5ceeac08af48642095e8b6c4d6fa7d0f2799fb22
[paraslash.git] / web / manual.m4
1 dnl To generate the html version, execute
2 dnl m4 web/manual.m4 | grutatxt --toc
3
4 define(`LOCAL_LINK_NAME', `translit(`$1', `A-Z
5 ', `a-z__')')
6 define(`REMOVE_NEWLINE', `translit(`$1',`
7 ', ` ')')
8
9 define(`REFERENCE', ./``#''`LOCAL_LINK_NAME($1)' (`REMOVE_NEWLINE($2)'))
10 define(`XREFERENCE', `$1' (`REMOVE_NEWLINE($2)'))
11 define(`EMPH', ``_''`REMOVE_NEWLINE($1)'``_'')
12
13 Paraslash user manual
14 =====================
15
16 This document describes how to install, configure and use the paraslash
17 network audio streaming system. Most chapters start with a chapter
18 overview and conclude with an example section. We try to focus on
19 general concepts and on the interaction of the various pieces of the
20 paraslash package. Hence this user manual is not meant as a replacement
21 for the manual pages that describe all command line options of each
22 paraslash executable.
23
24 ------------
25 Introduction
26 ------------
27
28 In this chapter we give an REFERENCE(Overview, overview) of the
29 interactions of the two main programs contained in the paraslash
30 package, followed by REFERENCE(The paraslash executables, brief
31 descriptions) of all executables.
32
33 Overview
34 ~~~~~~~~
35
36 The core functionality of the para suite is provided by two main
37 executables, para_server and para_audiod. The former maintains a
38 database of audio files and streams these files to para_audiod which
39 receives and plays the stream.
40
41 In a typical setting, both para_server and para_audiod act as
42 background daemons whose functionality is controlled by client
43 programs: the para_audioc client controls para_audiod over a local
44 socket while the para_client program connects to para_server over a
45 local or remote networking connection.
46
47 Typically, these two daemons run on different hosts but a local setup
48 is also possible.
49
50 A simplified picture of a typical setup is as follows
51 <<
52 <pre>
53 server_host client_host
54 ~~~~~~~~~~~ ~~~~~~~~~~~
55
56 +-----------+ audio stream +-----------+
57 |para_server| -----------------------------> |para_audiod|
58 +-----------+ +-----------+
59 ^ ^
60 | |
61 | | connect
62 | |
63 | |
64 | +-----------+
65 | |para_audioc|
66 | +-----------+
67 |
68 |
69 | connect +-----------+
70 +-------------------------------------- |para_client|
71 +-----------+
72 </pre>
73 >>
74
75 The paraslash executables
76 ~~~~~~~~~~~~~~~~~~~~~~~~~
77
78 *para_server*
79
80 para_server streams binary audio data (MP3, OGG/Vorbis, OGG/Speex,
81 M4A, WMA files) over local and/or remote networks. It listens on a
82 TCP port and accepts commands such as play, stop, pause, next from
83 authenticated clients. There are many more commands though, see the
84 man page of para_server for a description of all commands.
85
86 It supports three built-in network streaming protocols
87 (senders/receivers): HTTP, DCCP, or UDP. This is explained in more
88 detail in the section on REFERENCE(Networking, networking).
89
90 The built-in audio file selector of paraslash is used to manage your
91 audio files. It maintains statistics on the usage of all available
92 audio files such as last-played time, and the number of times each
93 file was selected.
94
95 Additional information may be added to the database to allow
96 fine-grained selection based on various properties of the audio file,
97 including information found in (ID3) tags. However, old-fashioned
98 playlists are also supported.
99
100 It is also possible to store images (album covers) and lyrics in the
101 database and associate these to the corresponding audio files.
102
103 The section on the REFERENCE(The audio file selector, audio file
104 selector) discusses this topic.
105
106
107 *para_client*
108
109 The client program to connect to para_server. paraslash commands
110 are sent to para_server and the response is dumped to STDOUT. This
111 can be used by any scripting language to produce user interfaces with
112 little programming effort.
113
114 All connections between para_server and para_client are encrypted
115 with a symmetric RC4 session key. For each user of paraslash you must
116 create a public/secret RSA key pair for authentication.
117
118 If para_client is started without non-option arguments, an interactive
119 session (shell) is started. Command history and command completion are
120 supported through libreadline.
121
122 *para_audiod*
123
124 The local daemon that collects information from para_server.
125
126 It runs on the client side and connects to para_server. As soon as
127 para_server announces the availability of an audio stream, para_audiod
128 starts an appropriate receiver, any number of filters and a paraslash
129 writer to play the stream.
130
131 Moreover, para_audiod listens on a local socket and sends status
132 information about para_server and para_audiod to local clients on
133 request. Access via this local socket may be restricted by using Unix
134 socket credentials, if available.
135
136
137 *para_audioc*
138
139 The client program which talks to para_audiod. Used to control
140 para_audiod, to receive status info, or to grab the stream at any
141 point of the decoding process. Like para_client, para_audioc supports
142 interactive sessions on systems with libreadline.
143
144 *para_recv*
145
146 A command line HTTP/DCCP/UDP stream grabber. The http mode is
147 compatible with arbitrary HTTP streaming sources (e.g. icecast).
148 In addition to the three network streaming modes, para_recv can also
149 operate in local (afh) mode. In this mode it writes the content of
150 an audio file on the local file system in complete chunks to stdout,
151 optionally 'just in time'. This allows to cut an audio file without
152 first decoding it, and it enables third-party software which is unaware
153 of the particular audio format to send complete frames in real time.
154
155 *para_filter*
156
157 A filter program that reads from STDIN and writes to STDOUT.
158 Like para_recv, this is an atomic building block which can be used to
159 assemble higher-level audio receiving facilities. It combines several
160 different functionalities in one tool: decoders for multiple audio
161 formats and a number of processing filters, among these a normalizer
162 for audio volume.
163
164 *para_afh*
165
166 A small stand-alone program that prints tech info about the given
167 audio file to STDOUT. It can be instructed to print a "chunk table",
168 an array of offsets within the audio file.
169
170 *para_write*
171
172 A modular audio stream writer. It supports a simple file writer
173 output plug-in and optional WAV/raw players for ALSA (Linux) and for
174 coreaudio (Mac OS). para_write can also be used as a stand-alone WAV
175 or raw audio player.
176
177
178 *para_gui*
179
180 Curses-based gui that presents status information obtained in a curses
181 window. Appearance can be customized via themes. para_gui provides
182 key-bindings for the most common server commands and new key-bindings
183 can be added easily.
184
185
186 *para_fade*
187
188 An alarm clock and volume-fader for OSS and ALSA.
189
190 -----------
191 Quick start
192 -----------
193
194 This chapter lists the REFERENCE(Requirements, necessary software)
195 that must be installed to compile the paraslash package, describes
196 how to REFERENCE(Installation, compile and install) the paraslash
197 source code and the steps that have to be performed in order to
198 REFERENCE(Quick start, set up) a typical server and client.
199
200 Requirements
201 ~~~~~~~~~~~~
202
203 In any case you'll need
204
205 - XREFERENCE(http://systemlinux.org/~maan/osl/, libosl).
206 The _object storage layer_ library is used by para_server. To
207 clone the source code repository, execute
208
209 git clone git://git.tuebingen.mpg.de/osl
210
211 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gcc, gcc) or
212 XREFERENCE(http://clang.llvm.org, clang). All gcc versions
213 >= 3.3 are currently supported. Clang version 1.1 or newer
214 should work as well.
215
216 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/make, gnu make) is
217 also shipped with the disto. On BSD systems the gnu make
218 executable is often called gmake.
219
220 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/bash, bash). Some
221 scripts which run during compilation require the EMPH(Bourne
222 again shell). It is most likely already installed.
223
224 - XREFERENCE(http://www.openssl.org/, openssl) or
225 XREFERENCE(ftp://ftp.gnupg.org/gcrypt/libgcrypt/, libgcrypt).
226 At least one of these two libraries is needed as the backend
227 for cryptographic routines on both the server and the client
228 side. Both openssl and libgcrypt are usually shipped with the
229 distro, but you might have to install the development package
230 (libssl-dev or libgcrypt-dev on debian systems) as well.
231
232 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gengetopt/, gengetopt)
233 is needed to generate the C code for the command line parsers
234 of all paraslash executables.
235
236 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/help2man, help2man)
237 is used to create the man pages.
238
239 Optional:
240
241 - XREFERENCE(http://www.underbit.com/products/mad/, libmad).
242 To compile in MP3 support for paraslash, the development
243 package must be installed. It is called libmad0-dev on
244 debian-based systems. Note that libmad is not necessary on
245 the server side, i.e. for sending MP3 files.
