4 ----------------------------------------------
5 0.7.3 (to be announced) "weighted correctness"
6 ----------------------------------------------
8 - Old style PEM keys are now deprecated. They still work but their
9 use results in a run-time warning. The removal of PEM key support is
10 scheduled for paraslash-0.8.0.
11 - Version 1.0 of the openssl library has been deprecated. A warning
12 is printed at compile-time on systems which have this outdated version
13 because it will no longer be supported once paraslash-0.8.0 comes out.
14 - A spring cleanup for the senescent code in fd.c.
17 [tarball](./releases/paraslash-git.tar.xz)
19 -------------------------------------
20 0.7.2 (2023-03-08) "optical friction"
21 -------------------------------------
23 The improved error reporting of afs commands and the two new options
24 for the sleep subcommand of para_mixer are the most prominent features
25 of this minor release. The bulk of the changes are cleanups of the
26 afs and net subsystems, which should both have no user-visible impact.
28 - A major cleanup of the audio file selector.
29 - The client no longer prints error messages from afs commands to
31 - The sleep subcommand of para_mixer gained two options to control
32 the startup mood and the time period before fade-out starts. A bunch
33 of further improvements for this subcommand went in as well.
34 - Minor cleanup of the net subsystem.
35 - The openssl specific code now employs the EVP API to compute hashes.
36 It should compile without warnings against openssl-3.
37 - The deprecated syntax for specifying negative offsets in the argument
38 to the "ff" server command has been removed.
41 [tarball](./releases/paraslash-0.7.2.tar.xz),
42 [signature](./releases/paraslash-0.7.2.tar.xz.asc)
44 --------------------------------------
45 0.7.1 (2022-10-03) "digital spindrift"
46 --------------------------------------
48 The two new ll commands and the internal mp4ff library are the most
49 user-visible changes of this release. On top of that there are two
50 core changes which aim to improve the robustness of the code but which
51 are otherwise invisible: the switch from select(2) to poll(2) and the
52 revised memory allocation API which checks for integer overflows. The
53 release also comes with a slight change to the build system and the
54 usual mix of bug fixes and minor improvements not mentioned here.
56 - The autogen.sh script now only creates the autoconf specific files
57 but no longer runs configure, make and the test suite.
58 - A stripped down copy of the discontinued libmp4ff library has become
59 part of the paraslash code base. As a result it is no longer necessary
60 to install faad from source to get support for aac/m4a files. The
61 faad decoder package must still be installed.
62 - The log level of the running daemon can now be changed with the
63 new ll command. It is available for para_server and para_audiod.
64 - All calls to select(2) have been replaced by calls to poll(2)
65 to avoid known shortcomings of the select API.
66 - All allocation functions now check for integer overflow. Since this
67 requires support from the compiler, the oldest supported gcc version
68 has been bumped to gcc-5.4 (released in 2015).
71 [tarball](./releases/paraslash-0.7.1.tar.xz),
72 [signature](./releases/paraslash-0.7.1.tar.xz.asc)
74 ----------------------------------
75 0.7.0 (2022-03-12) "seismic orbit"
76 ----------------------------------
78 The major incompatible change which requires to bump the major version
79 is the switch from sha1 to sha256, see below for details. However,
80 there are many other improvements, the usual amount of bug fixes and
81 a couple of new features.
83 - Starting with paraslash-0.7.0, the sha256 hash value of each known
84 audio file is stored in the database while older versions employed the
85 sha1 hash algorithm which has been considered insecure since 2005
86 and should no longer be used today. The switch from sha1 to sha256
87 requires users to upgrade their database using the new para_upgrade_db
88 program, followed by re-adding all files to recompute the hashes. With
89 this approach all metadata stored in the database (last played date,
90 num played value, moods, playlists, attributes etc.) are kept. An
91 simpler alternative is to start over from scratch by running the
92 "init" command but this will lose these metadata.
93 - Server and client now hash the session keys with sha256 rather
94 than sha1 during the initial handshake. This feature is optional and
95 backwards compatible: old clients can still connect to a new server
96 (using sha1). Also new clients can connect to an old server (again
98 - The new "duration" keyword of the mood grammar makes it possible to
99 impose a constraint on the duration of the admissible files.
100 - The long deprecated version 1 mood syntax is no longer supported.
101 - Paraslash writers handle early end-of-file more gracefully.
102 - The alsa writer no longer warns about spurious underruns.
103 - The score formula of the audio file selector has been reworked.
104 - Cleanups of the doubly linked lists code.
105 - New option for configure: --enable-ubsan to detect and report undefined
107 - The "tasks" server command has been removed.
108 - The fancy new logo and a minor overhaul of the web pages.
111 [tarball](./releases/paraslash-0.7.0.tar.xz),
112 [signature](./releases/paraslash-0.7.0.tar.xz.asc)
114 --------------------------------------
115 0.6.4 (2021-11-04) "fuzzy calibration"
116 --------------------------------------
118 This point release contains a fair number of fixes but no new features.
119 This marks the end of the 0.6 development, although paraslash-0.6 will
120 still be supported for some time and subsequent maintenance releases
123 - The udp sender no longer crashes when empty chunks are encountered.
124 - Fix a double-free bug in the exit path of the server.
125 - The "jmp" command now errors out when given a negative percentage.
126 - A fix for a bug in para_afh which triggered on the attempt to modify
127 the tags of an invalid mp4 file.
128 - A memory leak in para_afh has been fixed.
129 - The udp sender no longer sends multiple EOF packets.
130 - Some gcc warnings have been silenced.
131 - Minor log level adjustments and documentation improvements.
134 [tarball](./releases/paraslash-0.6.4.tar.xz),
135 [signature](./releases/paraslash-0.6.4.tar.xz.asc)
137 ---------------------------------------
138 0.5.9 (2021-11-04) "reversed dimension"
139 ---------------------------------------
141 This release contains a few important fixes which have accumulated in
142 the maint branch. The paraslash-0.5.x series has now reached its end
143 of life and will no longer be supported. All users should upgrade to
144 a more recent version at this point.
146 - Fix an issue with the bash completion script.
147 - Initialize the random seed also when using libgrypt.
148 - Fix some compiler warnings in the resample filter
149 - Don't return spurious errors from the ff server command.
152 [tarball](./releases/paraslash-0.5.9.tar.bz2),
153 [signature](./releases/paraslash-0.5.9.tar.bz2.asc)
155 ----------------------------------------
156 0.6.3 (2021-02-18) "generalized activity"
157 -----------------------------------------
159 It has been a while since the last release, but paraslash-0.6.3
160 finally has arrived. It contains a fair number of improvements,
161 features and fixes all over the place.
163 - The ff command now accepts a negative argument to instruct the
164 virtual streaming system to jump backwards in the current audio
165 stream. The old syntax (e.g., "ff 30-") is still supported but it
166 is deprecated and no longer documented. The compatibility code is
167 sheduled for removal after 0.7.0.
168 - para_afh: New option: --preserve to reset the modification time to
169 the value of the original file after meta data modification.
170 - Overhaul of the compress filter code. The refined algorithm should
171 reduce clipping. The meaning of --aggressiveness has changed, see the
172 updated and extended documentation of the compress filter for details.
173 - Cleanup of the audio format handler code.
174 - We now build the tree using the .ONESHELL feature of GNU make,
175 which results in a significant speedup.
176 - Two robustness fixes for FreeBSD.
177 - para_client now supports RFC4716 private keys as generated with
178 ssh-keygen -m RFC4716. In fact, this key format has been made the
179 default, and the former PEM keys will be depreciated at some point.
180 - The ogg audio format handlers learned to detect holes and now report
181 the correct duration also if ogg pages are missing in the file. This
182 affects ogg/vorbis ogg/speex and ogg/opus.
183 - Robustness improvements for para_mixer.
184 - A fix for an old bug that could cause the server to crash or report
185 garbage in its status output.
186 - New para_play option: --end-of-playlist
187 - Streaming m4a files over udp has been improved.
190 [tarball](./releases/paraslash-0.6.3.tar.xz),
191 [signature](./releases/paraslash-0.6.3.tar.xz.asc)
193 --------------------------------------
194 0.6.2 (2018-06-30) "elastic diversity"
195 --------------------------------------
197 The main new feature of this release is the --listen-address option
198 for para_server and para_audiod. But there are also a large number of
199 small improvements and bug fixes all over the place. The user-visible
202 - para_gui no longer waits up to one second to update the screen when
203 the geometry of the terminal changes.
