sideband: Fix use after free bug.
[paraslash.git] / NEWS
1 -------------------------------------------
2 0.4.11 (to be announced) "mutual diversity"
3 -------------------------------------------
4
5 - Sideband connections: If both para_server and para_client
6 support this feature, data is sent as a multiplexed stream.
7 - The --no_default_filters option of para_filter has been
8 removed.
9 - Several fixes and latency improvements to various decoders.
10 - Documentation improvements.
11
12 ------------------------------------------
13 0.4.10 (2012-03-30) "heterogeneous vacuum"
14 ------------------------------------------
15
16 Nothing earth-shaking in this release, but quite a few usability
17 improvements and the usual mix of cleanups and fixes.
18
19 - The --no_default_filters option of para_filter has been
20 deprecated. It still works but has no effect and will be
21 removed in the next version.
22 - para_gui now prints also the stderr output of the executing
23 command in the bottom window.
24 - Cleanup and consolidation of the various wrappers for
25 write(), writev(), send() and friends.
26 - The obscure error messages on mmap() failures have been
27 replaced by meaningful messages. This affects mainly
28 para_afh.
29 - para_audioc: Cleanups and memory leak fixes.
30 - Test 0004-server no longer fails if para_server is not
31 being built.
32 - New configure options: --with-id3tag-{headers,libs}.
33
34 -------------------------------------
35 0.4.9 (2011-12-06) "hybrid causality"
36 -------------------------------------
37
38 Support for another audio format, interactive mode for para_client
39 and para_audiod and many small improvements/fixes all over the place.
40
41 - Support for flac, the free lossless audio codec.
42 - Fix for an endless loop in the mp3 decoder for certain
43 (corrupt) mp3 files.
44 - When executed without specifying a command, para_client
45 and para_audioc start an interactive shell (requires
46 libreadline being installed). The interactive mode offers
47 full tab completion and command line history.
48 - autogen.sh now detects a distcc setup and adjusts the
49 parameter for the -j option of make accordingly.
50 - Shared memory areas are no longer restricted to 64K. We now
51 detect the maximal size of a shared memory area at runtime.
52 - cleanup of the internal uptime API.
53 - para_server prefaults the mmapped audio file to avoid
54 delays on slow media.
55 - A new test for the test-suite that exercises the
56 communication between para_server and para_audiod.
57 - The alsa writer eats up less CPU cycles when configured to
58 use the DMIX plugin.
59 - Simplified and unified receiver code.
60 - Makefile cleanups.
61 - Commands which print a list of matching audio files now
62 emit a meaningful error message if no audio file matched the
63 given pattern(s).
64
65 --------------------------------------
66 0.4.8 (2011-08-19) "nested assignment"
67 --------------------------------------
68
69 Gcrypt support, the overhauled osx writer and regex format specifiers
70 are the highlights of this release.
71
72 - support for libgcrypt as a drop-in replacement for openssl.
73 Run configure --enable-cryptolib=gcrypt to link against
74 libgcrypt. The two crypto implementations are compatible to
75 each other, i.e. a para_client executable linked against
76 gcrypt can connect to para_server linked against libssl
77 and vice versa.
78 - Rewrite of the osx writer (output plugin for Mac OS).
79 - audiod: The format specifier for receivers, filters and
80 writers is now treated as a regular expression. This allows
81 to replace 5 lines in the config file (one for each audio
82 format) by one single line. See the manual for details.
83 - The *.cmdline.[ch] files are no longer contained in the released
84 tarballs. This reduces the size of the tarballs but requires
85 gengetopt to build the tarball.
86 - Compiles cleanly also with llvm/clang.
87 - Corrupt mp3 files are handled more gracefully.
88 - The alsa writer uses poll fds instead of computing timeouts.
89 - Cleanup of the generic writer API.
90 - sched: Optimized zero timeouts.
91 - vss timeout cleanups.
92 - oggdec fixes and improvements.
93
94 --------------------------------------
95 0.4.7 (2011-06-01) "infinite rollback"
96 --------------------------------------
97
98 The new ao writer, support for ssh RSA keys and a couple of other
99 enhancements.
100
101 - Support for ESD, Pulseaudio, AIX, Solaris, IRIX and other
102 platforms through the libao audio library.
103 - Support for RSA keys generated with ssh-keygen.
104 - configure: improved options for ogg/vorbis/speex.
105 - The git version reported by --version always matches HEAD.
106 - The autogen script detects the number of processors and
107 runs a parallel make if possible.
108 - Major cleanup of the crypto API.
109 - Documentation updates.
110
111 ------------------------------------------
112 0.4.6 (2011-03-31) "deterministic entropy"
113 ------------------------------------------
114
115 Lots of ogg/vorbis improvements, the new test suite, enhancements
116 for para_gui and a fair amount of other bug fixes.
117
118 - For DCCP/OGG streams the audio file header is only sent once
119 at the beginning of the stream rather than periodically
120 every five seconds. This reduces network traffic and the
121 FEC group size.
122 - The vorbis comment header is replaced by an empty dummy header
123 before the header is sent over the network. This also results in
124 less network traffic and smaller FEC groups.
125 - The new "test" make target allows to perform some sanity checks prior
126 to installing the package.
127 - ogg timing fixes and performance improvements
128 - Scheduler improvements
129 - Proper exit codes for para_write
130 - para_gui: New option --theme to select a startup theme. Several
131 other improvements and fixes.