246
247 - XREFERENCE(http://www.underbit.com/products/mad/,
248 libid3tag). For version-2 ID3 tag support, you'll need
249 the libid3tag development package libid3tag0-dev. Without
250 libid3tag, only version one tags are recognized.
251
252 - XREFERENCE(http://www.xiph.org/downloads/, ogg vorbis).
253 For ogg vorbis streams you'll need libogg, libvorbis,
254 libvorbisfile. The corresponding Debian packages are called
255 libogg-dev and libvorbis-dev.
256
257 - XREFERENCE(http://www.audiocoding.com/, libfaad). For aac
258 files (m4a) you'll need libfaad (libfaad-dev).
259
260 - XREFERENCE(http://www.speex.org/, speex). In order to stream
261 or decode speex files, libspeex (libspeex-dev) is required.
262
263 - XREFERENCE(http://flac.sourceforge.net/, flac). To stream
264 or decode files encoded with the _Free Lossless Audio Codec_,
265 libFLAC (libFLAC-dev) must be installed.
266
267 - XREFERENCE(ftp://ftp.alsa-project.org/pub/lib/, alsa-lib). On
268 Linux, you'll need to have ALSA's development package
269 libasound2-dev installed.
270
271 - XREFERENCE(http://downloads.xiph.org/releases/ao/,
272 libao). Needed to build the ao writer (ESD, PulseAudio,...).
273 Debian package: libao-dev.
274
275 - XREFERENCE(http://cnswww.cns.cwru.edu/php/chet/readline/rltop.html,
276 GNU Readline). If this library (libreadline-dev) is installed,
277 para_client and para_audioc support interactive sessions.
278
279 Installation
280 ~~~~~~~~~~~~
281
282 First make sure all non-optional packages listed in the section on
283 REFERENCE(Requirements, required software) are installed on your
284 system.
285
286 You don't need everything listed there. In particular, MP3, OGG/Vorbis,
287 OGG/Speex and AAC support are all optional. The configure script will
288 detect what is installed on your system and will only try to build
289 those executables that can be built with your setup.
290
291 Note that no special decoder library (not even the MP3 decoding library
292 libmad) is needed for para_server if you only want to stream MP3 or WMA
293 files. Also, it's fine to use para_server on a box without sound card.
294
295 Next, install the paraslash package on all machines, you'd like this
296 software to run on. If you compile from a released tarball, execute
297
298 (./configure && make) > /dev/null
299
300 When compiling from git or from snapshots downloaded via gitweb,
301 the above command will not work because the configure script is not
302 included in the git repository. In this case the following command
303 should be used instead:
304
305 ./autogen.sh
306
307 This runs autoconf to generate the configure script, then runs it as
308 above. Therefore you'll need autoconf for this to work.
309
310 There should be no errors but probably some warnings about missing
311 packages which usually implies that not all audio formats will be
312 supported. If headers or libs are installed at unusual locations you
313 might need to tell the configure script where to find them. Try
314
315 ./configure --help
316
317 to see a list of options. If the paraslash package was compiled
318 successfully, execute (optionally)
319
320 make test
321
322 to run the paraslash test suite. If all tests pass, execute as root
323
324 make install
325
326 to install executables under /usr/local/bin and the man pages under
327 /usr/local/man.
328
329 Configuration
330 ~~~~~~~~~~~~~
331
332 *Step 1*: Create a paraslash user
333
334 In order to control para_server at runtime you must create a paraslash
335 user. As authentication is based on the RSA crypto system you'll have
336 to create an RSA key pair. If you already have a user and an RSA key
337 pair, you may skip this step.
338
339 In this section we'll assume a typical setup: You would like to run
340 para_server on some host called server_host as user foo, and you want
341 to connect to para_server from another machine called client_host as
342 user bar.
343
344 As foo@server_host, create ~/.paraslash/server.users by typing the
345 following commands:
346
347 user=bar
348 target=~/.paraslash/server.users
349 key=~/.paraslash/id_rsa.pub.$user
350 perms=AFS_READ,AFS_WRITE,VSS_READ,VSS_WRITE
351 mkdir -p ~/.paraslash
352 echo "user $user $key $perms" >> $target
353
354 Next, change to the "bar" account on client_host and generate the
355 key pair with the commands
356
357 ssh-keygen -t rsa -b 2048
358 # hit enter twice to create a key with no passphrase
359
360 This generates the two files id_rsa and id_rsa.pub in ~/.ssh. Note
361 that paraslash can also read keys generated by the "openssl genrsa"
362 command. However, since keys created with ssh-keygen can also be used
363 for ssh, this method is recommended.
364
365 Note that para_server refuses to use a key if it is shorter than 2048
366 bits. In particular, the RSA keys of paraslash 0.3.x will not work
367 with version 0.4.x. Moreover, para_client refuses to use a (private)
368 key which is world-readable.
369
370 para_server only needs to know the public key of the key pair just
371 created. Copy this public key to server_host:
372
373 src=~/.ssh/id_rsa.pub
374 dest=.paraslash/id_rsa.pub.$LOGNAME
375 scp $src foo@server_host:$dest
376
377 Finally, tell para_client to connect to server_host:
378
379 conf=~/.paraslash/client.conf
380 echo 'hostname server_host' > $conf
381
382
383 *Step 2*: Start para_server
384
385 Before starting the server make sure you have write permissions to
386 the directory /var/paraslash that has been created during installation:
387
388 sudo chown $LOGNAME /var/paraslash
389
390 Alternatively, use the --afs_socket Option to specify a different
391 location for the AFS command socket.
392
393 For this first try, we'll use the info loglevel to make the output
394 of para_server more verbose.
395
396 para_server -l info
397
398 Now you can use para_client to connect to the server and issue
399 commands. Open a new shell as bar@client_host and try
400
401 para_client help
402 para_client si
403
404 to retrieve the list of available commands and some server info.
405 Don't proceed if this doesn't work.
406
407 *Step 3*: Create and populate the database
408
409 An empty database is created with
410
411 para_client init
412
413 This initializes a couple of empty tables under
414 ~/.paraslash/afs_database-0.4. You normally don't need to look at these
415 tables, but it's good to know that you can start from scratch with
416
417 rm -rf ~/.paraslash/afs_database-0.4
418
419 in case something went wrong.
420
421 Next, you need to add some audio files to that database so that
422 para_server knows about them. Choose an absolute path to a directory
423 containing some audio files and add them to the audio file table:
424
425 para_client add /my/mp3/dir
426
427 This might take a while, so it is a good idea to start with a directory
428 containing not too many files. Note that the table only contains data
429 about the audio files found, not the files themselves.
430
431 You may print the list of all known audio files with
432
433 para_client ls
434
435 *Step 4*: Configure para_audiod
436
437 para_audiod needs to create a "well-known" socket for the clients to
438 connect to. The default path for this socket is
439
440 /var/paraslash/audiod_socket.$HOSTNAME
441
442 In order to make this directory writable for para_audiod, execute
443 as bar@client_host
444
445 sudo chown $LOGNAME /var/paraslash
446
447
448 We will also have to tell para_audiod that it should receive the
449 audio stream from server_host via http:
450
451 para_audiod -l info -r '.:http -i server_host'
452
453 You should now be able to listen to the audio stream once para_server
454 starts streaming. To activate streaming, execute
455
456 para_client play
457
458 Since no playlist has been specified yet, the "dummy" mode which
459 selects all known audio files is activated automatically. See the
460 section on the REFERENCE(The audio file selector, audio file selector)
461 for how to use playlists and moods to specify which files should be
462 streamed in which order.
463
464 *Troubleshooting*
465
466 It did not work? To find out why, try to receive, decode and play the
467 stream manually using para_recv, para_filter and para_write as follows.
468
469 For simplicity we assume that you're running Linux/ALSA and that only
470 MP3 files have been added to the database.
471
472 para_recv -r 'http -i server_host' > file.mp3
473 # (interrupt with CTRL+C after a few seconds)
474 ls -l file.mp3 # should not be empty
475 para_filter -f mp3dec -f wav < file.mp3 > file.wav
476 ls -l file.wav # should be much bigger than file.mp3
477 para_write -w alsa < file.wav
478
479 Double check what is logged by para_server and use the --loglevel
480 option of para_recv, para_filter and para_write to increase verbosity.
481
482 ---------------
483 User management
484 ---------------
485
486 para_server uses a challenge-response mechanism to authenticate
487 requests from incoming connections, similar to ssh's public key
488 authentication method. Authenticated connections are encrypted using
489 the RC4 stream cipher.