204 - Minor documentation improvements.
205 - Improvements to the crypto subsystem.
206 - The server subcommand "task" has been deprecated. It still works,
207 but prints nothing. It will be removed in the next major release.
208 - Server log output is now serialized, avoiding issues with partial
210 - It is now possible to switch to a different afs database by changing
211 the server configuration and sending SIGHUP to the server process.
212 - New server options: --listen--address, --http-listen-address and
213 --dccp-listen-address. These options restrict the set of listening
214 addresses of the TCP and DCCP sockets of the server process.
215 - The help commands of server, audiod, play now support --long. By default,
216 the short help is shown.
217 - The code which merges the command line options with the config file
218 options has been consolidated.
219 - If the current audio file is renamed, the status items are now updated
221 - After the server process received SIGHUP, changes to the current audio
222 file did not trigger an update of the status items. This has been fixed.
225 [tarball](./releases/paraslash-0.6.2.tar.xz),
226 [signature](./releases/paraslash-0.6.2.tar.xz.asc)
228 ----------------------------------------
229 0.6.1 (2017-09-23) "segmented iteration"
230 ----------------------------------------
232 The highlight of this release is the version 2 mood parser. But there
233 is a lot more than that, as summarized in the list below. And of
234 course we have many small usability improvements and bug fixes not
237 - A more intuitive syntax for moods ("version 2 moods"). The
238 traditional version 1 moods are still supported but are deprecated
239 now. Removal of the version 1 mood parser is scheduled for the next
241 - Flex and bison are now required to build para_server.
242 - New sort order for the ls command: -s=h sorts the ls output by hash
243 value of the audio file.
244 - autogen.sh now runs the test suite after a successful build.
245 - The contents of overview.pdf have been integrated into the user
247 - Fixed sized audio format headers for ogg/opus streams.
248 - The doxygen source browser has been disabled temporarily. The
249 API reference is still online, though.
250 - Overhaul of the source code documentation.
251 - The deprecated --full-path option of the ls command has been
252 removed. It was a no-op since 0.6.0.
253 - The wma decoder has been cleaned up and its bitstream API made
255 - The image/lyrics ID status items of the current audio file are now
256 updated on changes. This affects para_gui, which used to report the
260 [tarball](./releases/paraslash-0.6.1.tar.xz),
261 [signature](./releases/paraslash-0.6.1.tar.xz.asc)
263 ---------------------------------------
264 0.5.9 (2021-11-04) "reversed dimension"
265 ---------------------------------------
266 This release contains a few important fixes which have accumulated in
267 the maint branch. The paraslash-0.5.x series has now reached its end
268 of life and will no longer be supported. All users should upgrade to
269 a more recent version at this point.
271 - Fix an issue with the bash completion script.
272 - Initialize the random seed also when using libgrypt.
273 - Fix some compiler warnings in the resample filter
274 - Don't return spurious errors from the ff server command.
277 [tarball](./releases/paraslash-0.5.9.tar.bz2),
278 [signature](./releases/paraslash-0.5.9.tar.bz2.asc)
280 ---------------------------------------
281 0.5.8 (2017-09-23) "branching parabola"
282 ---------------------------------------
284 A couple of fixes for serious old bugs have accummulated in the maint
285 branch since paraslash-0.5.7 was released nine months ago. So here
286 is another 0.5.x release. As usual, all commits in the maint branch
287 have been merged into master, so the upcoming paraslash-0.6.1 also
288 contains them. For details see the git log.
291 [tarball](./releases/paraslash-0.5.8.tar.bz2),
292 [signature](./releases/paraslash-0.5.8.tar.bz2.asc)
294 -------------------------------
295 0.6.0 (2017-04-28) "fuzzy flux"
296 -------------------------------
298 The highlights of this release are the conversion of all option parsers
299 to the lopsub library and the improved AAC support. But there are
300 many other user-visible changes all over the place. Also a lot of
301 old cruft has been removed, leading to incompatible changes. Hence
302 the new major version number.
304 - Support for Mac OS X has been removed.
305 - On Linux systems, glibc-2.17 or newer is required to build the
307 - Support for RSA public keys in ASN format (as generated by openssl
308 genrsa) has been removed. These keys have been deprecated since
309 2011, so users should have long switched to keys generated with
311 - If libgcrypt is used as the crypto library, we now require version
312 1.5.0 (released in 2011) or later.
313 - The insecure RC4 stream cipher has been removed. It was superseded
314 by aes_ctr128 three years ago but the RC4 code had been kept for
315 backwards compatibility.
316 - On Linux, abstract unix domain sockets are used unconditionally.
317 - The "install" target no longer strips executables, the new
318 install-strip target can be used to get the old behaviour.
319 - The clean targets have been renamed: clean2 is gone, and the new
320 mostlyclean removes only the executables and object files.
321 - New target: check (identical to test).
322 - The DESTDIR make variable is honored to prepend a path to the
323 installation directory. This feature is orthogonal to the --prefix
325 - Minor WMA cleanups.
326 - The aac audio format handler has been rewritten to use the mp4ff library.
327 See the manual for how to install the library on your system.
328 - New status item: max_chunk_size. The value is stored in a previously
329 unused field of the afhi object of the aft table. Although backwards
330 compatible, users are encouraged to re-add m4a files to populate
332 - No more chunk tables for aac. Chunk boundaries are determined
333 dynamically at stream time.
334 - Release and master branch tarballs are now compressed with xz rather
336 - The lopsub package is required to install the paraslash package.
337 Gengetopt is no longer needed.
338 - make dep is gone. Dependencies have been created automatically for
339 a long time and it was never necessary to run make dep manually.
340 - para_gui lost its --timeout option.
341 - Most manual pages have been extended to include an overall
342 description of the command.
343 - The --stream-delay option of para_audiod has been removed. It had
344 been a no-op for many years.
345 - The deprecated --path option of the server ls command has been
346 removed. The command now prints full paths by default, making
347 --full-path a no-op. Hence --full-path has been depreacted and is
348 scheduled for removal in v0.6.1.
349 - It is now possible to use 'CFLAGS' to override the default compiler
351 - para_fade has been renamed to para_mixer. The four modes of operation
352 (set, fade, snooze, sleep) are implemented as subcommands of the
354 - The sleep subcommand of para_mixer (former sleep mode of para_fade)
355 lost its --wake-hour and --wake-min options in favor of the new
356 --wake-time option which also accepts relative wakeup times like
358 - The new --fade-exponent option of para_mixer allows for non-linear
363 [tarball](./releases/paraslash-0.6.0.tar.xz),
364 [signature](./releases/paraslash-0.6.0.tar.xz.asc)
366 -------------------------------------
367 0.5.7 (2016-12-31) "semantic density"
368 -------------------------------------
370 Mostly a bug fix release, and a bunch of internal improvements.
371 The only user-visible changes are the sanity checks for the touch
372 command and the new options to the ls command.
374 - Speedup of the base64 decoder.
375 - One of the two source browsers has been removed from the web pages.
376 The doxygen API reference still contains an HTML version of each
378 - Two race conditions in para_server have been fixed.
379 - ls -p is now deprecated in favor of -F or -b. See the help text of
380 the ls command for details.
381 - The openssl code has been adjusted to work also with openssl-1.1.
382 - The wma decoder and audio format handler now correctly decodes
383 files with unusual block sizes.
384 - We now compile with -Wformat-signedness if possible.
385 - The touch command now refuses to set an invalid image or lyrics ID.
386 - New section on contributing for the user manual.
387 - Major simplification of the error subsystem.
390 [tarball](./releases/paraslash-0.5.7.tar.bz2),
391 [signature](./releases/paraslash-0.5.7.tar.bz2.asc)
393 -------------------------------------------
394 0.4.14 (2016-12-31) "branching oscillation"
395 -------------------------------------------
397 Many fixes for nasty bugs have accumulated over the years since 0.4.13
398 was released in 2013. This release contains all of them. It is the
399 last v0.4.x release, so v0.4.x is now end of life. The maint branch
400 of the git repository has been forwarded to v0.5.7 and will start to
401 receive important fixes for 0.5.x, which is now in maintenance mode.