132 - aacdec error message cleanups
133 - simplified color error handling
134
135 --------------------------------------------
136 0.4.5 (2010-12-17) "symmetric randomization"
137 --------------------------------------------
138
139 Bug fixes, internal cleanups and variable-sized FEC slices.
140
141 - Contains a fix for an invalid-free-bug in the ogg audio format
142 handler code.
143 - Switching off the DCCP sender works again.
144 - para_audiod handles crashes of para_server more robustly.
145 - Internal scheduler and writer cleanups.
146 - Reduced latency due to variable-sized FEC slices.
147 - Improved documentation and error diagnostics.
148 - The build of para_server is now optional, allowing the build
149 to succeed in case libosl is not installed.
150
151 ------------------------------------------
152 0.4.4 (2010-08-06) "persistent regularity"
153 ------------------------------------------
154
155 Support for yet another audio format, para_write improvements and
156 bug fixes.
157
158 - Support for the speex codec.
159 - Support for sample formats other than 16 bit little endian.
160 - error2.h is now created by a perl script which speeds up configure
161 considerably.
162 - Fix a bug in the aac decoder which could lead to segfaults in
163 para_filter/para_audiod.
164 - Fixes for autoconf-2.66.
165
166 ----------------------------------------
167 0.4.3 (2010-07-05) "imaginary radiation"
168 ----------------------------------------
169
170 Many improvements for the DCCP and the UDP transport, the new user
171 manual and the usual mix of bug fixes and internal improvements.
172
173 - FEC support for the DCCP sender (Gerrit Renker). The new
174 --dccp_max_slice_size, --dccp_data_slices_per_group and
175 --dccp_slices_per_group options can be used to set the FEC
176 parameters for the DCCP transport.
177 - DNS lookups for UDP targets (Gerrit Renker).
178 - The new user manual replaces the README, README.afs, REQUIREMENTS
179 and INSTALL documents.
180 - Fix an end-of-file detection bug in the oggdec filter.
181 - The new nonblock API.
182 - Both options of the oggdec filter have been removed.
183 - New debug mode for the internal scheduler.
184
185 ------------------------------------------
186 0.4.2 (2010-04-23) "associative expansion"
187 ------------------------------------------
188
189 It's been some time since the last release, but finally here is
190 paraslash-0.4.2. The bulk of the changes comes from the new buffer
191 tree API, but there are changes all over the tree. Mainly performance
192 and usability improvements, but also quite some bug fixes.
193
194 - The new buffer tree API.
195 - DCCP: Support for CCID negotiation (Gerrit Renker).
196 - UDP robustness fixes.
197 - The --bufsize option for mp3dec is gone as it no longer makes sense
198 for the new buffer tree API.
199 - Fix audible buffer underruns for wma streams.
200 - The alsa writer no longer prints meaningless underrun durations.
201 - audiod: Defaults work also for udp streams. If no filter is
202 given for an audio format that is received via upd, fecdec is
203 automatically added as the first filter (along with the decoder).
204
205 ---------------------------------------
206 0.4.1 (2009-12-22) "concurrent horizon"
207 ---------------------------------------
208
209 Support for another audio format, minor feature enhancements and lots of bug
210 fixes. All fixes that have been accumulated in the maint branch (in particular
211 those mentionened in the 0.3.6 release notes) appear in this release as well.
212
213 - wma support.
214 - new afh option: --human to activate human-readable output.
215 - new server/audiod option: --log-timing to print timing information.
216 - build system improvements.
217 - source code documentation updates.
218
219 -------------------------------------
220 0.3.6 (2009-12-07) "cubic continuity"
221 -------------------------------------
222
223 Quite a few bugs have been found and fixed since 0.3.5, so here's
224 another 0.3.x release. No new features.
225
226 - Always check return value of malloc().
227 - ogg vorbis/FEC: Do not write garbage after the audio file header.
228 - exit if root privileges could not be dropped.
229 - FEC: Fix computation of extra slices.
230 - oss: Fix check for empty input buffer.
231 - Avoid buffer underruns due to filter chain output buffer constraints.
232 - server: Fix assignment of afs_pid.
233 - Don't panic if the afs database contains unknown audio formats.
234 - http/dccp: Do not send the audio file header twice.
235 - FEC: Timing improvements.
236
237 ----------------------------------------------
238 0.4.0 (2009-11-10) "simultaneous independence"
239 ----------------------------------------------
240
241 Two significant changes which require the new version number: The
242 improved authentication dialog and the fact that the database code
243 has been moved to a library, libosl. To use the new version, you have
244 to generate new RSA keys, see INSTALL for details. A shell script is
245 provided for conversion of the 0.3 database to the new 0.4 format.
246
247 - stronger crypto for client authentication
248 - the database code has been moved to a library
249 - improved status item handling
250 - cleanup of the build system
251 - The "-V" option now also prints the git version
252 - the new parser-friendly listing mode for the ls and stat commands
253 - mandatory rc4 encryption
254 - major audio format handler cleanups
255 - (id3,...) tags are no longer stored as a combined string in the database
256 - new mood methods: artist_matches, title_matches, comment_matches,
257 album_matches, year_maches, year.