490
491 In this chapter we briefly describe RSA and RC4 and sketch the
492 REFERENCE(Client-server authentication, authentication handshake)
493 between para_client and para_server. User management is discussed
494 in the section on REFERENCE(The user_list file, the user_list file).
495 These sections are all about communication between the client and the
496 server. Connecting para_audiod is a different matter and is described
497 in a REFERENCE(Connecting para_audiod, separate section).
498
499
500
501 RSA and RC4
502 ~~~~~~~~~~~
503
504 RSA is an asymmetric block cipher which is used in many applications,
505 including ssh and gpg. An RSA key consists in fact of two keys,
506 called the public key and the private key. A message can be encrypted
507 with either key and only the counterpart of that key can decrypt
508 the message. While RSA can be used for both signing and encrypting
509 a message, paraslash uses RSA only for the latter purpose. The
510 RSA public key encryption and signatures algorithms are defined in
511 detail in RFC 2437.
512
513 RC4 is a stream cipher, i.e. the input is XORed with a pseudo-random
514 key stream to produce the output. Decryption uses the same function
515 calls as encryption. While RC4 supports variable key lengths,
516 paraslash uses a fixed length of 256 bits, which is considered a
517 strong encryption by today's standards. Since the same key must never
518 be used twice, a different, randomly-generated key is used for every
519 new connection.
520
521 Client-server authentication
522 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
523
524 The authentication handshake between para_client and para_server goes
525 as follows:
526
527 - para_client connects to para_server and sends an
528 authentication request for a user. It does so by connecting
529 to TCP port 2990 of the server host. This port is called the
530 para_server _control port_.
531
532 - para_server accepts the connection and forks a child process
533 which handles the incoming request. The parent process keeps
534 listening on the control port while the child process (also
535 called para_server below) continues as follows.
536
537 - para_server loads the RSA public key of that user, fills a
538 fixed-length buffer with random bytes, encrypts that buffer
539 using the public key and sends the encrypted buffer to the
540 client. The first part of the buffer is the challenge which
541 is used for authentication while the second part is the RC4
542 session key.
543
544 - para_client receives the encrypted buffer and decrypts it
545 with the user's private key, thereby obtaining the challenge
546 buffer and the session key. It sends the SHA1 hash value of
547 the challenge back to para_server and stores the session key
548 for further use.
549
550 - para_server also computes the SHA1 hash of the challenge
551 and compares it against what was sent back by the client.
552
553 - If the two hashes do not match, the authentication has
554 failed and para_server closes the connection.
555
556 - Otherwise the user is considered authenticated and the client
557 is allowed to proceed by sending a command to be executed. From
558 this point on the communication is encrypted using the RC4
559 stream cipher with the session key known to both peers.
560
561 paraslash relies on the quality of the pseudo-random bytes provided
562 by the crypto library (openssl or libgcrypt), on the security of
563 the implementation of the RSA and RC4 crypto routines and on the
564 infeasibility to invert the SHA1 function.
565
566 Neither para_server or para_client create RSA keys on their own. This
567 has to be done once for each user as sketched in REFERENCE(Quick start,
568 Quick start) and discussed in more detail REFERENCE(The user_list
569 file, below).
570
571 The user_list file
572 ~~~~~~~~~~~~~~~~~~
573
574 At startup para_server reads the user list file which contains one
575 line per user. The default location of the user list file may be
576 changed with the --user_list option.
577
578 There should be at least one user in this file. Each user must have
579 an RSA key pair. The public part of the key is needed by para_server
580 while the private key is needed by para_client. Each line of the
581 user list file must be of the form
582
583 user <username> <key> <perms>
584
585 where _username_ is an arbitrary string (usually the user's login
586 name), _key_ is the full path to that user's public RSA key, and
587 _perms_ is a comma-separated list of zero or more of the following
588 permission bits:
589
590 +---------------------------------------------------------+
591 | AFS_READ | read the contents of the databases |
592 +-----------+---------------------------------------------+
593 | AFS_WRITE | change database contents |
594 +-----------+---------------------------------------------+
595 | VSS_READ | obtain information about the current stream |
596 +-----------+---------------------------------------------+
597 | VSS_WRITE | change the current stream |
598 +---------------------------------------------------------+
599
600 The permission bits specify which commands the user is allowed to
601 execute. The output of
602
603 para_client help
604
605 contains in the third column the permissions needed to execute the
606 command.
607
608 It is possible to make para_server reread the user_list file by
609 executing the paraslash "hup" command or by sending SIGHUP to the
610 PID of para_server.
611
612
613 Connecting para_audiod
614 ~~~~~~~~~~~~~~~~~~~~~~
615
616 para_audiod listens on a Unix domain socket. Those sockets are
617 for local communication only, so only local users can connect to
618 para_audiod. The default is to let any user connect but this can be
619 restricted on platforms that support UNIX socket credentials which
620 allow para_audiod to obtain the Unix credentials of the connecting
621 process.
622
623 Use para_audiod's --user_allow option to allow connections only for
624 a limited set of users.
625
626 -----------------------
627 The audio file selector
628 -----------------------
629
630 paraslash comes with a sophisticated audio file selector (AFS),
631 whose main task is to determine which file to stream next, based on
632 information on the audio files stored in a database. It communicates
633 also with para_client whenever an AFS command is executed, for example
634 to answer a database query.
635
636 Besides the traditional playlists, AFS supports audio file selection
637 based on _moods_ which act as a filter that limits the set of all
638 known audio files to those which satisfy certain criteria. It also
639 maintains tables containing images (e.g. album cover art) and lyrics
640 that can be associated with one or more audio files.
641
642 AFS uses XREFERENCE(http://systemlinux.org/~maan/osl/, libosl), the
643 object storage layer library, as the backend library for storing
644 information on audio files, playlists, etc. This library offers
645 functionality similar to a relational database, but is much more
646 lightweight than a full database backend.
647
648 In this chapter we sketch the setup of the REFERENCE(The AFS process,
649 AFS process) during server startup and proceed with the description
650 of the REFERENCE(Database layout, layout) of the various database
651 tables. The section on REFERENCE(Playlists and moods, playlists
652 and moods) explains these two audio file selection mechanisms
653 in detail and contains pratical examples. The way REFERENCE(File
654 renames and content changes, file renames and content changes) are
655 detected is discussed briefly before the REFERENCE(Troubleshooting,
656 Troubleshooting) section concludes the chapter.
657
658 The AFS process
659 ~~~~~~~~~~~~~~~
660
661 On startup, para_server forks to create the AFS process which opens
662 the OSL database tables. The server process communicates with the
663 AFS process via pipes and shared memory. Usually, the AFS process
664 awakes only briefly whenever the current audio file changes. The AFS
665 process determines the next audio file, opens it, verifies it has
666 not been changed since it was added to the database and passes the
667 open file descriptor to the server process, along with audio file
668 meta-data such as file name, duration, audio format and so on. The
669 server process then starts to stream the audio file.
670
671 The AFS process also accepts connections from local clients via
672 a well-known socket. However, only child processes of para_server
673 may connect through this socket. All server commands that have the
674 AFS_READ or AFS_WRITE permission bits use this mechanism to query or
675 change the database.
676
677 Database layout
678 ~~~~~~~~~~~~~~~
679
680 *The audio file table*
681
682 This is the most important and usually also the largest table of the
683 AFS database. It contains the information needed to stream each audio
684 file. In particular the following data is stored for each audio file.
685
686 - SHA1 hash value of the audio file contents. This is computed
687 once when the file is added to the database. Whenever AFS
688 selects this audio file for streaming the hash value is
689 recomputed and checked against the value stored in the
690 database to detect content changes.
691
692 - The time when this audio file was last played.
693
694 - The number of times the file has been played so far.
695
696 - The attribute bitmask.
697
698 - The image id which describes the image associated with this
699 audio file.
700
701 - The lyrics id which describes the lyrics associated with
702 this audio file.
703
704 - The audio format id (MP3, OGG, ...).
705
706 - An amplification value that can be used by the amplification
707 filter to pre-amplify the decoded audio stream.
708
709 - The chunk table. It describes the location and the timing
710 of the building blocks of the audio file. This is used by
711 para_server to send chunks of the file at appropriate times.
712
713 - The duration of the audio file.
714
715 - Tag information contained in the audio file (ID3 tags,
716 Vorbis comments, ...).
717
718 - The number of channels
719
720 - The encoding bitrate.
721
722 - The sampling frequency.
723
724 To add or refresh the data contained in the audio file table, the _add_
725 command is used. It takes the full path of either an audio file or a
726 directory. In the latter case, the directory is traversed recursively
727 and all files which are recognized as valid audio files are added to
728 the database.
729
730 *The attribute table*
731
732 The attribute table contains two columns, _name_ and _bitnum_. An
733 attribute is simply a name for a certain bit number in the attribute
734 bitmask of the audio file table.
