not all public keys could be loaded.
952 - Audiod performance improvements.
953 - fix a bug in the "off" command of the http sender.
954 - fix some fd and memory leaks.
955 - Update to gengetopt-2.22.
956
957 -------------------------------------
958 0.3.0 (2008-01-12) "solar saturation"
959 -------------------------------------
960
961 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
962
963
964 Usually one might expect lots of new features AND a big increase in size
965 for a major release like this.
966
967 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
968 ever. The decrease in size is mostly due to the removal of some
969 graphical tools (which were only quick hacks anyway). But also the
970 fact that the mysql code is gone cuts down the size a bit.
971
972 Being independent of mysql comes at a cost: The fact that paraslash
973 now contains its own database (the object storage layer, osl) increases
974 the (stripped) binary size of para_server by ~50K on i386.
975
976 - no more restrictions on unique basenames.
977 - independent of mysql: The new self-contained object
978 storage layer (osl) replaces the mysql database.
979 - New executable para_fsck: Check integrity of osl tables.
980 - Lyrics support.
981 - Reliable audio file move/rename detection.
982 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
983 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
984 Solaris (sparc64).
985 - the new osl-based audio file selector (afs) replaces the random,
986 the playlist and the mysql selector of paraslash-0.2.x.
987 - IPv6 support (thanks to Gerrit Renker).
988 - paraslash-0.2.x streams are now called "moods". Writing
989 0.3.x-mood definitions should be both easier and more
990 powerful than writing 0.2.x-stream definitions.
991 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
992 been removed. The world is a better place without them. However,
993 para_gui is still there.
994 - afs tracks audio file selection also in playlist mode.
995 - few easy-to-use afs commands replace the many not-so-easy-to-use
996 mysql commands (and are available also in playlist mode).
997 - Improved error subsystem.
998 - The earth-shaking new logo.
999
1000 -----------------------------------------
1001 0.2.17 (2007-11-20) "isotropic threshold"
1002 -----------------------------------------
1003
1004 Mainly bugfixes and cleanups in this version which marks the end of
1005 the 0.2.x series if no serious bugs show up after the release.
1006
1007 - mysql_selector: fix a locking bug.
1008 - universal chunk queueing.
1009 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
1010 to Gerrit Renker).
1011 - be more carful wrt. signed vs. unsigned argument passing.
1012 - cleanup error.h and fix some references to invalid error
1013 codes.
1014 - update to gengetopt-2.21.
1015 - update to ortp-0.13.1.
1016 - autoconf: extend checks for headers, library functions and
1017 compiler characteristics.
1018 - Fix streaming of large mp3 files.
1019 - Fix an off-by-one bug in playlist handling.
1020
1021 --------------------------------------
1022 0.2.16 (2007-04-05) "neural discharge"
1023 --------------------------------------
1024
1025 The main change in this release is the major audio format handler
1026 cleanup which removes some similar/duplicate code and makes it easier
1027 to implement plugins for other audio formats. Of course, the usual mix
1028 of other improvements/changes/bugfixes also made it into the release.
1029
1030 - simplified audio format handlers (most of the handling functions
1031 were moved one layer up to the virtual streaming system).
1032 - para_server uses mmap to read audio files
1033 - repositioning of mp3 streams is much faster, in particular for
1034 jumping near the end of large mp3 files.
1035 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
1036 and AFS_WRITE.
1037 - fix a bug in para_filter that caused decoding of aac files
1038 to start only after a few seconds.
1039 - fix osx_writer hangs
1040 - simplified dccp code (thanks to Gerrit Renker)
1041 - the compress filter works also on big endian systems (ppc)
1042
1043 -----------------------------------------
1044 0.2.15 (2007-02-16) "inductive resonance"
1045 -----------------------------------------
1046
1047 Minor improvements, more documentation and a bunch of bug fixes.
1048
1049 - para_server: The server.users file is only read once on server
1050 startup rather than for each connection
1051 - mp3dec: Fix decoding of corrupt mp3 files
1052 - afs (audio file sender) is now called vss (virtual streaming
1053 system). Consequently, the permission flags specified in
1054 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
1055 become VSS_READ and VSS_WRITE respectively.