258
259 --------------------------------------------
260 0.3.5 (2009-09-21) "symplectic separability"
261 --------------------------------------------
262
263 Full client support for *BSD Unixes, complete re-write of the ogg
264 vorbis audio format handler, various improvements all over the place
265 and the usual mix of bugfixes. This release marks the end of the 0.3
266 series if no serious problems show up.
267
268 - the new oss writer (supported on *BSD and Linux)
269 - rewrite of the ogg vorbis audio format handler. It's
270 recommended to replace the chunk tables of existing ogg
271 vorbis files in the afs database by re-adding these files
272 with "add -f".
273 - support for netmask subsets (Gerrit Renker)
274 - the new prebuffer filter
275 - improved signal handling
276 - variable fec output buffer size
277 - improved FEC timing fixes audible buffer underruns in UDP mode
278 - --log_color actually works
279 - new ls option: -d (print dates as seconds after the epoch)
280 - update to gengetopt 2.22.2
281 - support for RSA keys of size > 512 bits
282 - new option "mixer_channel" for para_fade
283
284 -----------------------------------------
285 0.3.4 (2009-05-07) "elliptic inheritance"
286 -----------------------------------------
287
288 The new udp sender, forward error correction, colored logs and various
289 other improvements. As the udp sender does not depend on any special
290 libraries, it is built unconditionally.
291
292 - The udp sender replaces the ortp sender. The new code uses forward
293 error correction to protect against packet losses. Many thanks to
294 Gerrit Renker for providing ipv6 support.
295 - The default port for udp streaming now defaults to 8000, like
296 for the http and the dccp senders/receivers.
297 - Loglevels are now specified as symbolic names, e.g.
298 "--loglevel info".
299 - improved ipv4 and ipv6 URI parser (Gerrit Renker).
300 - para_server/para_audiod: Color support for log messages.
301 - new options for mp3dec: --ignore-crc, --bufsize
302 - new audiod option: --config-file.
303 - gengetopt cleanups.
304 - Improved help/man pages: The documentation of para_audiod,
305 para_recv, para_filter and para_write now also contains
306 all options of the available receivers/filters/writers. The
307 man page of para_fade contains a description of the different
308 modes of operation.
309 - More source code documentation.
310 - vss timing fixes.
311
312 --------------------------------------------
313 0.3.3 (2008-12-01) "axiomatic perspectivity"
314 --------------------------------------------
315
316 Internal code cleanups, bug fixes, improved tag handling and the new
317 amplification filter.
318
319 - para_server uses the generic scheduling code.
320 - overhaul of the virtual streaming system.
321 - mp3: id3 version 2 support via libid3tag (optional)
322 - ogg: vorbis comment support.
323 - aac meta info support.
324 - mp3 audio format handler cleanups.
325 - new filter: "amp" to amplify the amplitude of the audio stream
326 - new status item/database entry: amplification. It is
327 used by the amp filter to pre-amplify the audio stream.
328 - fix a close-without-open bug in para_write.
329 - fix a bug in com_init() which was introduced in 0.3.2.
330 - better error diagnostics for para_client.
331
332 -----------------------------------------
333 0.3.2 (2008-04-11) "probabilistic parity"
334 -----------------------------------------
335
336 The new para_afh executable, scheduling and documentation improvements.
337
338 - new ls option: -lc (list chunk table)
339 - new executable: para_afh, the stand-alone audio file handler tool
340 - afs commands can send output more than SHMMAX (32MB on Linux). This
341 also reduces the memory usage of commands that produce large amounts
342 of output.
343 - major scheduler and audiod cleanups.
344 - more detailed and much nicer man pages.
345
346 ---------------------------------------
347 0.3.1 (2008-02-23) "liquid interaction"
348 ---------------------------------------
349
350 A mix of cleanups, bug fixes, improvements, and some new features. No
351 significant changes to the new database (osl) code, which is generally
352 a good sign.
353
354 - Share some similar/duplicate code between the http and the
355 dccp sender.
356 - Generic access control lists for paraslash senders.
357 - dccp sender: Access control lists, connection limiting and support
358 for the allow,deny,on,off,help sender commands.
359 - The default dccp port changed from 5001 to 8000 (suggested by
360 Gerrit Renker).
361 - para_server starts even if not all public keys could be loaded.
362 - Audiod performance improvements.
363 - fix a bug in the "off" command of the http sender.
364 - fix some fd and memory leaks.
365 - Update to gengetopt-2.22.
366
367 -------------------------------------
368 0.3.0 (2008-01-12) "solar saturation"
369 -------------------------------------
370
371 paraslash.0.3.0 -- 'WWDBND --what would databases never do?'.
372
373
374 Usually one might expect lots of new features AND a big increase in size
375 for a major release like this.
376
377 However, paraslash-0.3.0.tar.bz2 is the smallest paraslash tarball
378 ever. The decrease in size is mostly due to the removal of some
379 graphical tools (which were only quick hacks anyway). But also the
380 fact that the mysql code is gone cuts down the size a bit.
381
382 Being independent of mysql comes at a cost: The fact that paraslash
383 now contains its own database (the object storage layer, osl) increases
384 the (stripped) binary size of para_server by ~50K on i386.
385
386 - no more restrictions on unique basenames.
387 - independent of mysql: The new self-contained object
388 storage layer (osl) replaces the mysql database.
389 - New executable para_fsck: Check integrity of osl tables.
390 - Lyrics support.