403 - Fix for the "double encryption bug" with gcrypt, which caused
404 garbage to be sent to stat clients.
405 - The "prev" command of para_play now works as expected.
406 - Invalid parameters for the begin and end chunk no longer cause
407 the afh receiver of para_recv to segfault.
408 - A fix for an assertion that triggered in the FEC slice setup code.
409 - Closing the terminal with a running para_gui causes the process to
410 die, rather than using 100% of CPU time.
411 - para_gui no longer shows stale information about the previous
413 - A fix for a bug in the fec decoder which caused aborts if the buffer
415 - The assertion in the buffer tree code no longer triggers.
416 - The "afs_event: table moods, event 0: key already exists in rbtree"
417 error no longer occurs on attribute changes.
418 - A NULL pointer dereference in com_sender() has been fixed.
419 - The -i option of lsatt works as advertised.
420 - Various fixes for the build system.
423 [tarball](./releases/paraslash-0.4.14.tar.bz2),
424 [signature](./releases/paraslash-0.4.14.tar.bz2.asc)
426 ---------------------------------------
427 0.5.6 (2016-07-10) "cascading gradient"
428 ---------------------------------------
430 The highlight of this release is the new -m flag for para_afh which
431 lets it modify the meta tags of the given audio file(s). This feature
432 is supported for all audio formats. Many small cleanups and bug fixes
433 not mentioned here have accumulated and are also part of the release.
435 - para_afh learned to modify meta tags of mp3 wma ogg spx
437 - afs commands propagate error codes to the client.
438 - The check command now also checks the attribute table for
440 - New -v flag for the version command (print verbose version string)
441 - New option --priority for para_server and para_audiod.
442 - New mood methods: image_id and lyrics_id.
443 - The manual and this NEWS file have been converted to markdown.
444 - Support for the compile-time loglevel feature has been removed.
445 - Cleanup of the wma decoder and bitstream code.
446 - Improved wide-character support and fixes related to signal handling.
447 - para_gui no longer reports 100% playing time at the stream start.
451 [tarball](./releases/paraslash-0.5.6.tar.bz2),
452 [signature](./releases/paraslash-0.5.6.tar.bz2.asc)
454 --------------------------------------
455 0.5.5 (2015-09-20) "magnetic momentum"
456 --------------------------------------
458 Many new features and a lot of other improvements.
460 - On Linux systems, local sockets are now created in the
461 abstract name space by default. This allows to get rid of
462 the socket specials in /var/paraslash.
463 - The --user-allow option of para_audiod now accepts also
464 usernames rather than only user IDs.
465 - New autoconf macros to avoid duplication in configure.ac.
466 - Status items (as shown by para_gui) are updated correctly
467 when the meta information of the current audio changes.
468 - para_server and para_audiod no longer refuse to start in
469 the background if no log file is given. Instead, all log
470 messages go to /dev/null in this case.
472 - New syntax for the -l and -s options of the ls command.
473 These options should now be specified as -l=v rather than
474 -lv, for example. The old syntax still works, but support
475 will be dropped in v0.6.0.
478 [tarball](./releases/paraslash-0.5.5.tar.bz2),
479 [signature](./releases/paraslash-0.5.5.tar.bz2.asc)
481 ------------------------------------------
482 0.5.4 (2015-01-23) "exponential alignment"
483 ------------------------------------------
485 Another cleanup and bugfix release.
487 - New server command: tasks.
488 - Minor cleanups to daemon.c.
489 - New URLs for home page and git services.
490 - Improved error diagnostics for the mvblob commands.
491 - New sender subcommand: status.
492 - Improved help text for server and afs commands.
493 - audiod memory leak fixes.
494 - Miscellaneous improvements to the build system.
495 - oss_writer improvements.
496 - Improved handling of mp3 files with both id3v1 and id3v2 tags.
499 [tarball](./releases/paraslash-0.5.4.tar.bz2),
500 [signature](./releases/paraslash-0.5.4.tar.bz2.asc)
502 ---------------------------------------------
503 0.5.3 (2014-08-01) "symbolic synchronization"
504 ---------------------------------------------
506 Not many new features, but lots of fixes and usability improvements.
508 - para_gui has been converted to use the paraslash scheduler.
509 - Various alsa-related fixes, mostly for the raspberry pi.
510 - Many scheduler improvements and cleanups.
511 - The test suite has been extended to include sanity checks
512 for the generated man pages.
513 - ao_writer fixes. This writer was in a quite bad shape. Many
514 serious bugs have been fixed.
515 - new audiod command: version.
516 - Minor improvements to the bitstream API.
517 - The cpsi command now prints a meaningful error message if
518 none of the given patterns matched any audio file.
521 [tarball](./releases/paraslash-0.5.3.tar.bz2),
522 [signature](./releases/paraslash-0.5.3.tar.bz2.asc)
524 ----------------------------------------
525 0.5.2 (2014-04-11) "orthogonal interior"
526 ----------------------------------------
528 The new sync filter, the AES_CTR128 stream cipher and the overhauled
529 network code are the highlights of this release. It also includes a
530 fair number of smaller fixes and improvements not mentioned here.
532 - The new sync filter synchronizes playback between multiple
534 - Connections between para_server and para_client are now
535 encrypted by means of AES rather than RC4 if both sides
536 support it. RC4 is still available as a fallback. This
537 feature is fully transparent, i.e. no command line options
538 are necessary, and a client linked against openssl can
539 speak with a server linked against libgcrypt and vice versa.
540 - Major cleanup of the networking subsystem.
541 - Improvements to para_fade: the new set mode, multi-channel
542 initial volumes, better error logging.
543 - The man pages of para_audiod, para_filter, para_recv, and
544 para_write contain the relevant options for receivers, filters,
545 writers. This broke in 0.5.0.
546 - ogg/vorbis latency improvements.
547 - Improved user manual.
548 - Minor fixes to avoid clang warnings.
551 [tarball](./releases/paraslash-0.5.2.tar.bz2),
552 [signature](./releases/paraslash-0.5.2.tar.bz2.asc)
554 ------------------------------------------
555 0.5.1 (2013-12-20) "temporary implication"
556 ------------------------------------------
558 Lots of fixes and improvements all over the place, and a major overhaul
561 - Audiod improvements and fixes.
562 - Buffer tree robustness improvements.
563 - Cleanup of the mood subsystem.
564 - Fixes and cleanups for the flac decoder.
565 - Latency improvements for the ogg/opus decoder.
566 - Crypto support is now optional. On systems without
567 openssl/gcrypt, the build succeeds but para_server,
568 para_audiod, para_client won't be built.
569 - The build system now works for cross-compile setups.
570 - The dependency tree has been flattened, which speeds up
571 builds and avoids to recreate the man pages on every change.
572 - The error code helper has been rewritten from perl to C,
573 which further improves build time.
574 - Many small bugs in the build system have been identified
578 [tarball](./releases/paraslash-0.5.1.tar.bz2),
579 [signature](./releases/paraslash-0.5.1.tar.bz2.asc)
581 ----------------------------------------
582 0.5.0 (2013-08-23) "invertible validity"
583 ----------------------------------------
585 Some API-breaking changes, one serious bug fix, and a lot of bike-shedding.
587 - The sideband compatibility code has been removed, hence
588 sideband connections (introduced in 0.4.11) are now mandatory.
589 - Addblob commands can produce output.
590 - The stat command no longer sends garbage when para_server was
591 compiled against libgcrypt.
592 - Dependencies for gengetopt files are computed automatically.
593 This eliminates a constant source of build bugs.
594 - The setatt command now accepts file name patterns rather than only
596 - overview.pdf is now based on dia, a simple diagram creation program.
597 The new version is much more detailed and contains descriptions of
598 the various programs of the paraslash package.
599 - The separator of all multi-word options has been changed from
600 underscore to dash. For example --log_color becomes --log-color.
601 - Overhauled web pages and the new logo.