1056 - para_audiod/para_filter: Fix a bug that caused the last chunk
1057 of audio data not being written under certain circumstances
1058 - audiod: compute the difference of server time and local time
1059 correctly
1060 - para_server/para_audiod: Fix some memory leaks
1061 - documentation improvements
1062 - configure.ac: fix checks for para_krell
1063 - new man pages
1064
1065 -------------------------------------------
1066 0.2.14 (2006-10-15) "transient singularity"
1067 -------------------------------------------
1068
1069 The only major enhancement of this version is the osx writer which completes
1070 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
1071 until now for reasons beyond the scope of this changelog entry.
1072
1073 - new output plugin for Mac Os: the osx writer
1074 - rename configure command line options from --enable-xxx-headers to
1075 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
1076 - configure: new command line options: --with-mad-headers,
1077 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
1078 - some robustness fixes
1079 - dymamic audio format recognition for audiod
1080 - para_server: new command line option: --autoplay_delay
1081 - para_audiod: new command line option: --clock_diff_count
1082
1083 ---------------------------------------
1084 0.2.13 (2006-07-14) "sonic convolution"
1085 ---------------------------------------
1086
1087 A bunch of new features and core changes.
1088
1089 - the new paraslash scheduler, short and sweet.
1090 - Support for m4a/mp4 files via the new aac audio format
1091 handler/filter (requires libfaad).
1092 - each writer has its own command line parser, just like
1093 para_recv and para_filter.
1094 - para_client and para_audioc use the error subsystem
1095 - writers are integrated in para_audiod (currently linux-only)
1096 - para_client is integrated in para_audiod
1097 - random/playlist selector: improved info strings
1098 - new audiod commands: tasks, kill
1099 - update to libortp-0.10.1
1100 - para_fade: wake time defaults to 8 hours from now
1101 - update to autoconf-2.60
1102
1103 ------------------------------------------
1104 0.2.12 (2006-05-12) "oriented abstraction"
1105 ------------------------------------------
1106
1107 Many user-visible changes in this release and lots of new
1108 features:
1109
1110 - the new optional dccp sender/receiver. It uses the datagram
1111 congestion control protocol. You'll need a fairly new kernel
1112 for this.
1113 - paraslash works on Mac OS X (thanks to Gerd Becker)
1114 - para_play renamed to para_write
1115 - modular output plugin design (writers) for para_write
1116 - new file_writer output plugin for para_write
1117 - compress filter speed improvements
1118 - update to libortp-0.9.1
1119 - update to gengetopt-2.17rc
1120 - para_client no longer depends on libreadline (as the
1121 code for the interactive mode was removed).
1122 - gcc-2-95 is no longer a supported compiler. It may still
1123 work, but it gets no more testing.
1124 - the tarball no longer contains the screenshot images which
1125 reduces its size quite a bit.
1126 - configure: new command line options: --enable-mysql-headers
1127 and --enable-mysql-libs
1128
1129 ------------------------------------
1130 0.2.11 (2006-03-11) "atomic duality"
1131 ------------------------------------
1132
1133 Here it is, the first paraslash release developed with git. There
1134 are fairly many user-visible changes in this release. As two out of
1135 the three "database tools" of paraslash don't use a database at all,
1136 they are now called "audio file selectors" instead.
1137
1138 - the cdt command (change database tool) becomes chs (change
1139 selector)
1140 - no more colon separators: The syntax of some options of
1141 para_audiod and para_filter have changed. Use --help for
1142 more info (and some examples).
1143 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
1144 - switch from cvs to git (should've done that earlier)
1145 - the new ipc subsystem
1146 - new audio file selector: playlist
1147 - para_server: the dopey selector is now called "random",
1148 and is the default selector. Use the --selector option to
1149 choose another selector at startup, or the chs command to
1150 change the selector at runtime.