391 - Reliable audio file move/rename detection.
392 - More portable than ever: Tested on Linux (x86_32, x86_64, sparc64),
393 MacOS (ppc32, x86_32), FreeBSD (x86_32), NetBSD (x86_32) and
394 Solaris (sparc64).
395 - the new osl-based audio file selector (afs) replaces the random,
396 the playlist and the mysql selector of paraslash-0.2.x.
397 - IPv6 support (thanks to Gerrit Renker).
398 - paraslash-0.2.x streams are now called "moods". Writing
399 0.3.x-mood definitions should be both easier and more
400 powerful than writing 0.2.x-stream definitions.
401 - para_krell, para_slider, para_para_sdl_gui, para_dbadm have
402 been removed. The world is a better place without them. However,
403 para_gui is still there.
404 - afs tracks audio file selection also in playlist mode.
405 - few easy-to-use afs commands replace the many not-so-easy-to-use
406 mysql commands (and are available also in playlist mode).
407 - Improved error subsystem.
408 - The earth-shaking new logo.
409
410 -----------------------------------------
411 0.2.17 (2007-11-20) "isotropic threshold"
412 -----------------------------------------
413
414 Mainly bugfixes and cleanups in this version which marks the end of
415 the 0.2.x series if no serious bugs show up after the release.
416
417 - mysql_selector: fix a locking bug.
418 - universal chunk queueing.
419 - dccp sender uses chunk queueing if write() returns EAGAIN (thanks
420 to Gerrit Renker).
421 - be more carful wrt. signed vs. unsigned argument passing.
422 - cleanup error.h and fix some references to invalid error
423 codes.
424 - update to gengetopt-2.21.
425 - update to ortp-0.13.1.
426 - autoconf: extend checks for headers, library functions and
427 compiler characteristics.
428 - Fix streaming of large mp3 files.
429 - Fix an off-by-one bug in playlist handling.
430
431 --------------------------------------
432 0.2.16 (2007-04-05) "neural discharge"
433 --------------------------------------
434
435 The main change in this release is the major audio format handler
436 cleanup which removes some similar/duplicate code and makes it easier
437 to implement plugins for other audio formats. Of course, the usual mix
438 of other improvements/changes/bugfixes also made it into the release.
439
440 - simplified audio format handlers (most of the handling functions
441 were moved one layer up to the virtual streaming system).
442 - para_server uses mmap to read audio files
443 - repositioning of mp3 streams is much faster, in particular for
444 jumping near the end of large mp3 files.
445 - permission flags DB_READ,DB_WRITE have been renamed to AFS_READ
446 and AFS_WRITE.
447 - fix a bug in para_filter that caused decoding of aac files
448 to start only after a few seconds.
449 - fix osx_writer hangs
450 - simplified dccp code (thanks to Gerrit Renker)
451 - the compress filter works also on big endian systems (ppc)
452
453 -----------------------------------------
454 0.2.15 (2007-02-16) "inductive resonance"
455 -----------------------------------------
456
457 Minor improvements, more documentation and a bunch of bug fixes.
458
459 - para_server: The server.users file is only read once on server
460 startup rather than for each connection
461 - mp3dec: Fix decoding of corrupt mp3 files
462 - afs (audio file sender) is now called vss (virtual streaming
463 system). Consequently, the permission flags specified in
464 ~/.paraslash/server.users have also changed: AFS_READ and AFS_WRITE
465 become VSS_READ and VSS_WRITE respectively.
466 - para_audiod/para_filter: Fix a bug that caused the last chunk
467 of audio data not being written under certain circumstances
468 - audiod: compute the difference of server time and local time
469 correctly
470 - para_server/para_audiod: Fix some memory leaks
471 - documentation improvements
472 - configure.ac: fix checks for para_krell
473 - new man pages
474
475 -------------------------------------------
476 0.2.14 (2006-10-15) "transient singularity"
477 -------------------------------------------
478
479 The only major enhancement of this version is the osx writer which completes
480 the Mac OS Port and was originally planned already for 0.2.13 but had to wait
481 until now for reasons beyond the scope of this changelog entry.
482
483 - new output plugin for Mac Os: the osx writer
484 - rename configure command line options from --enable-xxx-headers to
485 --with-xxx-headers and --enable-xxx-libs to --with-xxx-libs
486 - configure: new command line options: --with-mad-headers,
487 --with-mad-libs, --with-oggvorbis-headers, and --with-oggvorbis-libs
488 - some robustness fixes
489 - dymamic audio format recognition for audiod
490 - para_server: new command line option: --autoplay_delay
491 - para_audiod: new command line option: --clock_diff_count
492
493 ---------------------------------------
494 0.2.13 (2006-07-14) "sonic convolution"
495 ---------------------------------------
496
497 A bunch of new features and core changes.
498
499 - the new paraslash scheduler, short and sweet.
500 - Support for m4a/mp4 files via the new aac audio format
501 handler/filter (requires libfaad).
502 - each writer has its own command line parser, just like
503 para_recv and para_filter.