604 [tarball](./releases/paraslash-0.5.0.tar.bz2),
605 [signature](./releases/paraslash-0.5.0.tar.bz2.asc)
607 --------------------------------------
608 0.4.13 (2013-07-29) "spectral gravity"
609 --------------------------------------
611 One more 0.4.x release before the API-breaking changes for 0.5.0 go
612 in. The main features of this release are the ogg/opus audio format,
613 and UTF-8 support, but it includes also tons of other improvements
614 and fixes all over the place.
616 - New audio format: ogg/opus.
617 - UTF8 support for para_gui and the mp3 audio format handler.
618 - Scheduler improvements and fixes.
619 - The obsolete gettimeofday() function has been replaced
620 by clock_gettime() on systems which support it.
621 - Speed and usability improvements for para_gui.
622 - para_client now restores the fd flags of stdin and stdout
624 - Improved manual pages.
625 - Consistent version strings for all executables.
626 - Reduced dependencies on generated files result in fewer
627 recompilations on changes.
628 - Performance improvements for the compress filter.
629 - Improved downloads web page.
631 -----------------------------------------
632 0.4.12 (2012-12-20) "volatile relativity"
633 -----------------------------------------
634 The new command line player, the resample filter, ALSA support for
635 para_fade, and the improved build system are the highlights of this
636 release which probably marks the end of the 0.4.x series.
638 - The afh receiver and the para_play executable.
639 - The resample filter: A sample rate converter based on
641 - The "versions" directory has been removed from the master
642 branch. The tarballs of the old releases are now available
643 in the new "releases" branch.
644 - Overhaul of the build system: All generated files are now
645 written to the "build" directory.
646 - The modular mixer API and the alsa mixer.
647 - Minor fixes for the osx writer.
649 --------------------------------------
650 0.4.11 (2012-07-20) "mutual diversity"
651 --------------------------------------
653 The major feature in this release is the new sideband API for
654 client-server communication. This API will be used exclusively starting
655 with 0.5.0, which breaks backward compatibility but allows to get rid
656 of quite some compatibility code. Other noteworthy changes include
657 decoder latency improvements and a long-standing bug fix for the
660 - Sideband connections: If both para_server and para_client
661 support this feature, data is sent as a multiplexed stream.
662 - The --no_default_filters option of para_filter has been
664 - Several fixes and latency improvements to various decoders.
665 - The ALSA writer now limits the prebuffer time to 500ms.
666 - Documentation improvements.
667 - Overhaul of the command_util.sh script.
668 - Fixes for some minor problems found by the clang analyzer.
669 - Compiles (almost) without warnings on gcc-3.
670 - Robustness improvements of the buffer tree code.
672 ------------------------------------------
673 0.4.10 (2012-03-30) "heterogeneous vacuum"
674 ------------------------------------------
676 Nothing earth-shaking in this release, but quite a few usability
677 improvements and the usual mix of cleanups and fixes.
679 - The --no_default_filters option of para_filter has been
680 deprecated. It still works but has no effect and will be
681 removed in the next version.
682 - para_gui now prints also the stderr output of the executing
683 command in the bottom window.
684 - Cleanup and consolidation of the various wrappers for
685 write(), writev(), send() and friends.
686 - The obscure error messages on mmap() failures have been
687 replaced by meaningful messages. This affects mainly
689 - para_audioc: Cleanups and memory leak fixes.
690 - Test 0004-server no longer fails if para_server is not
692 - New configure options: --with-id3tag-{headers,libs}.
694 -------------------------------------
695 0.4.9 (2011-12-06) "hybrid causality"
696 -------------------------------------
698 Support for another audio format, interactive mode for para_client
699 and para_audiod and many small improvements/fixes all over the place.
701 - Support for flac, the free lossless audio codec.
702 - Fix for an endless loop in the mp3 decoder for certain
704 - When executed without specifying a command, para_client
705 and para_audioc start an interactive shell (requires
706 libreadline being installed). The interactive mode offers
707 full tab completion and command line history.
708 - autogen.sh now detects a distcc setup and adjusts the
709 parameter for the -j option of make accordingly.
710 - Shared memory areas are no longer restricted to 64K. We now
711 detect the maximal size of a shared memory area at runtime.
712 - cleanup of the internal uptime API.
713 - para_server prefaults the mmapped audio file to avoid
714 delays on slow media.
715 - A new test for the test-suite that exercises the
716 communication between para_server and para_audiod.
717 - The alsa writer eats up less CPU cycles when configured to
719 - Simplified and unified receiver code.
721 - Commands which print a list of matching audio files now
722 emit a meaningful error message if no audio file matched the
725 --------------------------------------
726 0.4.8 (2011-08-19) "nested assignment"
727 --------------------------------------
729 Gcrypt support, the overhauled osx writer and regex format specifiers
730 are the highlights of this release.
732 - support for libgcrypt as a drop-in replacement for openssl.
733 Run configure --enable-cryptolib=gcrypt to link against
734 libgcrypt. The two crypto implementations are compatible to
735 each other, i.e. a para_client executable linked against
736 gcrypt can connect to para_server linked against libssl
738 - Rewrite of the osx writer (output plugin for Mac OS).
739 - audiod: The format specifier for receivers, filters and
740 writers is now treated as a regular expression. This allows
741 to replace 5 lines in the config file (one for each audio
742 format) by one single line. See the manual for details.
743 - The \*.cmdline.[ch] files are no longer contained in the released
744 tarballs. This reduces the size of the tarballs but requires
745 gengetopt to build the tarball.
746 - Compiles cleanly also with llvm/clang.
747 - Corrupt mp3 files are handled more gracefully.
748 - The alsa writer uses poll fds instead of computing timeouts.
749 - Cleanup of the generic writer API.
750 - sched: Optimized zero timeouts.
751 - vss timeout cleanups.
752 - oggdec fixes and improvements.
754 --------------------------------------
755 0.4.7 (2011-06-01) "infinite rollback"
756 --------------------------------------
758 The new ao writer, support for ssh RSA keys and a couple of other
761 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
762 platforms through the libao audio library.
763 - Support for RSA keys generated with ssh-keygen.
764 - configure: improved options for ogg/vorbis/speex.
765 - The git version reported by --version always matches HEAD.
766 - The autogen script detects the number of processors and
767 runs a parallel make if possible.
768 - Major cleanup of the crypto API.
769 - Documentation updates.
771 ------------------------------------------
772 0.4.6 (2011-03-31) "deterministic entropy"
773 ------------------------------------------
775 Lots of ogg/vorbis improvements, the new test suite, enhancements
776 for para_gui and a fair amount of other bug fixes.
778 - For DCCP/OGG streams the audio file header is only sent once
779 at the beginning of the stream rather than periodically
780 every five seconds. This reduces network traffic and the
782 - The vorbis comment header is replaced by an empty dummy header
783 before the header is sent over the network. This also results in
784 less network traffic and smaller FEC groups.
785 - The new "test" make target allows to perform some sanity checks prior
786 to installing the package.
787 - ogg timing fixes and performance improvements
788 - Scheduler improvements
789 - Proper exit codes for para_write
790 - para_gui: New option --theme to select a startup theme. Several
791 other improvements and fixes.
792 - aacdec error message cleanups
793 - simplified color error handling
795 --------------------------------------------
796 0.4.5 (2010-12-17) "symmetric randomization"
797 --------------------------------------------
799 Bug fixes, internal cleanups and variable-sized FEC slices.
801 - Contains a fix for an invalid-free-bug in the ogg audio format
803 - Switching off the DCCP sender works again.
804 - para_audiod handles crashes of para_server more robustly.
805 - Internal scheduler and writer cleanups.
806 - Reduced latency due to variable-sized FEC slices.
807 - Improved documentation and error diagnostics.
808 - The build of para_server is now optional, allowing the build
809 to succeed in case libosl is not installed.
811 ------------------------------------------
812 0.4.4 (2010-08-06) "persistent regularity"
813 ------------------------------------------
815 Support for yet another audio format, para_write improvements and
818 - Support for the speex codec.
819 - Support for sample formats other than 16 bit little endian.
820 - error2.h is now created by a perl script which speeds up configure
822 - Fix a bug in the aac decoder which could lead to segfaults in
823 para_filter/para_audiod.
824 - Fixes for autoconf-2.66.
826 ----------------------------------------
827 0.4.3 (2010-07-05) "imaginary radiation"
828 ----------------------------------------
830 Many improvements for the DCCP and the UDP transport, the new user
831 manual and the usual mix of bug fixes and internal improvements.