735
736 Each of the 64 bits of the attribute bitmask can be set for each
737 audio file individually. Hence up to 64 different attributes may be
738 defined. For example, "pop", "rock", "blues", "jazz", "instrumental",
739 "german_lyrics", "speech", whatever. You are free to choose as
740 many attributes as you like and there are no naming restrictions
741 for attributes.
742
743 A new attribute "test" is created by
744
745 para_client addatt test
746 and
747 para_client lsatt
748
749 lists all available attributes. You can set the "test" attribute for
750 an audio file by executing
751
752 para_client setatt test+ /path/to/the/audio/file
753
754 Similarly, the "test" bit can be removed from an audio file with
755
756 para_client setatt test- /path/to/the/audio/file
757
758 Instead of a path you may use a shell wildcard pattern. The attribute
759 is applied to all audio files matching this pattern:
760
761 para_client setatt test+ '/test/directory/*'
762
763 The command
764
765 para_client -- ls -lv
766
767 gives you a verbose listing of your audio files also showing which
768 attributes are set.
769
770 In case you wonder why the double-dash in the above command is needed:
771 It tells para_client to not interpret the options after the dashes. If
772 you find this annoying, just say
773
774 alias para='para_client --'
775
776 and be happy. In what follows we shall use this alias.
777
778 The "test" attribute can be dropped from the database with
779
780 para rmatt test
781
782 Read the output of
783
784 para help ls
785 para help setatt
786
787 for more information and a complete list of command line options to
788 these commands.
789
790 *Blob tables*
791
792 The image, lyrics, moods and playlists tables are all blob tables.
793 Blob tables consist of three columns each: The identifier which is
794 a positive non-negative number that is auto-incremented, the name
795 (an arbitrary string) and the content (the blob).
796
797 All blob tables support the same set of actions: cat, ls, mv, rm
798 and add. Of course, _add_ is used for adding new blobs to the table
799 while the other actions have the same meaning as the corresponding
800 Unix commands. The paraslash commands to perform these actions are
801 constructed as the concatenation of the table name and the action. For
802 example addimg, catimg, lsimg, mvimg, rmimg are the commands that
803 manipulate or query the image table.
804
805 The add variant of these commands is special as these commands read
806 the blob contents from stdin. To add an image to the image table the
807 command
808
809 para addimg image_name < file.jpg
810
811 can be used.
812
813 Note that the images and lyrics are not interpreted at all, and also
814 the playlist and the mood blobs are only investigated when the mood
815 or playlist is activated with the select command.
816
817 *The score table*
818
819 Unlike all other tables the contents of the score table remain in
820 memory and are never stored on disk. The score table contains two
821 columns: The SHA1 hash value (of an audio file) and its current
822 score.
823
824 However, only those files which are admissible for the current mood
825 or playlist are contained in the score table. The audio file selector
826 always chooses the row with the highest score as the file to stream
827 next. While doing so, it computes the new score and updates the
828 last_played and the num_played fields in the audio file table.
829
830 The score table is recomputed by the select command which loads a
831 mood or playlist. Audio files are chosen for streaming from the rows
832 of the score table on a highest-score-first basis.
833
834
835 Playlists and moods
836 ~~~~~~~~~~~~~~~~~~~
837
838 Playlists and moods offer two different ways of specifying the set of
839 admissible files. A playlist in itself describes a set of admissible
840 files. A mood, in contrast, describes the set of admissible files in
841 terms of attributes and other type of information available in the
842 audio file table. As an example, a mood can define a filename pattern,
843 which is then matched against the names of audio files in the table.
844
845 *Playlists*
846
847 Playlists are accommodated in the playlist table of the afs database,
848 using the aforementioned blob format for tables. A new playlist is
849 created with the addpl command by specifying the full (absolute)
850 paths of all desired audio files, separated by newlines. Example:
851
852 find /my/mp3/dir -name "*.mp3" | para addpl my_playlist
853
854 If _my_playlist_ already exists it is overwritten. To activate the
855 new playlist, execute
856
857 para select p/my_playlist
858
859 The audio file selector will assign scores to each entry of the list,
860 in descending order so that files will be selected in order. If a
861 file could not be opened for streaming, its entry is removed from
862 the score table (but not from the playlist).
863
864 *Moods*
865
866 A mood consists of a unique name and its *mood definition*, which is
867 a set of *mood lines* containing expressions in terms of attributes
868 and other data contained in the database.
869
870 At any time at most one mood can be *active* which means that
871 para_server is going to select only files from that subset of
872 admissible files.
873
874 So in order to create a mood definition one has to write a set of
875 mood lines. Mood lines come in three flavours: Accept lines, deny
876 lines and score lines.
877
878 The general syntax of the three types of mood lines is
879
880
881 accept [with score <score>] [if] [not] <mood_method> [options]
882 deny [with score <score>] [if] [not] <mood_method> [options]
883 score <score> [if] [not] <mood_method> [options]
884
885
886 Here <score> is either an integer or the string "random" which assigns
887 a random score to all matching files. The score value changes the
888 order in which admissible files are going to be selected, but is of
889 minor importance for this introduction.
890
891 So we concentrate on the first two forms, i.e. accept and deny
892 lines. As usual, everything in square brackets is optional, i.e.
893 accept/deny lines take the following form when ignoring scores:
894
895 accept [if] [not] <mood_method> [options]
896
897 and analogously for the deny case. The "if" keyword is only syntactic
898 sugar and has no function. The "not" keyword just inverts the result,
899 so the essence of a mood line is the mood method part and the options
900 following thereafter.
901
902 A *mood method* is realized as a function which takes an audio file
903 and computes a number from the data contained in the database.
904 If this number is non-negative, we say the file *matches* the mood
905 method. The file matches the full mood line if it either
906
907 - matches the mood method and the "not" keyword is not given,
908 or
909 - does not match the mood method, but the "not" keyword is given.
910
911 The set of admissible files for the whole mood is now defined as those
912 files which match at least one accept mood line, but no deny mood line.
913 More formally, an audio file F is admissible if and only if
914
915 (F ~ AL1 or F ~ AL2...) and not (F ~ DL1 or F ~ DN2 ...)
916
917 where AL1, AL2... are the accept lines, DL1, DL2... are the deny
918 lines and "~" means "matches".
919
920 The cases where no mood lines of accept/deny type are defined need
921 special treatment:
922
923 - Neither accept nor deny lines: This treats all files as
924 admissible (in fact, that is the definition of the dummy mood
925 which is activated automatically if no moods are available).
926
927 - Only accept lines: A file is admissible iff it matches at
928 least one accept line:
929
930 F ~ AL1 or F ~ AL2 or ...
931
932 - Only deny lines: A file is admissible iff it matches no
933 deny line:
934
935 not (F ~ DL1 or F ~ DN2 ...)
936
937
938
939 *List of mood_methods*
940
941 no_attributes_set
942
943 Takes no arguments and matches an audio file if and only if no
944 attributes are set.
945
946 is_set <attribute_name>
947
948 Takes the name of an attribute and matches iff that attribute is set.
949
950 path_matches <pattern>
951
952 Takes a filename pattern and matches iff the path of the audio file
953 matches the pattern.
954
955 artist_matches <pattern>
956 album_matches <pattern>
957 title_matches <pattern>
958 comment_matches <pattern>
959
960 Takes an extended regular expression and matches iff the text of the
961 corresponding tag of the audio file matches the pattern. If the tag
962 is not set, the empty string is matched against the pattern.
963
964 year ~ <num>
965 bitrate ~ <num>
966 frequency ~ <num>
967 channels ~ <num>
968 num_played ~ <num>
969
970 Takes a comparator ~ of the set {<, =, <=, >, >=, !=} and a number
971 <num>. Matches an audio file iff the condition <val> ~ <num> is
972 satisfied where val is the corresponding value of the audio file
973 (value of the year tag, bitrate in kbit/s, frequency in Hz, channel
974 count, play count).
975
976 The year tag is special as its value is undefined if the audio file
977 has no year tag or the content of the year tag is not a number. Such
978 audio files never match. Another difference is the special treatment
979 if the year tag is a two-digit number. In this case either 1900 or
980 2000 is added to the tag value, depending on whether the number is
981 greater than 2000 plus the current year.