504 - para_client and para_audioc use the error subsystem
505 - writers are integrated in para_audiod (currently linux-only)
506 - para_client is integrated in para_audiod
507 - random/playlist selector: improved info strings
508 - new audiod commands: tasks, kill
509 - update to libortp-0.10.1
510 - para_fade: wake time defaults to 8 hours from now
511 - update to autoconf-2.60
512
513
514 ------------------------------------------
515 0.2.12 (2006-05-12) "oriented abstraction"
516 ------------------------------------------
517
518 Many user-visible changes in this release and lots of new
519 features:
520
521
522 - the new optional dccp sender/receiver. It uses the datagram
523 congestion control protocol. You'll need a fairly new kernel
524 for this.
525 - paraslash works on Mac OS X (thanks to Gerd Becker)
526 - para_play renamed to para_write
527 - modular output plugin design (writers) for para_write
528 - new file_writer output plugin for para_write
529 - compress filter speed improvements
530 - update to libortp-0.9.1
531 - update to gengetopt-2.17rc
532 - para_client no longer depends on libreadline (as the
533 code for the interactive mode was removed).
534 - gcc-2-95 is no longer a supported compiler. It may still
535 work, but it gets no more testing.
536 - the tarball no longer contains the screenshot images which
537 reduces its size quite a bit.
538 - configure: new command line options: --enable-mysql-headers
539 and --enable-mysql-libs
540
541 ------------------------------------
542 0.2.11 (2006-03-11) "atomic duality"
543 ------------------------------------
544
545 Here it is, the first paraslash release developed with git. There
546 are fairly many user-visible changes in this release. As two out of
547 the three "database tools" of paraslash don't use a database at all,
548 they are now called "audio file selectors" instead.
549
550
551 - the cdt command (change database tool) becomes chs (change
552 selector)
553 - no more colon separators: The syntax of some options of
554 para_audiod and para_filter have changed. Use --help for
555 more info (and some examples).
556 - update to gengetopt-2.16 (thanks to Lorenzo Bettini)
557 - switch from cvs to git (should've done that earlier)
558 - the new ipc subsystem
559 - new audio file selector: playlist
560 - para_server: the dopey selector is now called "random",
561 and is the default selector. Use the --selector option to
562 choose another selector at startup, or the chs command to
563 change the selector at runtime.
564 - X86_64 fixes (thanks to Steffen Klassert)
565 - para_play fixes
566
567
568 --------------------------------------
569 0.2.10 (2006-02-17) "cyclic attractor"
570 --------------------------------------
571
572 Huge documentation update, a scrollable window for para_gui, ortp
573 improvements, and of course many small fixes not mentioned here.
574 The diffstat below is rather misleading as most insertions are due
575 to the new source documentation.
576
577 - autoconf cleanup
578 - para_server also uses the new error subsystem
579 - lots of new documentation (UTSL)
580 - gui improvements:
581 - keysyms for cursor keys and for next/previous page keys
582 - scrollable output window
583 - new internal commands: scroll up/down, page up/down
584 - fix color of command output.
585 - ortp: the --chunk_time and --header flags are no longer needed
586 for para_recv/para_audiod as this information is now encoded in
587 each rtp packet sent by para_server.
588
589
590 -------------------------------------------
591 0.2.9 (2006-01-24) "progressive turbulence"
592 -------------------------------------------
593
594 Internal audiod receivers/filters, the new error subsystem and
595 a lot of small improvements.
596
597 - para_recv and para_filter are integrated into the para_audiod
598 binary, i.e. audiod no longer spawns a new process for
599 each receiver/filter. As para_recv and para_filter might be
600 useful as standalone programs, they still get built (linked
601 against the same object files that are also used for audiod).
602 - further ortp timing improvements should reduce the CPU usage
603 of the ortp receiver.
604 - improved audio grabbing. The 'grab' command of para_audiod
605 has its own set of command line options. Read the output of
606 "para_audioc help grab" for more info.
607 - oggdec: configurable input prebuffer size.
608 - the new error subsystem gives better error diagnostics
609 and reduces code size.
610
611
612 -----------------------------------------
613 0.2.8 (2006-01-02) "dynamic accumulation"
614 -----------------------------------------
615
616 The new modular filter design and the para_play-hangs bugfix.
617
618 - new executable: para_filter. It combines para_mp3dec,
619 para_oggdec and para_compress. It also adds a further filter
620 type, wav, that just inserts a wave header at the desired point
621 of the filter chain. All 'piping' is done in-memory (i.e. no
622 read/write operations are used).
623 - para_play: fix a stupid bug that caused it to hang under
624 certain circumstances.
625
626
627 -------------------------------------------
628 0.2.7 (2006-12-27) "transparent invariance"
629 -------------------------------------------
630
631 Not many user-visible changes but a fair amount of internal improvements.
632
633
634 - The http sender buffers data if it can not be sent
635 out immediately (because the socket is not writable). This
636 should prevent para_server from shutting down the connection
637 too early on a loaded network.
638 - para_play also prebuffers data if it is told to start at a
639 future time by the --start_time option.
640 - The return of para_recv: It combines para_ortp_recv and
641 para_http_recv. Use the --receiver option to switch between
642 the two. para_recv builds without libortp, but contains
643 only the http receiver in this case.
644 - update to ortp 0.8.1. As this ortp release contains incompatible
645 changes, para_recv-0.2.7 won't link against older ortp libs.
646 - improved ortp timings.
647 - use of gcc-extensions that #define away for non-gcc and
648 gcc < 3.0.
649
650
651 -------------------------------------------
652 0.2.6 (2005-10-29) "recursive compensation"
653 -------------------------------------------
654
655 Transparent session encryption (uses openssl's Alleged RC4 cipher),
656 the internal find command and several other improvements and cleanups.