833 - FEC support for the DCCP sender (Gerrit Renker). The new
834 --dccp_max_slice_size, --dccp_data_slices_per_group and
835 --dccp_slices_per_group options can be used to set the FEC
836 parameters for the DCCP transport.
837 - DNS lookups for UDP targets (Gerrit Renker).
838 - The new user manual replaces the README, README.afs, REQUIREMENTS
839 and INSTALL documents.
840 - Fix an end-of-file detection bug in the oggdec filter.
841 - The new nonblock API.
842 - Both options of the oggdec filter have been removed.
843 - New debug mode for the internal scheduler.
845 ------------------------------------------
846 0.4.2 (2010-04-23) "associative expansion"
847 ------------------------------------------
849 It's been some time since the last release, but finally here is
850 paraslash-0.4.2. The bulk of the changes comes from the new buffer
851 tree API, but there are changes all over the tree. Mainly performance
852 and usability improvements, but also quite some bug fixes.
854 - The new buffer tree API.
855 - DCCP: Support for CCID negotiation (Gerrit Renker).
856 - UDP robustness fixes.
857 - The --bufsize option for mp3dec is gone as it no longer makes sense
858 for the new buffer tree API.
859 - Fix audible buffer underruns for wma streams.
860 - The alsa writer no longer prints meaningless underrun durations.
861 - audiod: Defaults work also for udp streams. If no filter is
862 given for an audio format that is received via upd, fecdec is
863 automatically added as the first filter (along with the decoder).
865 ---------------------------------------
866 0.4.1 (2009-12-22) "concurrent horizon"
867 ---------------------------------------
869 Support for another audio format, minor feature enhancements and lots of bug
870 fixes. All fixes that have been accumulated in the maint branch (in particular
871 those mentionened in the 0.3.6 release notes) appear in this release as well.
874 - new afh option: --human to activate human-readable output.
875 - new server/audiod option: --log-timing to print timing information.
876 - build system improvements.
877 - source code documentation updates.
879 -------------------------------------
880 0.3.6 (2009-12-07) "cubic continuity"
881 -------------------------------------
883 Quite a few bugs have been found and fixed since 0.3.5, so here's
884 another 0.3.x release. No new features.
886 - Always check return value of malloc().
887 - ogg vorbis/FEC: Do not write garbage after the audio file header.
888 - exit if root privileges could not be dropped.
889 - FEC: Fix computation of extra slices.
890 - oss: Fix check for empty input buffer.
891 - Avoid buffer underruns due to filter chain output buffer constraints.
892 - server: Fix assignment of afs_pid.
893 - Don't panic if the afs database contains unknown audio formats.
894 - http/dccp: Do not send the audio file header twice.
895 - FEC: Timing improvements.
897 ----------------------------------------------
898 0.4.0 (2009-11-10) "simultaneous independence"
899 ----------------------------------------------
901 Two significant changes which require the new version number: The
902 improved authentication dialog and the fact that the database code
903 has been moved to a library, libosl. To use the new version, you have
904 to generate new RSA keys, see INSTALL for details. A shell script is
905 provided for conversion of the 0.3 database to the new 0.4 format.
907 - stronger crypto for client authentication
908 - the database code has been moved to a library
909 - improved status item handling
910 - cleanup of the build system
911 - The "-V" option now also prints the git version
912 - the new parser-friendly listing mode for the ls and stat commands
913 - mandatory rc4 encryption
914 - major audio format handler cleanups
915 - (id3,...) tags are no longer stored as a combined string in the database
916 - new mood methods: artist_matches, title_matches, comment_matches,
917 album_matches, year_maches, year.
919 --------------------------------------------
920 0.3.5 (2009-09-21) "symplectic separability"
921 --------------------------------------------
923 Full client support for \*BSD Unixes, complete re-write of the ogg
924 vorbis audio format handler, various improvements all over the place
925 and the usual mix of bugfixes. This release marks the end of the 0.3
926 series if no serious problems show up.
928 - the new oss writer (supported on \*BSD and Linux)
929 - rewrite of the ogg vorbis audio format handler. It's
930 recommended to replace the chunk tables of existing ogg
931 vorbis files in the afs database by re-adding these files
933 - support for netmask subsets (Gerrit Renker)
934 - the new prebuffer filter
935 - improved signal handling
936 - variable fec output buffer size
937 - improved FEC timing fixes audible buffer underruns in UDP mode
938 - --log_color actually works
939 - new ls option: -d (print dates as seconds after the epoch)
940 - update to gengetopt 2.22.2
941 - support for RSA keys of size > 512 bits
942 - new option "mixer_channel" for para_fade
944 -----------------------------------------
945 0.3.4 (2009-05-07) "elliptic inheritance"
946 -----------------------------------------
948 The new udp sender, forward error correction, colored logs and various
949 other improvements. As the udp sender does not depend on any special
950 libraries, it is built unconditionally.
952 - The udp sender replaces the ortp sender. The new code uses forward
953 error correction to protect against packet losses. Many thanks to
954 Gerrit Renker for providing ipv6 support.
955 - The default port for udp streaming now defaults to 8000, like
956 for the http and the dccp senders/receivers.
957 - Loglevels are now specified as symbolic names, e.g.
959 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
960 - para_server/para_audiod: Color support for log messages.
961 - new options for mp3dec: --ignore-crc, --bufsize
962 - new audiod option: --config-file.
963 - gengetopt cleanups.
964 - Improved help/man pages: The documentation of para_audiod,
965 para_recv, para_filter and para_write now also contains
966 all options of the available receivers/filters/writers. The
967 man page of para_fade contains a description of the different
969 - More source code documentation.
972 --------------------------------------------
973 0.3.3 (2008-12-01) "axiomatic perspectivity"
974 --------------------------------------------
976 Internal code cleanups, bug fixes, improved tag handling and the new
977 amplification filter.
979 - para_server uses the generic scheduling code.
980 - overhaul of the virtual streaming system.
981 - mp3: id3 version 2 support via libid3tag (optional)
982 - ogg: vorbis comment support.
983 - aac meta info support.
984 - mp3 audio format handler cleanups.
985 - new filter: "amp" to amplify the amplitude of the audio stream
986 - new status item/database entry: amplification. It is
987 used by the amp filter to pre-amplify the audio stream.
988 - fix a close-without-open bug in para_write.
989 - fix a bug in com_init() which was introduced in 0.3.2.
990 - better error diagnostics for para_client.
992 -----------------------------------------
993 0.3.2 (2008-04-11) "probabilistic parity"
994 -----------------------------------------
996 The new para_afh executable, scheduling and documentation improvements.
998 - new ls option: -lc (list chunk table)
999 - new executable: para_afh, the stand-alone audio file handler tool
1000 - afs commands can send output more than SHMMAX (32MB on Linux). This
1001 also reduces the memory usage of commands that produce large amounts
1003 - major scheduler and audiod cleanups.
1004 - more detailed and much nicer man pages.
1006 ---------------------------------------
1007 0.3.1 (2008-02-23) "liquid interaction"
1008 ---------------------------------------
1010 A mix of cleanups, bug fixes, improvements, and some new features. No
1011 significant changes to the new database (osl) code, which is generally
1014 - Share some similar/duplicate code between the http and the
1016 - Generic access control lists for paraslash senders.
1017 - dccp sender: Access control lists, connection limiting and support
1018 for the allow,deny,on,off,help sender commands.
1019 - The default dccp port changed from 5001 to 8000 (suggested by
1021 - para_server starts even if not all public keys could be loaded.
1022 - Audiod performance improvements.
1023 - fix a bug in the "off" command of the http sender.
1024 - fix some fd and memory leaks.
1025 - Update to gengetopt-2.22.
1027 -------------------------------------
1028 0.3.0 (2008-01-12) "solar saturation"
1029 -------------------------------------
1031 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
1034 Usually one might expect lots of new features AND a big increase in size
1035 for a major release like this.
1037 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
1038 ever. The decrease in size is mostly due to the removal of some
1039 graphical tools (which were only quick hacks anyway). But also the
1040 fact that the mysql code is gone cuts down the size a bit.