982
983
984 *Mood usage*
985
986 To create a new mood called "my_mood", write its definition into
987 some temporary file, say "tmpfile", and add it to the mood table
988 by executing
989
990 para addmood my_mood < tmpfile
991
992 If the mood definition is really short, you may just pipe it to the
993 client instead of using temporary files. Like this:
994
995 echo "$MOOD_DEFINITION" | para addmood my_mood
996
997 There is no need to keep the temporary file since you can always use
998 the catmood command to get it back:
999
1000 para catmood my_mood
1001
1002 A mood can be activated by executing
1003
1004 para select m/my_mood
1005
1006 Once active, the list of admissible files is shown by the ls command
1007 if the "-a" switch is given:
1008
1009 para ls -a
1010
1011
1012 *Example mood definition*
1013
1014 Suppose you have defined attributes "punk" and "rock" and want to define
1015 a mood containing only Punk-Rock songs. That is, an audio file should be
1016 admissible if and only if both attributes are set. Since
1017
1018 punk and rock
1019
1020 is obviously the same as
1021
1022 not (not punk or not rock)
1023
1024 (de Morgan's rule), a mood definition that selects only Punk-Rock
1025 songs is
1026
1027 deny if not is_set punk
1028 deny if not is_set rock
1029
1030
1031
1032 File renames and content changes
1033 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1034
1035 Since the audio file selector knows the SHA1 of each audio file that
1036 has been added to the afs database, it recognizes if the content of
1037 a file has changed, e.g. because an ID3 tag was added or modified.
1038 Also, if a file has been renamed or moved to a different location,
1039 afs will detect that an entry with the same hash value already exists
1040 in the audio file table.
1041
1042 In both cases it is enough to just re-add the new file. In the
1043 first case (file content changed), the audio table is updated, while
1044 metadata such as the num_played and last_played fields, as well as
1045 the attributes, remain unchanged. In the other case, when the file
1046 is moved or renamed, only the path information is updated, all other
1047 data remains as before.
1048
1049 It is possible to change the behaviour of the add command by using the
1050 "-l" (lazy add) or the "-f" (force add) option.
1051
1052 Troubleshooting
1053 ~~~~~~~~~~~~~~~
1054
1055 Use the debug loglevel (option -l debug for most commands) to show
1056 debugging info. Almost all paraslash executables have a brief online
1057 help which is displayed by using the -h switch. The --detailed-help
1058 option prints the full help text.
1059
1060 If para_server crashed or was killed by SIGKILL (signal 9), it
1061 may refuse to start again because of "dirty osl tables". In this
1062 case you'll have to run the oslfsck program of libosl to fix your
1063 database. It might be necessary to use --force (even if your name
1064 isn't Luke). However, make sure para_server isn't running before
1065 executing oslfsck --force.
1066
1067 If you don't mind to recreate your database you can start
1068 from scratch by removing the entire database directory, i.e.
1069
1070 rm -rf ~/.paraslash/afs_database-0.4
1071
1072 Be aware that this removes all attribute definitions, all playlists
1073 and all mood definitions and requires to re-initialize the tables.
1074
1075 Although oslfsck fixes inconsistencies in database tables it doesn't
1076 care about the table contents. To check for invalid table contents, use
1077
1078 para_client check
1079
1080 This prints out references to missing audio files as well as invalid
1081 playlists and mood definitions.
1082
1083 ---------------------------------------
1084 Audio formats and audio format handlers
1085 ---------------------------------------
1086
1087 Audio formats
1088 ~~~~~~~~~~~~~
1089
1090 The following audio formats are supported by paraslash:
1091
1092 *MP3*
1093
1094 Mp3, MPEG-1 Audio Layer 3, is a common audio format for audio storage,
1095 designed as part of its MPEG-1 standard. An MP3 file is made up of
1096 multiple MP3 frames, which consist of a header and a data block. The
1097 size of an MP3 frame depends on the bit rate and on the number
1098 of channels. For a typical CD-audio file (sample rate of 44.1 kHz
1099 stereo), encoded with a bit rate of 128 kbit, an MP3 frame is about
1100 400 bytes large.
1101
1102 *OGG/Vorbis*
1103
1104 OGG is a standardized audio container format, while Vorbis is an
1105 open source codec for lossy audio compression. Since Vorbis is most
1106 commonly made available via the OGG container format, it is often
1107 referred to as OGG/Vorbis. The OGG container format divides data into
1108 chunks called OGG pages. A typical OGG page is about 4KB large. The
1109 Vorbis codec creates variable-bitrate (VBR) data, where the bitrate
1110 may vary considerably.
1111
1112 *OGG/Speex*
1113
1114 Speex is an open-source speech codec that is based on CELP (Code
1115 Excited Linear Prediction) coding. It is designed for voice
1116 over IP applications, has modest complexity and a small memory
1117 footprint. Wideband and narrowband (telephone quality) speech are
1118 supported. As for Vorbis audio, Speex bit-streams are often stored
1119 in OGG files.
1120
1121 *AAC*
1122
1123 Advanced Audio Coding (AAC) is a standardized, lossy compression
1124 and encoding scheme for digital audio which is the default audio
1125 format for Apple's iPhone, iPod, iTunes. Usually MPEG-4 is used as
1126 the container format and audio files encoded with AAC have the .m4a
1127 extension. A typical AAC frame is about 700 bytes large.
1128
1129 *WMA*
1130
1131 Windows Media Audio (WMA) is an audio data compression technology
1132 developed by Microsoft. A WMA file is usually encapsulated in the
1133 Advanced Systems Format (ASF) container format, which also specifies
1134 how meta data about the file is to be encoded. The bit stream of WMA
1135 is composed of superframes, each containing one or more frames of
1136 2048 samples. For 16 bit stereo a WMA superframe is about 8K large.
1137
1138 *FLAC*
1139
1140 The Free Lossless Audio Codec (FLAC) compresses audio without quality
1141 loss. It gives better compression ratios than a general purpose
1142 compressor like zip or bzip2 because FLAC is designed specifically
1143 for audio. A FLAC-encoded file consits of frames of varying size, up
1144 to 16K. Each frame starts with a header that contains all information
1145 necessary to decode the frame.
1146
1147 Meta data
1148 ~~~~~~~~~
1149
1150 Unfortunately, each audio format has its own conventions how meta
1151 data is added as tags to the audio file.
1152
1153 For MP3 files, ID3, version 1 and 2 are widely used. ID3 version 1
1154 is rather simple but also very limited as it supports only artist,
1155 title, album, year and comment tags. Each of these can only be at most
1156 32 characters long. ID3, version 2 is much more flexible but requires
1157 a separate library being installed for paraslash to support it.
1158
1159 Ogg vorbis, ogg speex and flac files contain meta data as Vorbis
1160 comments, which are typically implemented as strings of the form
1161 "[TAG]=[VALUE]". Unlike ID3 version 1 tags, one may use whichever
1162 tags are appropriate for the content.
1163
1164 AAC files usually use the MPEG-4 container format for storing meta
1165 data while WMA files wrap meta data as special objects within the
1166 ASF container format.
1167
1168 paraslash only tracks the most common tags that are supported by
1169 all tag variants: artist, title, year, album, comment. When a file
1170 is added to the AFS database, the meta data of the file is extracted
1171 and stored in the audio file table.
1172
1173 Chunks and chunk tables
1174 ~~~~~~~~~~~~~~~~~~~~~~~
1175
1176 paraslash uses the word "chunk" as common term for the building blocks
1177 of an audio file. For MP3 files, a chunk is the same as an MP3 frame,
1178 while for OGG files a chunk is an OGG page, etc. Therefore the chunk
1179 size varies considerably between audio formats, from a few hundred
1180 bytes (MP3) up to 16K (FLAC).
1181
1182 The chunk table contains the offsets within the audio file that
1183 correspond to the chunk boundaries of the file. Like the meta data,
1184 the chunk table is computed and stored in the database whenever an
1185 audio file is added.
1186
1187 The paraslash senders (see below) always send complete chunks. The
1188 granularity for seeking is therefore determined by the chunk size.
1189
1190 Audio format handlers
1191 ~~~~~~~~~~~~~~~~~~~~~
1192
1193 For each audio format paraslash contains an audio format handler whose
1194 first task is to tell whether a given file is a valid audio file of
1195 this type. If so, the audio file handler extracts some technical data
1196 (duration, sampling rate, number of channels etc.), computes the
1197 chunk table and reads the meta data.
1198
1199 The audio format handler code is linked into para_server and executed
1200 via the _add_ command. The same code is also available as a stand-alone
1201 tool, para_afh, which can be used to print the technical data, the
1202 chunk table and the meta data of a file. Furthermore, one can use
1203 para_afh to cut an audio file, i.e. to select some of its chunks to
1204 produce a new file containing only these chunks.