1151 - X86_64 fixes (thanks to Steffen Klassert)
1152 - para_play fixes
1153
1154 --------------------------------------
1155 0.2.10 (2006-02-17) "cyclic attractor"
1156 --------------------------------------
1157
1158 Huge documentation update, a scrollable window for para_gui, ortp
1159 improvements, and of course many small fixes not mentioned here.
1160 The diffstat below is rather misleading as most insertions are due
1161 to the new source documentation.
1162
1163 - autoconf cleanup
1164 - para_server also uses the new error subsystem
1165 - lots of new documentation (UTSL)
1166 - gui improvements:
1167 - keysyms for cursor keys and for next/previous page keys
1168 - scrollable output window
1169 - new internal commands: scroll up/down, page up/down
1170 - fix color of command output.
1171 - ortp: the --chunk_time and --header flags are no longer needed
1172 for para_recv/para_audiod as this information is now encoded in
1173 each rtp packet sent by para_server.
1174
1175 -------------------------------------------
1176 0.2.9 (2006-01-24) "progressive turbulence"
1177 -------------------------------------------
1178
1179 Internal audiod receivers/filters, the new error subsystem and
1180 a lot of small improvements.
1181
1182 - para_recv and para_filter are integrated into the para_audiod
1183 binary, i.e. audiod no longer spawns a new process for
1184 each receiver/filter. As para_recv and para_filter might be
1185 useful as standalone programs, they still get built (linked
1186 against the same object files that are also used for audiod).
1187 - further ortp timing improvements should reduce the CPU usage
1188 of the ortp receiver.
1189 - improved audio grabbing. The 'grab' command of para_audiod
1190 has its own set of command line options. Read the output of
1191 "para_audioc help grab" for more info.
1192 - oggdec: configurable input prebuffer size.
1193 - the new error subsystem gives better error diagnostics
1194 and reduces code size.
1195
1196 -----------------------------------------
1197 0.2.8 (2006-01-02) "dynamic accumulation"
1198 -----------------------------------------
1199
1200 The new modular filter design and the para_play-hangs bugfix.
1201
1202 - new executable: para_filter. It combines para_mp3dec,
1203 para_oggdec and para_compress. It also adds a further filter
1204 type, wav, that just inserts a wave header at the desired point
1205 of the filter chain. All 'piping' is done in-memory (i.e. no
1206 read/write operations are used).
1207 - para_play: fix a stupid bug that caused it to hang under
1208 certain circumstances.
1209
1210 -------------------------------------------
1211 0.2.7 (2006-12-27) "transparent invariance"
1212 -------------------------------------------
1213
1214 Not many user-visible changes but a fair amount of internal improvements.
1215
1216 - The http sender buffers data if it can not be sent
1217 out immediately (because the socket is not writable). This
1218 should prevent para_server from shutting down the connection
1219 too early on a loaded network.
1220 - para_play also prebuffers data if it is told to start at a
1221 future time by the --start_time option.
1222 - The return of para_recv: It combines para_ortp_recv and
1223 para_http_recv. Use the --receiver option to switch between
1224 the two. para_recv builds without libortp, but contains
1225 only the http receiver in this case.
1226 - update to ortp 0.8.1. As this ortp release contains incompatible
1227 changes, para_recv-0.2.7 won't link against older ortp libs.
1228 - improved ortp timings.
1229 - use of gcc-extensions that #define away for non-gcc and
1230 gcc < 3.0.
1231
1232 -------------------------------------------
1233 0.2.6 (2005-10-29) "recursive compensation"
1234 -------------------------------------------
1235
1236 Transparent session encryption (uses openssl's Alleged RC4 cipher),
1237 the internal find command and several other improvements and cleanups.
1238
1239 - Encrypt the session if encryption is requested by the client
1240 (default for para_client 0.2.6). This is backwards
1241 compatible, so older clients can still connect to para_server
1242 0.2.6. Use the new client option --plain to request an
1243 uncrypted session (off by default, must be set to on in
1244 order to connect to para_server 0.2.x with 0 <= x <= 5).