1042 Being independent of mysql comes at a cost: The fact that paraslash
1043 now contains its own database (the object storage layer, osl) increases
1044 the (stripped) binary size of para_server by ~50K on i386.
1046 - no more restrictions on unique basenames.
1047 - independent of mysql: The new self-contained object
1048 storage layer (osl) replaces the mysql database.
1049 - New executable para_fsck: Check integrity of osl tables.
1051 - Reliable audio file move/rename detection.
1052 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
1053 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
1055 - the new osl-based audio file selector (afs) replaces the random,
1056 the playlist and the mysql selector of paraslash-0.2.x.
1057 - IPv6 support (thanks to Gerrit Renker).
1058 - paraslash-0.2.x streams are now called "moods". Writing
1059 0.3.x-mood definitions should be both easier and more
1060 powerful than writing 0.2.x-stream definitions.
1061 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
1062 been removed. The world is a better place without them. However,
1063 para_gui is still there.
1064 - afs tracks audio file selection also in playlist mode.
1065 - few easy-to-use afs commands replace the many not-so-easy-to-use
1066 mysql commands (and are available also in playlist mode).
1067 - Improved error subsystem.
1068 - The earth-shaking new logo.
1070 -----------------------------------------
1071 0.2.17 (2007-11-20) "isotropic threshold"
1072 -----------------------------------------
1074 Mainly bugfixes and cleanups in this version which marks the end of
1075 the 0.2.x series if no serious bugs show up after the release.
1077 - mysql_selector: fix a locking bug.
1078 - universal chunk queueing.
1079 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
1081 - be more carful wrt. signed vs. unsigned argument passing.
1082 - cleanup error.h and fix some references to invalid error
1084 - update to gengetopt-2.21.
1085 - update to ortp-0.13.1.
1086 - autoconf: extend checks for headers, library functions and
1087 compiler characteristics.
1088 - Fix streaming of large mp3 files.
1089 - Fix an off-by-one bug in playlist handling.
1091 --------------------------------------
1092 0.2.16 (2007-04-05) "neural discharge"
1093 --------------------------------------
1095 The main change in this release is the major audio format handler
1096 cleanup which removes some similar/duplicate code and makes it easier
1097 to implement plugins for other audio formats. Of course, the usual mix
1098 of other improvements/changes/bugfixes also made it into the release.
1100 - simplified audio format handlers (most of the handling functions
1101 were moved one layer up to the virtual streaming system).
1102 - para_server uses mmap to read audio files
1103 - repositioning of mp3 streams is much faster, in particular for
1104 jumping near the end of large mp3 files.
1105 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
1107 - fix a bug in para_filter that caused decoding of aac files
1108 to start only after a few seconds.
1109 - fix osx_writer hangs
1110 - simplified dccp code (thanks to Gerrit Renker)
1111 - the compress filter works also on big endian systems (ppc)
1113 -----------------------------------------
1114 0.2.15 (2007-02-16) "inductive resonance"
1115 -----------------------------------------
1117 Minor improvements, more documentation and a bunch of bug fixes.
1119 - para_server: The server.users file is only read once on server
1120 startup rather than for each connection
1121 - mp3dec: Fix decoding of corrupt mp3 files
1122 - afs (audio file sender) is now called vss (virtual streaming
1123 system). Consequently, the permission flags specified in
1124 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
1125 become VSS_READ and VSS_WRITE respectively.
1126 - para_audiod/para_filter: Fix a bug that caused the last chunk
1127 of audio data not being written under certain circumstances
1128 - audiod: compute the difference of server time and local time
1130 - para_server/para_audiod: Fix some memory leaks
1131 - documentation improvements
1132 - configure.ac: fix checks for para_krell
1135 -------------------------------------------
1136 0.2.14 (2006-10-15) "transient singularity"
1137 -------------------------------------------
1139 The only major enhancement of this version is the osx writer which completes
1140 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
1141 until now for reasons beyond the scope of this changelog entry.
1143 - new output plugin for Mac Os: the osx writer
1144 - rename configure command line options from --enable-xxx-headers to
1145 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
1146 - configure: new command line options: --with-mad-headers,
1147 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
1148 - some robustness fixes
1149 - dymamic audio format recognition for audiod
1150 - para_server: new command line option: --autoplay_delay
1151 - para_audiod: new command line option: --clock_diff_count
1153 ---------------------------------------
1154 0.2.13 (2006-07-14) "sonic convolution"
1155 ---------------------------------------
1157 A bunch of new features and core changes.
1159 - the new paraslash scheduler, short and sweet.
1160 - Support for m4a/mp4 files via the new aac audio format
1161 handler/filter (requires libfaad).
1162 - each writer has its own command line parser, just like
1163 para_recv and para_filter.
1164 - para_client and para_audioc use the error subsystem
1165 - writers are integrated in para_audiod (currently linux-only)
1166 - para_client is integrated in para_audiod
1167 - random/playlist selector: improved info strings
1168 - new audiod commands: tasks, kill
1169 - update to libortp-0.10.1
1170 - para_fade: wake time defaults to 8 hours from now
1171 - update to autoconf-2.60
1173 ------------------------------------------
1174 0.2.12 (2006-05-12) "oriented abstraction"
1175 ------------------------------------------
1177 Many user-visible changes in this release and lots of new
1180 - the new optional dccp sender/receiver. It uses the datagram
1181 congestion control protocol. You'll need a fairly new kernel
1183 - paraslash works on Mac OS X (thanks to Gerd Becker)
1184 - para_play renamed to para_write
1185 - modular output plugin design (writers) for para_write
1186 - new file_writer output plugin for para_write
1187 - compress filter speed improvements
1188 - update to libortp-0.9.1
1189 - update to gengetopt-2.17rc
1190 - para_client no longer depends on libreadline (as the
1191 code for the interactive mode was removed).
1192 - gcc-2-95 is no longer a supported compiler. It may still
1193 work, but it gets no more testing.
1194 - the tarball no longer contains the screenshot images which
1195 reduces its size quite a bit.
1196 - configure: new command line options: --enable-mysql-headers
1197 and --enable-mysql-libs
1199 ------------------------------------
1200 0.2.11 (2006-03-11) "atomic duality"
1201 ------------------------------------
1203 Here it is, the first paraslash release developed with git. There
1204 are fairly many user-visible changes in this release. As two out of
1205 the three "database tools" of paraslash don't use a database at all,
1206 they are now called "audio file selectors" instead.
1208 - the cdt command (change database tool) becomes chs (change
1210 - no more colon separators: The syntax of some options of
1211 para_audiod and para_filter have changed. Use --help for
1212 more info (and some examples).
1213 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
1214 - switch from cvs to git (should've done that earlier)
1215 - the new ipc subsystem
1216 - new audio file selector: playlist
1217 - para_server: the dopey selector is now called "random",
1218 and is the default selector. Use the --selector option to
1219 choose another selector at startup, or the chs command to
1220 change the selector at runtime.
1221 - X86_64 fixes (thanks to Steffen Klassert)
1224 --------------------------------------
1225 0.2.10 (2006-02-17) "cyclic attractor"
1226 --------------------------------------
1228 Huge documentation update, a scrollable window for para_gui, ortp
1229 improvements, and of course many small fixes not mentioned here.
1230 The diffstat below is rather misleading as most insertions are due
1231 to the new source documentation.
1234 - para_server also uses the new error subsystem
1235 - lots of new documentation (UTSL)
1237 - keysyms for cursor keys and for next/previous page keys
1238 - scrollable output window
1239 - new internal commands: scroll up/down, page up/down
1240 - fix color of command output.
1241 - ortp: the --chunk_time and --header flags are no longer needed
1242 for para_recv/para_audiod as this information is now encoded in
1243 each rtp packet sent by para_server.
1245 -------------------------------------------
1246 0.2.9 (2006-01-24) "progressive turbulence"
1247 -------------------------------------------
1249 Internal audiod receivers/filters, the new error subsystem and
1250 a lot of small improvements.
1252 - para_recv and para_filter are integrated into the para_audiod
1253 binary, i.e. audiod no longer spawns a new process for
1254 each receiver/filter. As para_recv and para_filter might be
1255 useful as standalone programs, they still get built (linked
1256 against the same object files that are also used for audiod).