1205
1206 ----------
1207 Networking
1208 ----------
1209
1210 Paraslash uses different network connections for control and data.
1211 para_client communicates with para_server over a dedicated TCP control
1212 connection. To transport audio data, separate data connections are
1213 used. For these data connections, a variety of transports (UDP, DCCP,
1214 HTTP) can be chosen.
1215
1216 The chapter starts with the REFERENCE(The paraslash control
1217 service, control service), followed by a section on the various
1218 REFERENCE(Streaming protocols, streaming protocols) in which the data
1219 connections are described. The way audio file headers are embedded into
1220 the stream is discussed REFERENCE(Streams with headers and headerless
1221 streams, briefly) before the REFERENCE(Networking examples, example
1222 section) which illustrates typical commands for real-life scenarios.
1223
1224 Both IPv4 and IPv6 are supported.
1225
1226 The paraslash control service
1227 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1228
1229 para_server is controlled at runtime via the paraslash control
1230 connection. This connection is used for server commands (play, stop,
1231 ...) as well as for afs commands (ls, select, ...).
1232
1233 The server listens on a TCP port and accepts connections from clients
1234 that connect the open port. Each connection causes the server to fork
1235 off a client process which inherits the connection and deals with that
1236 client only. In this classical accept/fork approach the server process
1237 is unaffected if the child dies or goes crazy for whatever reason. In
1238 fact, the child process can not change address space of server process.
1239
1240 The section on REFERENCE(Client-server authentication, client-server
1241 authentication) above described the early connection establishment
1242 from the crypto point of view. Here it is described what happens
1243 after the connection (including crypto setup) has been established.
1244 There are four processes involved during command dispatch as sketched
1245 in the following diagram.
1246
1247 <<
1248 <pre>
1249 server_host client_host
1250 ~~~~~~~~~~~ ~~~~~~~~~~~
1251
1252 +-----------+ connect +-----------+
1253 |para_server|<------------------------------ |para_client|
1254 +-----------+ +-----------+
1255 | ^
1256 | fork +---+ |
1257 +----------> |AFS| |
1258 | +---+ |
1259 | ^ |
1260 | | |
1261 | | connect (cookie) |
1262 | | |
1263 | | |
1264 | fork +-----+ inherited connection |
1265 +---------->|child|<--------------------------+
1266 +-----+
1267 </pre>
1268 >>
1269
1270 Note that the child process is not a child of the afs process,
1271 so communication of these two processes has to happen via local
1272 sockets. In order to avoid abuse of the local socket by unrelated
1273 processes, a magic cookie is created once at server startup time just
1274 before the server process forks off the AFS process. This cookie is
1275 known to the server, AFS and the child, but not to unrelated processes.
1276
1277 There are two different kinds of commands: First there are commands
1278 that cause the server to respond with some answer such as the list
1279 of all audio files. All but the addblob commands (addimg, addlyr,
1280 addpl, addmood) are of this kind. The addblob commands add contents
1281 to the database, so they need to transfer data the other way round,
1282 from the client to the server.
1283
1284 There is no knowledge about the server commands built into para_client,
1285 so it does not know about addblob commands. Instead, it inspects the
1286 first data package sent by the server for a magic string. If this
1287 string was found, it sends STDIN to the server, otherwise it dumps
1288 data from the server to STDOUT.
1289
1290 Streaming protocols
1291 ~~~~~~~~~~~~~~~~~~~
1292
1293 A network (audio) stream usually consists of one streaming source,
1294 the _sender_, and one or more _receivers_ which read data over the
1295 network from the streaming source.
1296
1297 Senders are thus part of para_server while receivers are part of
1298 para_audiod. Moreover, there is the stand-alone tool para_recv which
1299 can be used to manually download a stream, either from para_server
1300 or from a web-based audio streaming service.
1301
1302 The following three streaming protocols are supported by paraslash:
1303
1304 - HTTP. Recommended for public streams that can be played by
1305 any player like mpg123, xmms, itunes, winamp, etc. The HTTP
1306 sender is supported on all operating systems and all platforms.
1307
1308 - DCCP. Recommended for LAN streaming. DCCP is currently
1309 available only for Linux.
1310
1311 - UDP. Recommended for multicast LAN streaming.
1312
1313 See the Appendix on REFERENCE(Network protocols, network protocols)
1314 for brief descriptions of the various protocols relevant for network
1315 audio streaming with paraslash.
1316
1317 It is possible to activate more than one sender simultaneously.
1318 Senders can be controlled at run time and via config file and command
1319 line options.
1320
1321 Note that audio connections are _not_ encrypted. Transport or Internet
1322 layer encryption should be used if encrypted data connections are
1323 needed.
1324
1325 Since DCCP and TCP are both connection-oriented protocols, connection
1326 establishment/teardown and access control are very similar between
1327 these two streaming protocols. UDP is the most lightweight option,
1328 since in contrast to TCP/DCCP it is connectionless. It is also the
1329 only protocol supporting IP multicast.
1330
1331 The HTTP and the DCCP sender listen on a (TCP/DCCP) port waiting for
1332 clients to connect and establish a connection via some protocol-defined
1333 handshake mechanism. Both senders maintain two linked lists each:
1334 The list of all clients which are currently connected, and the list
1335 of access control entries which determines who is allowed to connect.
1336 IP-based access control may be configured through config file and
1337 command line options and via the "allow" and "deny" sender subcommands.
1338
1339 Upon receiving a GET request from the client, the HTTP sender sends
1340 back a status line and a message. The body of this message is the
1341 audio stream. This is common practice and is supported by many popular
1342 clients which can thus be used to play a stream offered by para_server.
1343 For DCCP things are a bit simpler: No messages are exchanged between
1344 the receiver and sender. The client simply connects and the sender
1345 starts to stream.
1346
1347 DCCP is an experimental protocol which offers a number of new features
1348 not available for TCP. Both ends can negotiate these features using
1349 a built-in negotiation mechanism. In contrast to TCP/HTTP, DCCP is
1350 datagram-based (no retransmissions) and thus should not be used over
1351 lossy media (e.g. WiFi networks). One useful feature offered by DCCP
1352 is access to a variety of different congestion-control mechanisms
1353 called CCIDs. Two different CCIDs are available per default on Linux:
1354
1355
1356 - _CCID 2_. A Congestion Control mechanism similar to that
1357 of TCP. The sender maintains a congestion window and halves
1358 this window in response to congestion.
1359
1360
1361 - _CCID-3_. Designed to be fair when competing for bandwidth.
1362 It has lower variation of throughput over time compared with
1363 TCP, which makes it suitable for streaming media.
1364
1365 Unlike the HTTP and DCCP senders, the UDP sender maintains only a
1366 single list, the _target list_. This list describes the set of clients
1367 to which the stream is sent. There is no list for access control and
1368 no "allow" and "deny" commands for the UDP sender. Instead, the "add"
1369 and "delete" commands can be used to modify the target list.
1370
1371 Since both UDP and DCCP offer an unreliable datagram-based transport,
1372 additional measures are necessary to guard against disruptions over
1373 networks that are lossy or which may be subject to interference (as
1374 is for instance the case with WiFi). Paraslash uses FEC (Forward
1375 Error Correction) to guard against packet losses and reordering. The
1376 stream is FEC-encoded before it is sent through the UDP socket and
1377 must be decoded accordingly on the receiver side.
1378
1379 The packet size and the amount of redundancy introduced by FEC can
1380 be configured via the FEC parameters which are dictated by server
1381 and may also be configured through the "sender" command. The FEC
1382 parameters are encoded in the header of each network packet, so no
1383 configuration is necessary on the receiver side. See the section on
1384 REFERENCE(Forward error correction, FEC) below.
1385
1386 Streams with headers and headerless streams
1387 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1388
1389 For OGG/Vorbis, OGG/Speex and wma streams, some of the information
1390 needed to decode the stream is only contained in the audio file
1391 header of the container format but not in each data chunk. Clients
1392 must be able to obtain this information in case streaming starts in
1393 the middle of the file or if para_audiod is started while para_server
1394 is already sending a stream.
1395
1396 This is accomplished in different ways, depending on the streaming
1397 protocol. For connection-oriented streams (HTTP, DCCP) the audio file
1398 header is sent prior to audio file data. This technique however does
1399 not work for the connectionless UDP transport. Hence the audio file
1400 header is periodically being embedded into the UDP audio data stream.
1401 By default, the header is resent after five seconds. The receiver has
1402 to wait until the next header arrives before it can start decoding
1403 the stream.
1404
1405 Examples
1406 ~~~~~~~~
1407
1408 The sender command of para_server allows to (de-)activate senders
1409 and to change the access permissions senders at runtime. The "si"
1410 (server info) command is used to list the streaming options of the
1411 currently running server as well as the various sender access lists.