1245 - para_server uses an internal function to locate audio files
1246 rather than calling find(1). The server option
1247 --mysql_audio_file_dir replaces --mysql_find_cmd.
1248 - documentation update
1249 - man pages
1250 - header file cleanup
1251 - para_client code cleanup
1252 - para_gui: faster display of output of display commands
1253
1254 ------------------------------------------
1255 0.2.5 (2005-10-13) "aggressive resolution"
1256 ------------------------------------------
1257
1258 This release adds internal senders, i.e. no more external programs are
1259 spawned for sending out the audio data. There are two different senders
1260 available: The http sender and the ortp sender (former para_send which
1261 is no longer needed).
1262
1263 The new sender code has a plugin-like design so it can be easily
1264 extended should there be be any future need for supporting another
1265 network streaming protocol. All senders are completely independent of
1266 each other. In particular, the http and the ortp sender can operate
1267 in parallel.
1268
1269 - new server command: sender to control senders at runtime.
1270 Read the output of "para_server -h" and "para_client help
1271 sender" for more information.
1272 - para_recv renamed to para_ortp_recv
1273 - new executable: para_http_recv, a simple command line
1274 http receiver.
1275 - major afs/mp3/ogg code simplifications due to internal
1276 senders.
1277 - ogg timing improvements
1278 - fix several minor memory leaks (found by valgrind)
1279 - empty stream definitions work again
1280 - com_ne(): ignore errors on remove
1281 - audiod: fix segfault on server restart
1282
1283 ---------------------------------------
1284 0.2.4 (2005-09-21) "toxic anticipation"
1285 ---------------------------------------
1286
1287 Several small improvements, fixes and the new grab command.
1288
1289 - audiod:
1290 - new command: "grab" to grab the output of the stream reader
1291 or any filters. Read the output of "para_audioc help grab"
1292 for more information.
1293 - fix memory leak
1294 - code cleanup
1295 - audioc: new command line option: --bufsize to specify a
1296 buffer size different from the default size 8192.
1297 - improved error diagnostics for para_play.
1298 - new configure option: --enable-ssldir so search for openssl in
1299 non-standard places
1300 - sdl_gui: Make it look nice again for 1024x768
1301 - server: report total size of memory allocated with sbrk by malloc,
1302 new command line option: --announce_time
1303
1304 -----------------------------------------
1305 0.2.3 (2005-09-01) "hydrophilic movement"
1306 -----------------------------------------
1307
1308 Two new executables and major feature enhancements.
1309
1310 - audiod filters: It is now possible to specify arbitrary many
1311 (including none) filters for each supported audio
1312 format. This can be used e.g. for normalizing volume,
1313 transforming or grabbing the audio stream, or for using
1314 visualizers. Read the output of "para_audiod -h" for the
1315 syntax of the new --filter_cmd option.
1316 - new executable: para_play, a tiny alsa player. It
1317 can play wave files or raw pcm (16 bit little endian)
1318 from stdin.
1319 - new executable: para_compress, a dynamic range compressor
1320 intended to keep audio output at a consistent volume. Derived
1321 from [AudioCompress](http://trikuare.cx/code/AudioCompress.html).
1322 - audiod: New option: --stream_delay. This can be used in
1323 a local network to syncronize the audio output of all
1324 clients that play the same stream.
1325
1326 ------------------------------------------
1327 0.2.2 (2005-08-19) "tangential excitation"
1328 ------------------------------------------
1329
1330 Mostly internal changes in this release, but also some new commands
1331 for the mysql database tool.
1332
1333 - cleanup exec.c, fix para_exec bug
1334 - compile time loglevel (log messages below the given level
1335 won't be compiled in, which reduces the size of the
1336 resulting binaries)
1337 - new log macros that shorten the size of the source code.
1338 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
1339 to send only zeros. With this workaround, para_audioc works
1340 again.