657
658 - Encrypt the session if encryption is requested by the client
659 (default for para_client 0.2.6). This is backwards
660 compatible, so older clients can still connect to para_server
661 0.2.6. Use the new client option --plain to request an
662 uncrypted session (off by default, must be set to on in
663 order to connect to para_server 0.2.x with 0 <= x <= 5).
664 - para_server uses an internal function to locate audio files
665 rather than calling find(1). The server option
666 --mysql_audio_file_dir replaces --mysql_find_cmd.
667 - documentation update
668 - man pages
669 - header file cleanup
670 - para_client code cleanup
671 - para_gui: faster display of output of display commands
672
673
674 ------------------------------------------
675 0.2.5 (2005-10-13) "aggressive resolution"
676 ------------------------------------------
677
678 This release adds internal senders, i.e. no more external programs are
679 spawned for sending out the audio data. There are two different senders
680 available: The http sender and the ortp sender (former para_send which
681 is no longer needed).
682
683 The new sender code has a plugin-like design so it can be easily
684 extended should there be be any future need for supporting another
685 network streaming protocol. All senders are completely independent of
686 each other. In particular, the http and the ortp sender can operate
687 in parallel.
688
689 - new server command: sender to control senders at runtime.
690 Read the output of "para_server -h" and "para_client help
691 sender" for more information.
692 - para_recv renamed to para_ortp_recv
693 - new executable: para_http_recv, a simple command line
694 http receiver.
695 - major afs/mp3/ogg code simplifications due to internal
696 senders.
697 - ogg timing improvements
698 - fix several minor memory leaks (found by valgrind)
699 - empty stream definitions work again
700 - com_ne(): ignore errors on remove
701 - audiod: fix segfault on server restart
702
703
704 ---------------------------------------
705 0.2.4 (2005-09-21) "toxic anticipation"
706 ---------------------------------------
707
708 Several small improvements, fixes and the new grab command.
709
710 - audiod:
711 - new command: "grab" to grab the output of the stream reader
712 or any filters. Read the output of "para_audioc help grab"
713 for more information.
714 - fix memory leak
715 - code cleanup
716 - audioc: new command line option: --bufsize to specify a
717 buffer size different from the default size 8192.
718 - improved error diagnostics for para_play.
719 - new configure option: --enable-ssldir so search for openssl in
720 non-standard places
721 - sdl_gui: Make it look nice again for 1024x768
722 - server: report total size of memory allocated with sbrk by malloc,
723 new command line option: --announce_time
724
725
726 -----------------------------------------
727 0.2.3 (2005-09-01) "hydrophilic movement"
728 -----------------------------------------
729
730 Two new executables and major feature enhancements.
731
732 - audiod filters: It is now possible to specify arbitrary many
733 (including none) filters for each supported audio
734 format. This can be used e.g. for normalizing volume,
735 transforming or grabbing the audio stream, or for using
736 visualizers. Read the output of "para_audiod -h" for the
737 syntax of the new --filter_cmd option.
738 - new executable: para_play, a tiny alsa player. It
739 can play wave files or raw pcm (16 bit little endian)
740 from stdin.
741 - new executable: para_compress, a dynamic range compressor
742 intended to keep audio output at a consistent volume. Derived
743 from AudioCompress, http://trikuare.cx/code/AudioCompress.html.
744 - audiod: New option: --stream_delay. This can be used in
745 a local network to syncronize the audio output of all
746 clients that play the same stream.
747
748
749 ------------------------------------------
750 0.2.2 (2005-08-19) "tangential excitation"
751 ------------------------------------------
752
753 Mostly internal changes in this release, but also some new commands
754 for the mysql database tool.
755
756 - cleanup exec.c, fix para_exec bug
757 - compile time loglevel (log messages below the given level
758 won't be compiled in, which reduces the size of the
759 resulting binaries)
760 - new log macros that shorten the size of the source code.
761 - workaround a gcc-4.1 bug (?) that caused send_cred_buffer()
762 to send only zeros. With this workaround, para_audioc works
763 again.
764 - avoid gcc-4 warning: conflicting types for built-in function 'clog'
765 - new mysql commands: "rm" (remove entry), "mv" (rename entry) "ne"
766 (new entry), "snp" (set numplayed). Read the manual for more
767 information.
768
769
770 ---------------------------------------
771 0.2.1 (2005-08-15) "surreal experience"
772 ---------------------------------------
773
774 Here comes paraslash-0.2.1. It contains a couple of new features and,
775 surprise, only minor bug fixes.
776
777 - kill noisy mp3 debug message
778 - cleanup of the build system
779 - para_server and para_client directly use the crypto routines
780 of the openssl library rather than invoking the openssl command
781 line utitlity
782 - server/audiod: new option --user to switch to the given user
783 when invoked as root. Read the output of "para_server -h" for
784 more information.
785 - gui/sdl_gui: new option --stat_cmd to be used to retrieve the
786 status. Default: "para_audioc stat"
787 - sdl_gui: new option --pic_cmd to be used to download the picture.