1412
1413 -> Show client/target/access lists:
1414
1415 para_client si
1416
1417 -> Obtain general help for the sender command:
1418
1419 para_client help sender
1420
1421 -> Get help for a specific sender (contains further examples):
1422
1423 s=http # or dccp or udp
1424 para_client sender $s help
1425
1426 By default para_server activates both the HTTP and th DCCP sender on
1427 startup. This can be changed via command line options or para_server's
1428 config file.
1429
1430 -> List config file options for senders:
1431
1432 para_server -h
1433
1434 All senders share the "on" and "off" commands, so senders may be
1435 activated and deactivated independently of each other.
1436
1437 -> Switch off the http sender:
1438
1439 para_client sender http off
1440
1441 -> Receive a DCCP stream using CCID2 and write the output into a file:
1442
1443 host=foo.org; ccid=2; filename=bar
1444 para_recv --receiver "dccp --host $host --ccid $ccid" > $filename
1445
1446 Note the quotes around the arguments for the dccp receiver. Each
1447 receiver has its own set of command line options and its own command
1448 line parser, so arguments for the dccp receiver must be protected
1449 from being interpreted by para_recv.
1450
1451 -> Start UDP multicast, using the default multicast address:
1452
1453 para_client sender udp add 224.0.1.38
1454
1455 -> Receive FEC-encoded multicast stream and write the output into a file:
1456
1457 filename=foo
1458 para_recv -r udp > $filename
1459
1460 -> Add an UDP unicast for a client to the target list of the UDP sender:
1461
1462 t=client.foo.org
1463 para_client sender udp add $t
1464
1465 -> Receive this (FEC-encoded) unicast stream:
1466
1467 filename=foo
1468 para_recv -r 'udp -i 0.0.0.0' > $filename
1469
1470 -> Create a minimal config for para_audiod for HTTP streams:
1471
1472 c=$HOME/.paraslash/audiod.conf.min; s=server.foo.com
1473 echo receiver \".:http -i $s\" > $c
1474 para_audiod --config $c
1475
1476 -------
1477 Filters
1478 -------
1479
1480 A paraslash filter is a module which transforms an input stream into
1481 an output stream. Filters are included in the para_audiod executable
1482 and in the stand-alone tool para_filter which usually contains the
1483 same modules.
1484
1485 While para_filter reads its input stream from STDIN and writes
1486 the output to STDOUT, the filter modules of para_audiod are always
1487 connected to a receiver which produces the input stream and a writer
1488 which absorbs the output stream.
1489
1490 Some filters depend on a specific library being installed and are
1491 not compiled in if this library was not found at compile time. To
1492 see the list of supported filters, run para_filter and para_audiod
1493 with the --help option. The output looks similar to the following:
1494
1495 Available filters:
1496 compress wav amp fecdec wmadec prebuffer oggdec aacdec mp3dec
1497
1498 Out of these filter modules, a chain of filters can be constructed,
1499 much in the way Unix pipes can be chained, and analogous to the use
1500 of modules in gstreamer: The output of the first filter becomes the
1501 input of the second filter. There is no limitation on the number of
1502 filters and the same filter may occur more than once.
1503
1504 Like receivers, each filter has its own command line options which
1505 must be quoted to protect them from the command line options of
1506 the driving application (para_audiod or para_filter). Example:
1507
1508 para_filter -f 'mp3dec --ignore-crc' -f 'compress --damp 1'
1509
1510 For para_audiod, each audio format has its own set of filters. The
1511 name of the audio format for which the filter should be applied can
1512 be used as the prefix for the filter option. Example:
1513
1514 para_audiod -f 'mp3:prebuffer --duration 300'
1515
1516 The "mp3" prefix above is actually interpreted as a POSIX extended
1517 regular expression. Therefore
1518
1519 para_audiod -f '.:prebuffer --duration 300'
1520
1521 activates the prebuffer filter for all supported audio formats (because
1522 "." matches all audio formats) while
1523
1524 para_audiod -f 'wma|ogg:prebuffer --duration 300'
1525
1526 activates it only for wma and ogg streams.
1527
1528 Decoders
1529 ~~~~~~~~
1530
1531 For each supported audio format there is a corresponding filter
1532 which decodes audio data in this format to 16 bit PCM data which
1533 can be directly sent to the sound device or any other software that
1534 operates on undecoded PCM data (visualizers, equalizers etc.). Such
1535 filters are called _decoders_ in general, and xxxdec is the name of
1536 the paraslash decoder for the audio format xxx. For example, the mp3
1537 decoder filter is called mp3dec.
1538
1539 Note that the output of the decoder is about 10 times larger than
1540 its input. This means that filters that operate on the decoded audio
1541 stream have to deal with much more data than filters that transform
1542 the audio stream before it is fed to the decoder.
1543
1544 Paraslash relies on external libraries for most decoders, so these
1545 libraries must be installed for the decoder to be included in the
1546 para_filter and para_audiod executables. The oggdec filter depends
1547 on the libogg and libvorbis libraries for example.
1548
1549 Forward error correction
1550 ~~~~~~~~~~~~~~~~~~~~~~~~
1551
1552 As already mentioned REFERENCE(Streaming protocols, earlier),
1553 paraslash uses forward error correction (FEC) for the unreliable UDP
1554 and DCCP transports. FEC is a technique which was invented already
1555 in 1960 by Reed and Solomon and which is widely used for the parity
1556 calculations of storage devices (RAID arrays). It is based on the
1557 algebraic concept of finite fields, today called Galois fields, in
1558 honour of the mathematician Galois (1811-1832). The FEC implementation
1559 of paraslash is based on code by Luigi Rizzo.
1560
1561 Although the details require a sound knowledge of the underlying
1562 mathematics, the basic idea is not hard to understand: For positive
1563 integers k and n with k < n it is possible to compute for any k given
1564 data bytes d_1, ..., d_k the corresponding r := n -k parity bytes p_1,
1565 ..., p_r such that all data bytes can be reconstructed from *any*
1566 k bytes of the set
1567
1568 {d_1, ..., d_k, p_1, ..., p_r}.
1569
1570 FEC-encoding for unreliable network transports boils down to slicing
1571 the audio stream into groups of k suitably sized pieces called _slices_
1572 and computing the r corresponding parity slices. This step is performed
1573 in para_server which then sends both the data and the parity slices
1574 over the unreliable network connection. If the client was able
1575 to receive at least k of the n = k + r slices, it can reconstruct
1576 (FEC-decode) the original audio stream.
1577
1578 From these observations it is clear that there are three different
1579 FEC parameters: The slice size, the number of data slices k, and the
1580 total number of slices n. It is crucial to choose the slice size
1581 such that no fragmentation of network packets takes place because
1582 FEC only guards against losses and reordering but fails if slices are
1583 received partially.
1584
1585 FEC decoding in paralash is performed through the fecdec filter which
1586 usually is the first filter (there can be other filters before fecdec
1587 if these do not alter the audio stream).
1588
1589
1590 Volume adjustment (amp and compress)
1591 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1592
1593 The amp and the compress filter both adjust the volume of the audio
1594 stream. These filters operate on uncompressed audio samples. Hence
1595 they are usually placed directly after the decoding filter. Each
1596 sample is multiplied with a scaling factor (>= 1) which makes amp
1597 and compress quite expensive in terms of computing power.
1598
1599 *amp*
1600
1601 The amp filter amplifies the audio stream by a fixed scaling factor
1602 that must be known in advance. For para_audiod this factor is derived
1603 from the amplification field of the audio file's entry in the audio
1604 file table while para_filter uses the value given at the command line.
1605
1606 The optimal scaling factor F for an audio file is the largest real
1607 number F >= 1 such that after multiplication with F all samples still
1608 fit into the sample interval [-32768, 32767]. One can use para_filter
1609 in combination with the sox utility to compute F:
1610
1611 para_filter -f mp3dec -f wav < file.mp3 | sox -t wav - -e stat -v
1612
1613 The amplification value V which is stored in the audio file table,
1614 however, is an integer between 0 and 255 which is connected to F
1615 through the formula
1616
1617 V = (F - 1) * 64.
1618
1619 To store V in the audio file table, the command
1620
1621 para_client -- touch -a=V file.mp3
1622
1623 is used. The reader is encouraged to write a script that performs
1624 these computations :)
1625
1626 *compress*
1627
1628 Unlike the amplification filter, the compress filter adjusts the volume
1629 of the audio stream dynamically without prior knowledge about the peak
1630 value. It maintains the maximal volume of the last n samples of the
1631 audio stream and computes a suitable amplification factor based on that
1632 value and the various configuration options. It tries to chose this
1633 factor such that the adjusted volume meets the desired target level.