1341 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
1342 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
1343 (new entry), "snp" (set numplayed). Read the manual for more
1344 information.
1345
1346 ---------------------------------------
1347 0.2.1 (2005-08-15) "surreal experience"
1348 ---------------------------------------
1349
1350 Here comes paraslash-0.2.1. It contains a couple of new features and,
1351 surprise, only minor bug fixes.
1352
1353 - kill noisy mp3 debug message
1354 - cleanup of the build system
1355 - para_server and para_client directly use the crypto routines
1356 of the openssl library rather than invoking the openssl command
1357 line utitlity
1358 - server/audiod: new option --user to switch to the given user
1359 when invoked as root. Read the output of "para_server -h" for
1360 more information.
1361 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
1362 status. Default: "para_audioc stat"
1363 - sdl_gui: new option --pic_cmd to be used to download the picture.
1364 Default: "para_client pic"
1365 - audiod: 5 slots ought to be enough for everybody
1366 - audiod: new status item: Uptime, kill hup command
1367
1368 ------------------------------------------
1369 0.2.0 (2005-08-06) "distributed diffusion"
1370 ------------------------------------------
1371
1372 After several month of increased development activity, paraslash-0.2.0
1373 has arrived. It contains many new features and is much more
1374 self-contained than the old 0.1.x series. Enjoy!
1375
1376 - para_server: fix hang on song change and crash on sighup.
1377 Speed up mysql queries. The DIR_LIKE macro is gone.
1378 - new executables: para_audiod, the local audio daemon that
1379 starts playback (uses SCM_CREDENTIALS socket magic) and
1380 para_audioc, the corresponding client.
1381 - new executables: para_mp3dec/para_oggdec, two really teensy
1382 decoders. para_mp3dec is based on libmad, para_oggdec requires
1383 libvorbisfile.
1384 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
1385 they are now called para_send and para_recv respectively.
1386 - documentation updates
1387 - para_gui is themable. For now there is the default theme that
1388 looks as before and the simple theme: blue and easy.
1389 - gui: audio streaming is now handled by audiod. Time display shows
1390 playback time rather than streaming time
1391 - slider: update to libzmw-0.2.0
1392 - para_krell: fix crash on server shutdown
1393 - switch from gzip to bzip2
1394
1395 ----------------------------------------
1396 0.1.7 (2005-04-18) "melting penetration"
1397 ----------------------------------------
1398
1399 The main change in this release is clearly the oggvorbis rewrite,
1400 but there are also lots of smaller changes. If you intend to use both
1401 the mp3 and the ogg plugin, it is recommended to use software mixing,
1402 e.g. the dmix plugin which is provided by ALSA.
1403
1404 - new executables: para_ovsend and para_ovrecv for sending/receiving
1405 oggvorbis files via rtp. Requires the open rtp library. Get it at
1406 http://www.linphone.org/ortp/
1407 - rewrite of the ogg_vorbis core code
1408 - configure detects libzmw and, if detected, includes
1409 para_slider to the list of binaries to be built by make
1410 - server stream writers read from their associated fifo rather
1411 than from stdin
1412 - slider: two new sliders, lastplayed and numplayed
1413 - fix nasty double free bug which caused random segfaults in case of
1414 mp3 files with invalid header information
1415 - gui: new command line option: --stream_timeout=seconds to
1416 deactivate a slot if it is idle for that many seconds (default=`5')
1417 - diffstats
1418
1419 ---------------------------------------
1420 0.1.6 (2005-03-05) "asymptotic balance"
1421 ---------------------------------------
1422
1423 Only little user-visible changes in this release. Mainly bugfixes and
1424 core code cleanup. This is probably the most stable version ever if you
1425 stick to mp3...
1426
1427 - fix several memory leaks
1428 - rename default name of mysql database from "music" to "paraslash".
1429 Use para_server's --mysql_database option if you do not want to
1430 rename your old database.