788 Default: "para_client pic"
789 - audiod: 5 slots ought to be enough for everybody
790 - audiod: new status item: Uptime, kill hup command
791
792
793
794 ------------------------------------------
795 0.2.0 (2005-08-06) "distributed diffusion"
796 ------------------------------------------
797
798 After several month of increased development activity, paraslash-0.2.0
799 has arrived. It contains many new features and is much more
800 self-contained than the old 0.1.x series. Enjoy!
801
802
803 - para_server: fix hang on song change and crash on sighup.
804 Speed up mysql queries. The DIR_LIKE macro is gone.
805 - new executables: para_audiod, the local audio daemon that
806 starts playback (uses SCM_CREDENTIALS socket magic) and
807 para_audioc, the corresponding client.
808 - new executables: para_mp3dec/para_oggdec, two really teensy
809 decoders. para_mp3dec is based on libmad, para_oggdec requires
810 libvorbisfile.
811 - ovsend/ovrecv are capable of streaming ogg as well as mp3, so
812 they are now called para_send and para_recv respectively.
813 - documentation updates
814 - para_gui is themable. For now there is the default theme that
815 looks as before and the simple theme: blue and easy.
816 - gui: audio streaming is now handled by audiod. Time display shows
817 playback time rather than streaming time
818 - slider: update to libzmw-0.2.0
819 - para_krell: fix crash on server shutdown
820 - switch from gzip to bzip2
821
822
823 ----------------------------------------
824 0.1.7 (2005-04-18) "melting penetration"
825 ----------------------------------------
826
827 The main change in this release is clearly the oggvorbis rewrite,
828 but there are also lots of smaller changes. If you intend to use both
829 the mp3 and the ogg plugin, it is recommended to use software mixing,
830 e.g. the dmix plugin which is provided by ALSA.
831
832 - new executables: para_ovsend and para_ovrecv for sending/receiving
833 oggvorbis files via rtp. Requires the open rtp library. Get it at
834 http://www.linphone.org/ortp/
835 - rewrite of the ogg_vorbis core code
836 - configure detects libzmw and, if detected, includes
837 para_slider to the list of binaries to be built by make
838 - server stream writers read from their associated fifo rather
839 than from stdin
840 - slider: two new sliders, lastplayed and numplayed
841 - fix nasty double free bug which caused random segfaults in case of
842 mp3 files with invalid header information
843 - gui: new command line option: --stream_timeout=seconds to
844 deactivate a slot if it is idle for that many seconds (default=`5')
845 - diffstats
846
847
848 ---------------------------------------
849 0.1.6 (2005-03-05) "asymptotic balance"
850 ---------------------------------------
851
852 Only little user-visible changes in this release. Mainly bugfixes and
853 core code cleanup. This is probably the most stable version ever if you
854 stick to mp3...
855
856 - fix several memory leaks
857 - rename default name of mysql database from "music" to "paraslash".
858 Use para_server's --mysql_database option if you do not want to
859 rename your old database.
860 - rework ogg vorbis code
861 - make update command work on mysql servers with LOCAL_INFILE
862 disabled
863 - gui: improved stream I/O (slots)
864 - simplified audio format API
865 - para_pob_ogg is gone
866
867 ------------------------------------
868 0.1.5 (2004-12-31) "opaque eternity"
869 ------------------------------------
870
871 Let's slide gently into the new year.
872
873 - new: para_slider (not built automatically, type "make
874 para_slider" to build). A toy for those who always felt that
875 creating stream definitions is difficult. See screenshots,
876 README and FEATURES for more info.
877 - improved signal handling. Fixes server segfault on SIGHUP
878 for linux kernels newer than Aug 24 2004 and makes para_gui
879 race-free.
880 - reload database tool on SIGHUP
881 - improved help message for sl
882 - do not log "broken pipe" messages as errors. They are
883 perfectly ok.
884 - fix wrong error message on permission errors
885
886 -----------------------------------------
887 0.1.4 (2004-12-19) "tunneling transition"
888 -----------------------------------------
889
890 Bugfix release. As expected, 0.1.3 introduced a bunch of new bugs.
891 Hopefully, most of them got wiped out with this release. Some
892 enhancements went also in.
893
894 - improved error diagnostics for all commands
895 - stradd/picadd: overwrite previous contents if entry already
896 exists, rather than returning errors
897 - stradd: use current stream if invoked without args
898 - faster (and hopefully more stable) ogg-vorbis handling
899 - para_krell: reap children to avoid zombie-flooding in case
900 no server is running
901 - si: report also server pid
902 - server: don't busy-loop if dbtool reports only invalid files.
903 - gui: CTRL+C works again, fix stream_read command line option
904 - fix pic_add, hist
905 - fix mysql dbtool startup in case no database exists
906 - many small fixes and cleanups
907
908 ---------------------------------------
909 0.1.3: (2004-12-10) "vanishing inertia"
910 ---------------------------------------
911
912 Starting from this release, the database tools are integrated in the
913 server binary. This decreases server startup time, reduces code size
914 and speeds up database commands. However, the layout of the underlying
915 mysql database changed only slightly and 0.1.3 should be backwards
916 compatible in that respect.
917
918 Visible changes:
919
920 - If mysql is not detected at compile time, or fails to init
921 at runtime, fall back to the dopey database tool which should
922 always work.
923 - para_dbtool and dbtool.conf are gone. All mysql specific
924 options are read from server.conf and are prefixed by 'mysql_'.
925 - new command: cdt (change database tool)
926 - new command line option: dbtool (choose startup database tool)
927 - The name of current stream is now stored in the database,
928 so paraslash remembers its current stream when restarted.