1257 - further ortp timing improvements should reduce the CPU usage
1258 of the ortp receiver.
1259 - improved audio grabbing. The 'grab' command of para_audiod
1260 has its own set of command line options. Read the output of
1261 "para_audioc help grab" for more info.
1262 - oggdec: configurable input prebuffer size.
1263 - the new error subsystem gives better error diagnostics
1264 and reduces code size.
1266 -----------------------------------------
1267 0.2.8 (2006-01-02) "dynamic accumulation"
1268 -----------------------------------------
1270 The new modular filter design and the para_play-hangs bugfix.
1272 - new executable: para_filter. It combines para_mp3dec,
1273 para_oggdec and para_compress. It also adds a further filter
1274 type, wav, that just inserts a wave header at the desired point
1275 of the filter chain. All 'piping' is done in-memory (i.e. no
1276 read/write operations are used).
1277 - para_play: fix a stupid bug that caused it to hang under
1278 certain circumstances.
1280 -------------------------------------------
1281 0.2.7 (2006-12-27) "transparent invariance"
1282 -------------------------------------------
1284 Not many user-visible changes but a fair amount of internal improvements.
1286 - The http sender buffers data if it can not be sent
1287 out immediately (because the socket is not writable). This
1288 should prevent para_server from shutting down the connection
1289 too early on a loaded network.
1290 - para_play also prebuffers data if it is told to start at a
1291 future time by the --start_time option.
1292 - The return of para_recv: It combines para_ortp_recv and
1293 para_http_recv. Use the --receiver option to switch between
1294 the two. para_recv builds without libortp, but contains
1295 only the http receiver in this case.
1296 - update to ortp 0.8.1. As this ortp release contains incompatible
1297 changes, para_recv-0.2.7 won't link against older ortp libs.
1298 - improved ortp timings.
1299 - use of gcc-extensions that #define away for non-gcc and
1302 -------------------------------------------
1303 0.2.6 (2005-10-29) "recursive compensation"
1304 -------------------------------------------
1306 Transparent session encryption (uses openssl's Alleged RC4 cipher),
1307 the internal find command and several other improvements and cleanups.
1309 - Encrypt the session if encryption is requested by the client
1310 (default for para_client 0.2.6). This is backwards
1311 compatible, so older clients can still connect to para_server
1312 0.2.6. Use the new client option --plain to request an
1313 uncrypted session (off by default, must be set to on in
1314 order to connect to para_server 0.2.x with 0 <= x <= 5).
1315 - para_server uses an internal function to locate audio files
1316 rather than calling find(1). The server option
1317 --mysql_audio_file_dir replaces --mysql_find_cmd.
1318 - documentation update
1320 - header file cleanup
1321 - para_client code cleanup
1322 - para_gui: faster display of output of display commands
1324 ------------------------------------------
1325 0.2.5 (2005-10-13) "aggressive resolution"
1326 ------------------------------------------
1328 This release adds internal senders, i.e. no more external programs are
1329 spawned for sending out the audio data. There are two different senders
1330 available: The http sender and the ortp sender (former para_send which
1331 is no longer needed).
1333 The new sender code has a plugin-like design so it can be easily
1334 extended should there be be any future need for supporting another
1335 network streaming protocol. All senders are completely independent of
1336 each other. In particular, the http and the ortp sender can operate
1339 - new server command: sender to control senders at runtime.
1340 Read the output of "para_server -h" and "para_client help
1341 sender" for more information.
1342 - para_recv renamed to para_ortp_recv
1343 - new executable: para_http_recv, a simple command line
1345 - major afs/mp3/ogg code simplifications due to internal
1347 - ogg timing improvements
1348 - fix several minor memory leaks (found by valgrind)
1349 - empty stream definitions work again
1350 - com_ne(): ignore errors on remove
1351 - audiod: fix segfault on server restart
1353 ---------------------------------------
1354 0.2.4 (2005-09-21) "toxic anticipation"
1355 ---------------------------------------
1357 Several small improvements, fixes and the new grab command.
1360 - new command: "grab" to grab the output of the stream reader
1361 or any filters. Read the output of "para_audioc help grab"
1362 for more information.
1365 - audioc: new command line option: --bufsize to specify a
1366 buffer size different from the default size 8192.
1367 - improved error diagnostics for para_play.
1368 - new configure option: --enable-ssldir so search for openssl in
1370 - sdl_gui: Make it look nice again for 1024x768
1371 - server: report total size of memory allocated with sbrk by malloc,
1372 new command line option: --announce_time
1374 -----------------------------------------
1375 0.2.3 (2005-09-01) "hydrophilic movement"
1376 -----------------------------------------
1378 Two new executables and major feature enhancements.
1380 - audiod filters: It is now possible to specify arbitrary many
1381 (including none) filters for each supported audio
1382 format. This can be used e.g. for normalizing volume,
1383 transforming or grabbing the audio stream, or for using
1384 visualizers. Read the output of "para_audiod -h" for the
1385 syntax of the new --filter_cmd option.
1386 - new executable: para_play, a tiny alsa player. It
1387 can play wave files or raw pcm (16 bit little endian)
1389 - new executable: para_compress, a dynamic range compressor
1390 intended to keep audio output at a consistent volume. Derived
1391 from [AudioCompress](http://trikuare.cx/code/AudioCompress.html).
1392 - audiod: New option: --stream_delay. This can be used in
1393 a local network to syncronize the audio output of all
1394 clients that play the same stream.
1396 ------------------------------------------
1397 0.2.2 (2005-08-19) "tangential excitation"
1398 ------------------------------------------
1400 Mostly internal changes in this release, but also some new commands
1401 for the mysql database tool.
1403 - cleanup exec.c, fix para_exec bug
1404 - compile time loglevel (log messages below the given level
1405 won't be compiled in, which reduces the size of the
1407 - new log macros that shorten the size of the source code.
1408 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
1409 to send only zeros. With this workaround, para_audioc works
1411 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
1412 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
1413 (new entry), "snp" (set numplayed). Read the manual for more
1416 ---------------------------------------
1417 0.2.1 (2005-08-15) "surreal experience"
1418 ---------------------------------------
1420 Here comes paraslash-0.2.1. It contains a couple of new features and,
1421 surprise, only minor bug fixes.
1423 - kill noisy mp3 debug message
1424 - cleanup of the build system
1425 - para_server and para_client directly use the crypto routines
1426 of the openssl library rather than invoking the openssl command
1428 - server/audiod: new option --user to switch to the given user
1429 when invoked as root. Read the output of "para_server -h" for
1431 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
1432 status. Default: "para_audioc stat"
1433 - sdl_gui: new option --pic_cmd to be used to download the picture.
1434 Default: "para_client pic"
1435 - audiod: 5 slots ought to be enough for everybody
1436 - audiod: new status item: Uptime, kill hup command
1438 ------------------------------------------
1439 0.2.0 (2005-08-06) "distributed diffusion"
1440 ------------------------------------------
1442 After several month of increased development activity, paraslash-0.2.0
1443 has arrived. It contains many new features and is much more
1444 self-contained than the old 0.1.x series. Enjoy!
1446 - para_server: fix hang on song change and crash on sighup.
1447 Speed up mysql queries. The DIR_LIKE macro is gone.
1448 - new executables: para_audiod, the local audio daemon that
1449 starts playback (uses SCM_CREDENTIALS socket magic) and
1450 para_audioc, the corresponding client.
1451 - new executables: para_mp3dec/para_oggdec, two really teensy
1452 decoders. para_mp3dec is based on libmad, para_oggdec requires
1454 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
1455 they are now called para_send and para_recv respectively.
1456 - documentation updates
1457 - para_gui is themable. For now there is the default theme that
1458 looks as before and the simple theme: blue and easy.
1459 - gui: audio streaming is now handled by audiod. Time display shows
1460 playback time rather than streaming time
1461 - slider: update to libzmw-0.2.0
1462 - para_krell: fix crash on server shutdown
1463 - switch from gzip to bzip2
1465 ----------------------------------------
1466 0.1.7 (2005-04-18) "melting penetration"
1467 ----------------------------------------
1469 The main change in this release is clearly the oggvorbis rewrite,
1470 but there are also lots of smaller changes. If you intend to use both
1471 the mp3 and the ogg plugin, it is recommended to use software mixing,
1472 e.g. the dmix plugin which is provided by ALSA.