1431 - rework ogg vorbis code
1432 - make update command work on mysql servers with LOCAL_INFILE
1433 disabled
1434 - gui: improved stream I/O (slots)
1435 - simplified audio format API
1436 - para_pob_ogg is gone
1437
1438 ------------------------------------
1439 0.1.5 (2004-12-31) "opaque eternity"
1440 ------------------------------------
1441
1442 Let's slide gently into the new year.
1443
1444 - new: para_slider (not built automatically, type "make
1445 para_slider" to build). A toy for those who always felt that
1446 creating stream definitions is difficult. See screenshots,
1447 README and FEATURES for more info.
1448 - improved signal handling. Fixes server segfault on SIGHUP
1449 for linux kernels newer than Aug 24 2004 and makes para_gui
1450 race-free.
1451 - reload database tool on SIGHUP
1452 - improved help message for sl
1453 - do not log "broken pipe" messages as errors. They are
1454 perfectly ok.
1455 - fix wrong error message on permission errors
1456
1457 -----------------------------------------
1458 0.1.4 (2004-12-19) "tunneling transition"
1459 -----------------------------------------
1460
1461 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
1462 Hopefully, most of them got wiped out with this release. Some
1463 enhancements went also in.
1464
1465 - improved error diagnostics for all commands
1466 - stradd/picadd: overwrite previous contents if entry already
1467 exists, rather than returning errors
1468 - stradd: use current stream if invoked without args
1469 - faster (and hopefully more stable) ogg-vorbis handling
1470 - para_krell: reap children to avoid zombie-flooding in case
1471 no server is running
1472 - si: report also server pid
1473 - server: don't busy-loop if dbtool reports only invalid files.
1474 - gui: CTRL+C works again, fix stream_read command line option
1475 - fix pic_add, hist
1476 - fix mysql dbtool startup in case no database exists
1477 - many small fixes and cleanups
1478
1479 ---------------------------------------
1480 0.1.3: (2004-12-10) "vanishing inertia"
1481 ---------------------------------------
1482
1483 Starting from this release, the database tools are integrated in the
1484 server binary. This decreases server startup time, reduces code size
1485 and speeds up database commands. However, the layout of the underlying
1486 mysql database changed only slightly and 0.1.3 should be backwards
1487 compatible in that respect.
1488
1489 Visible changes:
1490
1491 - If mysql is not detected at compile time, or fails to init
1492 at runtime, fall back to the dopey database tool which should
1493 always work.
1494 - para_dbtool and dbtool.conf are gone. All mysql specific
1495 options are read from server.conf and are prefixed by 'mysql_'.
1496 - new command: cdt (change database tool)
1497 - new command line option: dbtool (choose startup database tool)
1498 - The name of current stream is now stored in the database,
1499 so paraslash remembers its current stream when restarted.
1500 - new command: csp (change stream and play)
1501 - para_gui also reports current database tool and server uptime
1502
1503 -------------------------------------------
1504 0.1.2: (2004-11-28) "spherical fluctuation"
1505 -------------------------------------------
1506
1507 Point release before the big dbtool changes go in.
1508
1509 - dbtool: rename ca to cam (copy all meta data). It now also
1510 copies numplayed and lastplayed time as well as the picture
1511 id.
1512 - fix endless-loop-bug caused by mp3 files with invalid header
1513
1514 -----------------------------------------
1515 0.1.1: (2004-11-05) "floating atmosphere"
1516 -----------------------------------------
1517
1518 - gkrellm plugin
1519 - new dbtool command: mbox. Browse your sound-file collection
1520 with your favorite mail reader.
1521 - several small fixes
1522
1523 -------------------------------------
1524 0.1.0: (2004-10-22) "rotating cortex"
1525 -------------------------------------
1526
1527 - fix logging bug for loglevel > VERBOSE
1528 - fix skip command
1529 - correct timings for vbr mp3s
1530 - modular audio format support
1531 - ogg-vorbis support (experimental)
1532 - new server option: autoplay
1533
1534 -----------------------------------------
1535 0.0.99: (2004-07-25) "harmonic deviation"
1536 -----------------------------------------
1537
1538 - rename projectname from icc to paraslash (play, archive, rate
1539 and stream large audio sets happily)
1540 - paraslash is no longer restricted to one particular audio
1541 streaming software
1542 - new dbtool commands (stradd, strq, strdel) for easy stream
1543 managment w/o configuration file. That obsoletes stream_defs
1544 file/config option for dbtool.