929 - new command: csp (change stream and play)
930 - para_gui also reports current database tool and server uptime
931
932
933 -------------------------------------------
934 0.1.2: (2004-11-28) "spherical fluctuation"
935 -------------------------------------------
936
937 Point release before the big dbtool changes go in.
938
939 - dbtool: rename ca to cam (copy all meta data). It now also
940 copies numplayed and lastplayed time as well as the picture
941 id.
942 - fix endless-loop-bug caused by mp3 files with invalid header
943
944 -----------------------------------------
945 0.1.1: (2004-11-05) "floating atmosphere"
946 -----------------------------------------
947
948 - gkrellm plugin
949 - new dbtool command: mbox. Browse your sound-file collection
950 with your favorite mail reader.
951 - several small fixes
952
953 -------------------------------------
954 0.1.0: (2204-10-22) "rotating cortex"
955 -------------------------------------
956
957 - fix logging bug for loglevel > VERBOSE
958 - fix skip command
959 - correct timings for vbr mp3s
960 - modular audio format support
961 - ogg-vorbis support (experimental)
962 - new server option: autoplay
963
964 -----------------------------------------
965 0.0.99: (2004-07-25) "harmonic deviation"
966 -----------------------------------------
967
968 - rename projectname from icc to paraslash (play, archive, rate
969 and stream large audio sets happily)
970 - paraslash is no longer restricted to one particular audio
971 streaming software
972 - new dbtool commands (stradd, strq, strdel) for easy stream
973 managment w/o configuration file. That obsoletes stream_defs
974 file/config option for dbtool.
975 - picadd accepts jpeg data from stdin
976 - new server commands: ps (select previous stream), sc (song change)
977 - new default pictures for sdl_gui
978 - gui: new key_map option for binding commands and internal
979 functions to arbitrary keys, nice help screen, rip out
980 soundcard/linux specific stuff, avoid noise artefacts while jumping,
981 show silly logo on startup
982 - new executables: para_fade for fading volume, para_dbadm for
983 manipulating attributes
984 - cdb adds _all_ tables to mysql database
985 - revised and beautified documentation
986 - sample dbtool rewritten in C
987 - autoconf
988
989 ---------------------------------------------
990 0.0.98: (2003-12-26) "incremental smoothness"
991 ---------------------------------------------
992
993 - kick icecast in favour of poc. That removes some races and reduces
994 core code considerably.
995 - cbr/vbr is displayed by stat and gui/sdl_gui. New status flags
996 give finer info on afs' status.
997 - gui can start decoder (see config options). Further new gui
998 commands: refresh (^L), jmp (F1-F10)
999 - gui rereads conf on SIGUSR1 instead of SIGHUP. SIGHUP
1000 terminates gui. This fixes dead instances consuming memory
1001 continuously.
1002 - new dbtool command: verb for sending verbatim sql queries.
1003 - fix pid_list races (by removing pid_list)
1004 - codename funnies
1005
1006 --------------------
1007 0.0.97: (2003-10-26)
1008 --------------------
1009
1010 - installation prefix now defaults to /usr/local
1011 - new commands for gui: snozze, sleep and reread config
1012 - config file for gui and sdl_gui
1013 - fix problems with filenames containing funny characters
1014 (reported by Thomas Forell)
1015 - improved signal handling for gui, now it rereads conf on SIGHUP
1016 - new dbtool command: cdb (create database)
1017 - switch from argtable to gengetopt
1018 - major code cleanup and speed improvements
1019 - fix several potential buffer overflows
1020 - many small fixes and cleanups
1021
1022 -------------------
1023 0.0.96 (2003-08-30)
1024 -------------------
1025
1026 - easy stream_defs syntax
1027 - sdl_gui can display images associated to the file being played
1028 - Major feature enhancements for icc_gui including dynamic text
1029 placement and the top/bottom window design
1030 - vrfy/clean now also checks for NULL values in attributes as
1031 well as for invalid picture pointers
1032 - fix long outstanding case sensitivity bug
1033 - many small fixes and cleanups
1034
1035 -------------------
1036 0.0.95 (2003-06-29)
1037 -------------------
1038
1039 - sdl gui runs much faster
1040 - new dbtool command: ca (copy attributes)
1041 - count and display number of times the song has been played
1042 - new feature: scoring
1043 - command line options for sdl_gui
1044 - simpler syntax of streams file
1045 - decrease network traffic of stat
1046 - fix zombie bug
1047 - many small fixes and cleanups
1048
1049 -------------------
1050 0.0.94 (2003-05-04)
1051 -------------------
1052
1053 - new server command: ns (next stream)
1054 - new icc_gui command: c (change stream)
1055 - internal mp3info
1056 - stat shows also id3 tag info
1057 - new sdl based gui
1058 - log flodding bug fixed
1059 - many small fixes and cleanups
1060
1061 -------------------
1062 0.0.93 (2003-03-28)
1063 -------------------
1064
1065 - colors for icc_gui
1066 - icc_gui sets volume directly (linux only)
1067 - proper locking that fixes some races
1068 - fix security bug that caused commands to be executed even
1069 with unsufficient permissions
1070 - new command: hup to make all servers reread their configuration file
1071 - icecast meta data streaming
1072 - many small fixes and cleanups