1474 - new executables: para_ovsend and para_ovrecv for sending/receiving
1475 oggvorbis files via rtp. Requires the open rtp library. Get it at
1476 http://www.linphone.org/ortp/
1477 - rewrite of the ogg_vorbis core code
1478 - configure detects libzmw and, if detected, includes
1479 para_slider to the list of binaries to be built by make
1480 - server stream writers read from their associated fifo rather
1482 - slider: two new sliders, lastplayed and numplayed
1483 - fix nasty double free bug which caused random segfaults in case of
1484 mp3 files with invalid header information
1485 - gui: new command line option: --stream_timeout=seconds to
1486 deactivate a slot if it is idle for that many seconds (default=`5')
1489 ---------------------------------------
1490 0.1.6 (2005-03-05) "asymptotic balance"
1491 ---------------------------------------
1493 Only little user-visible changes in this release. Mainly bugfixes and
1494 core code cleanup. This is probably the most stable version ever if you
1497 - fix several memory leaks
1498 - rename default name of mysql database from "music" to "paraslash".
1499 Use para_server's --mysql_database option if you do not want to
1500 rename your old database.
1501 - rework ogg vorbis code
1502 - make update command work on mysql servers with LOCAL_INFILE
1504 - gui: improved stream I/O (slots)
1505 - simplified audio format API
1506 - para_pob_ogg is gone
1508 ------------------------------------
1509 0.1.5 (2004-12-31) "opaque eternity"
1510 ------------------------------------
1512 Let's slide gently into the new year.
1514 - new: para_slider (not built automatically, type "make
1515 para_slider" to build). A toy for those who always felt that
1516 creating stream definitions is difficult. See screenshots,
1517 README and FEATURES for more info.
1518 - improved signal handling. Fixes server segfault on SIGHUP
1519 for linux kernels newer than Aug 24 2004 and makes para_gui
1521 - reload database tool on SIGHUP
1522 - improved help message for sl
1523 - do not log "broken pipe" messages as errors. They are
1525 - fix wrong error message on permission errors
1527 -----------------------------------------
1528 0.1.4 (2004-12-19) "tunneling transition"
1529 -----------------------------------------
1531 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
1532 Hopefully, most of them got wiped out with this release. Some
1533 enhancements went also in.
1535 - improved error diagnostics for all commands
1536 - stradd/picadd: overwrite previous contents if entry already
1537 exists, rather than returning errors
1538 - stradd: use current stream if invoked without args
1539 - faster (and hopefully more stable) ogg-vorbis handling
1540 - para_krell: reap children to avoid zombie-flooding in case
1541 no server is running
1542 - si: report also server pid
1543 - server: don't busy-loop if dbtool reports only invalid files.
1544 - gui: CTRL+C works again, fix stream_read command line option
1546 - fix mysql dbtool startup in case no database exists
1547 - many small fixes and cleanups
1549 ---------------------------------------
1550 0.1.3: (2004-12-10) "vanishing inertia"
1551 ---------------------------------------
1553 Starting from this release, the database tools are integrated in the
1554 server binary. This decreases server startup time, reduces code size
1555 and speeds up database commands. However, the layout of the underlying
1556 mysql database changed only slightly and 0.1.3 should be backwards
1557 compatible in that respect.
1561 - If mysql is not detected at compile time, or fails to init
1562 at runtime, fall back to the dopey database tool which should
1564 - para_dbtool and dbtool.conf are gone. All mysql specific
1565 options are read from server.conf and are prefixed by 'mysql_'.
1566 - new command: cdt (change database tool)
1567 - new command line option: dbtool (choose startup database tool)
1568 - The name of current stream is now stored in the database,
1569 so paraslash remembers its current stream when restarted.
1570 - new command: csp (change stream and play)
1571 - para_gui also reports current database tool and server uptime
1573 -------------------------------------------
1574 0.1.2: (2004-11-28) "spherical fluctuation"
1575 -------------------------------------------
1577 Point release before the big dbtool changes go in.
1579 - dbtool: rename ca to cam (copy all meta data). It now also
1580 copies numplayed and lastplayed time as well as the picture
1582 - fix endless-loop-bug caused by mp3 files with invalid header
1584 -----------------------------------------
1585 0.1.1: (2004-11-05) "floating atmosphere"
1586 -----------------------------------------
1589 - new dbtool command: mbox. Browse your sound-file collection
1590 with your favorite mail reader.
1591 - several small fixes
1593 -------------------------------------
1594 0.1.0: (2004-10-22) "rotating cortex"
1595 -------------------------------------
1597 - fix logging bug for loglevel > VERBOSE
1599 - correct timings for vbr mp3s
1600 - modular audio format support
1601 - ogg-vorbis support (experimental)
1602 - new server option: autoplay
1604 -----------------------------------------
1605 0.0.99: (2004-07-25) "harmonic deviation"
1606 -----------------------------------------
1608 - rename projectname from icc to paraslash (play, archive, rate
1609 and stream large audio sets happily)
1610 - paraslash is no longer restricted to one particular audio
1612 - new dbtool commands (stradd, strq, strdel) for easy stream
1613 managment w/o configuration file. That obsoletes stream_defs
1614 file/config option for dbtool.
1615 - picadd accepts jpeg data from stdin
1616 - new server commands: ps (select previous stream), sc (song change)
1617 - new default pictures for sdl_gui
1618 - gui: new key_map option for binding commands and internal
1619 functions to arbitrary keys, nice help screen, rip out
1620 soundcard/linux specific stuff, avoid noise artefacts while jumping,
1621 show silly logo on startup
1622 - new executables: para_fade for fading volume, para_dbadm for
1623 manipulating attributes
1624 - cdb adds _all_ tables to mysql database
1625 - revised and beautified documentation
1626 - sample dbtool rewritten in C
1629 ---------------------------------------------
1630 0.0.98: (2003-12-26) "incremental smoothness"
1631 ---------------------------------------------
1633 - kick icecast in favour of poc. That removes some races and reduces
1634 core code considerably.
1635 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
1636 give finer info on afs' status.
1637 - gui can start decoder (see config options). Further new gui
1638 commands: refresh (^L), jmp (F1-F10)
1639 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1640 terminates gui. This fixes dead instances consuming memory
1642 - new dbtool command: verb for sending verbatim sql queries.
1643 - fix pid_list races (by removing pid_list)
1646 --------------------
1647 0.0.97: (2003-10-26)
1648 --------------------
1650 - installation prefix now defaults to /usr/local
1651 - new commands for gui: snozze, sleep and reread config
1652 - config file for gui and sdl_gui
1653 - fix problems with filenames containing funny characters
1654 (reported by Thomas Forell)
1655 - improved signal handling for gui, now it rereads conf on SIGHUP
1656 - new dbtool command: cdb (create database)
1657 - switch from argtable to gengetopt
1658 - major code cleanup and speed improvements
1659 - fix several potential buffer overflows
1660 - many small fixes and cleanups
1666 - easy stream_defs syntax
1667 - sdl_gui can display images associated to the file being played
1668 - Major feature enhancements for icc_gui including dynamic text
1669 placement and the top/bottom window design
1670 - vrfy/clean now also checks for NULL values in attributes as
1671 well as for invalid picture pointers
1672 - fix long outstanding case sensitivity bug
1673 - many small fixes and cleanups
1679 - sdl gui runs much faster
1680 - new dbtool command: ca (copy attributes)
1681 - count and display number of times the song has been played
1682 - new feature: scoring
1683 - command line options for sdl_gui
1684 - simpler syntax of streams file
1685 - decrease network traffic of stat
1687 - many small fixes and cleanups
1693 - new server command: ns (next stream)
1694 - new icc_gui command: c (change stream)
1696 - stat shows also id3 tag info
1698 - log flodding bug fixed
1699 - many small fixes and cleanups
1705 - colors for icc_gui
1706 - icc_gui sets volume directly (linux only)
1707 - proper locking that fixes some races
1708 - fix security bug that caused commands to be executed even
1709 with unsufficient permissions
1710 - new command: hup to make all servers reread their configuration file
1711 - icecast meta data streaming
1712 - many small fixes and cleanups