1545 - picadd accepts jpeg data from stdin
1546 - new server commands: ps (select previous stream), sc (song change)
1547 - new default pictures for sdl_gui
1548 - gui: new key_map option for binding commands and internal
1549 functions to arbitrary keys, nice help screen, rip out
1550 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1551 show silly logo on startup
1552 - new executables: para_fade for fading volume, para_dbadm for
1553 manipulating attributes
1554 - cdb adds _all_ tables to mysql database
1555 - revised and beautified documentation
1556 - sample dbtool rewritten in C
1557 - autoconf
1558
1559 ---------------------------------------------
1560 0.0.98: (2003-12-26) "incremental smoothness"
1561 ---------------------------------------------
1562
1563 - kick icecast in favour of poc. That removes some races and reduces
1564 core code considerably.
1565 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1566 give finer info on afs' status.
1567 - gui can start decoder (see config options). Further new gui
1568 commands: refresh (^L), jmp (F1-F10)
1569 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1570 terminates gui. This fixes dead instances consuming memory
1571 continuously.
1572 - new dbtool command: verb for sending verbatim sql queries.
1573 - fix pid_list races (by removing pid_list)
1574 - codename funnies
1575
1576 --------------------
1577 0.0.97: (2003-10-26)
1578 --------------------
1579
1580 - installation prefix now defaults to /usr/local
1581 - new commands for gui: snozze, sleep and reread config
1582 - config file for gui and sdl_gui
1583 - fix problems with filenames containing funny characters
1584 (reported by Thomas Forell)
1585 - improved signal handling for gui, now it rereads conf on SIGHUP
1586 - new dbtool command: cdb (create database)
1587 - switch from argtable to gengetopt
1588 - major code cleanup and speed improvements
1589 - fix several potential buffer overflows
1590 - many small fixes and cleanups
1591
1592 -------------------
1593 0.0.96 (2003-08-30)
1594 -------------------
1595
1596 - easy stream_defs syntax
1597 - sdl_gui can display images associated to the file being played
1598 - Major feature enhancements for icc_gui including dynamic text
1599 placement and the top/bottom window design
1600 - vrfy/clean now also checks for NULL values in attributes as
1601 well as for invalid picture pointers
1602 - fix long outstanding case sensitivity bug
1603 - many small fixes and cleanups
1604
1605 -------------------
1606 0.0.95 (2003-06-29)
1607 -------------------
1608
1609 - sdl gui runs much faster
1610 - new dbtool command: ca (copy attributes)
1611 - count and display number of times the song has been played
1612 - new feature: scoring
1613 - command line options for sdl_gui
1614 - simpler syntax of streams file
1615 - decrease network traffic of stat
1616 - fix zombie bug
1617 - many small fixes and cleanups
1618
1619 -------------------
1620 0.0.94 (2003-05-04)
1621 -------------------
1622
1623 - new server command: ns (next stream)
1624 - new icc_gui command: c (change stream)
1625 - internal mp3info
1626 - stat shows also id3 tag info
1627 - new sdl based gui
1628 - log flodding bug fixed
1629 - many small fixes and cleanups
1630
1631 -------------------
1632 0.0.93 (2003-03-28)
1633 -------------------
1634
1635 - colors for icc_gui
1636 - icc_gui sets volume directly (linux only)
1637 - proper locking that fixes some races
1638 - fix security bug that caused commands to be executed even
1639 with unsufficient permissions
1640 - new command: hup to make all servers reread their configuration file
1641 - icecast meta data streaming
1642 - many small fixes and cleanups