1634
1635 Note that it makes sense to combine amp and compress.
1636
1637 Misc filters (wav and prebuffer)
1638 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1639
1640 These filters are rather simple and do not modify the audio stream at
1641 all. The wav filter is only useful with para_filter and in connection
1642 with a decoder. It asks the decoder for the number of channels and the
1643 sample rate of the stream and adds a Microsoft wave header containing
1644 this information at the beginning. This allows to write wav files
1645 rather than raw PCM files (which do not contain any information about
1646 the number of channels and the sample rate).
1647
1648 The prebuffer filter simply delays the output until the given time has
1649 passed (starting from the time the first byte was available in its
1650 input queue) or until the given amount of data has accumulated. It
1651 is mainly useful for para_audiod if the standard parameters result
1652 in buffer underruns.
1653
1654 Both filters require almost no additional computing time, even when
1655 operating on uncompressed audio streams, since data buffers are simply
1656 "pushed down" rather than copied.
1657
1658 Examples
1659 ~~~~~~~~
1660
1661 -> Decode an mp3 file to wav format:
1662
1663 para_filter -f mp3dec -f wav < file.mp3 > file.wav
1664
1665 -> Amplify a raw audio file by a factor of 1.5:
1666
1667 para_filter -f amp --amp 32 < foo.raw > bar.raw
1668
1669 ------
1670 Output
1671 ------
1672
1673 Once an audio stream has been received and decoded to PCM format,
1674 it can be sent to a sound device for playback. This part is performed
1675 by paraslash _writers_ which are described in this chapter.
1676
1677 Writers
1678 ~~~~~~~
1679
1680 A paraslash writer acts as a data sink that consumes but does not
1681 produce audio data. Paraslash writers operate on the client side and
1682 are contained in para_audiod and in the stand-alone tool para_write.
1683
1684 The para_write program reads uncompressed audio data from STDIN. If
1685 this data starts with a wav header, sample rate, sample format and
1686 channel count are read from the header. Otherwise CD audio (44.1KHz
1687 16 bit little endian, stereo) is assumed but this can be overridden
1688 by command line options. para_audiod, on the other hand, obtains
1689 the sample rate and the number of channels from the decoder.
1690
1691 Like receivers and filters, each writer has an individual set of
1692 command line options, and for para_audiod writers can be configured
1693 per audio format separately. It is possible to activate more than
1694 one writer for the same stream simultaneously.
1695
1696 OS-dependent APIs
1697 ~~~~~~~~~~~~~~~~~
1698
1699 Unfortunately, the various flavours of Unix on which paraslash
1700 runs on have different APIs for opening a sound device and starting
1701 playback. Hence for each such API there is a paraslash writer that
1702 can play the audio stream via this API.
1703
1704 *ALSA*. The _Advanced Linux Sound Architecture_ is only available on
1705 Linux systems. Although there are several mid-layer APIs in use by
1706 the various Linux distributions (ESD, Jack, PulseAudio), paraslash
1707 currently supports only the low-level ALSA API which is not supposed
1708 to be change. ALSA is very feature-rich, in particular it supports
1709 software mixing via its DMIX plugin. ALSA is the default writer on
1710 Linux systems.
1711
1712 *OSS*. The _Open Sound System_ is the only API on *BSD Unixes and
1713 is also available on Linux systems, usually provided by ALSA as an
1714 emulation for backwards compatibility. This API is rather simple but
1715 also limited. For example only one application can open the device
1716 at any time. The OSS writer is activated by default on BSD Systems.
1717
1718 *OSX*. Mac OS X has yet another API called CoreAudio. The OSX writer
1719 for this API is only compiled in on such systems and is of course
1720 the default there.
1721
1722 *FILE*. The file writer allows to capture the audio stream and
1723 write the PCM data to a file on the file system rather than playing
1724 it through a sound device. It is supported on all platforms and is
1725 always compiled in.
1726
1727 *AO*. _Libao_ is a cross-platform audio library which supports a wide
1728 variety of platforms including PulseAudio (gnome), ESD (Enlightened
1729 Sound Daemon), AIX, Solaris and IRIX. The ao writer plays audio
1730 through an output plugin of libao.
1731
1732 Examples
1733 ~~~~~~~~
1734
1735 -> Use the OSS writer to play a wav file:
1736
1737 para_write --writer oss < file.wav
1738
1739 -> Enable ALSA software mixing for mp3 streams
1740
1741 para_audiod --writer 'mp3:alsa -d plug:swmix'
1742
1743
1744 ---
1745 Gui
1746 ---
1747
1748 para_gui executes an arbitrary command which is supposed to print
1749 status information to STDOUT. It then displays this information in
1750 a curses window. By default the command
1751
1752 para_audioc -- stat -p
1753
1754 is executed, but this can be customized via the --stat_cmd option. In
1755 particular it possible to use
1756
1757 para_client -- stat -p
1758
1759 to make para_gui work on systems on which para_audiod is not running.
1760
1761 Key bindings
1762 ~~~~~~~~~~~~
1763
1764 It is possible to bind keys to arbitrary commands via custom
1765 key-bindings. Besides the internal keys which can not be changed (help,
1766 quit, loglevel, version...), the following flavours of key-bindings
1767 are supported:
1768
1769 - external: Shutdown curses before launching the given command.
1770 Useful for starting other ncurses programs from within
1771 para_gui, e.g. aumix or dialog scripts. Or, use the mbox
1772 output format to write a mailbox containing one mail for each
1773 (admissible) file the audio file selector knows about. Then
1774 start mutt from within para_gui to browse your collection!
1775
1776 - display: Launch the command and display its stdout in
1777 para_gui's bottom window.
1778
1779 - para: Like display, but start "para_client <specified
1780 command>" instead of "<specified command>".
1781
1782 The general form of a key binding is
1783
1784 key_map k:m:c
1785
1786 which maps key k to command c using mode m. Mode may be x, d or p
1787 for external, display and paraslash commands, respectively.
1788
1789 Themes
1790 ~~~~~~
1791
1792 Currently there are only two themes for para_gui. It is easy, however,
1793 to add more themes. To create a new theme one has to define the
1794 position, color and geometry for for each status item that should be
1795 shown by this theme. See gui_theme.c for examples.
1796
1797 The "." and "," keys are used to switch between themes.
1798
1799 Examples
1800 ~~~~~~~~
1801
1802 -> Show server info:
1803
1804 key_map "i:p:si"
1805
1806 -> Jump to the middle of the current audio file by pressing F5:
1807
1808 key_map "<F5>:p:jmp 50"
1809
1810 -> vi-like bindings for jumping around:
1811
1812 key_map "l:p:ff 10"
1813 key_map "h:p:ff 10-"
1814 key_map "w:p:ff 60"
1815 key_map "b:p:ff 60-"
1816
1817 -> Print the current date and time:
1818
1819 key_map "D:d:date"
1820
1821 -> Call other curses programs:
1822
1823 key_map "U:x:aumix"
1824 key_map "!:x:/bin/bash"
1825 key_map "^E:x:/bin/sh -c 'vi ~/.paraslash/gui.conf'"
1826
1827 -----------
1828 Development
1829 -----------
1830
1831 Tools
1832 ~~~~~
1833
1834 In order to compile the sources from the git repository (rather than
1835 from tar balls) and for contributing non-trivial changes to the
1836 paraslash project, some additional tools should be installed on a
1837 developer machine.
1838
1839 http://git.or.cz/ (git). As described in more detail REFERENCE(Git
1840 branches, below), the git source code management tool is used for
1841 paraslash development. It is necessary for cloning the git repository
1842 and for getting updates.
1843
1844 ftp://ftp.gnu.org/pub/gnu/m4/ (m4). Some input files for gengetopt
1845 are generated from templates by the m4 macro processor.
1846
1847 ftp://ftp.gnu.org/pub/gnu/autoconf/ (autoconf) GNU autoconf creates
1848 the configure file which is shipped in the tarballs but has to be
1849 generated when compiling from git.
1850
1851 http://www.triptico.com/software/grutatxt.html (grutatxt). The
1852 HTML version of this manual and some of the paraslash web pages are
1853 generated by the grutatxt plain text to HTML converter. If changes
1854 are made to these text files the grutatxt package must be installed
1855 to regenerate the HTML files.
1856
1857 http://www.stack.nl/~dimitri/doxygen/ (doxygen). The documentation
1858 of paraslash's C sources uses the doxygen documentation system. The
1859 conventions for documenting the source code is described in the
1860 REFERENCE(Doxygen, Doxygen section).
1861
1862 ftp://ftp.gnu.org/pub/gnu/global (global). This is used to generate
1863 browsable HTML from the C sources. It is needed by doxygen.
1864
1865 Git branches
1866 ~~~~~~~~~~~~
1867
1868 Paraslash has been developed using the git source code management
1869 tool since 2006. Development is organized roughly in the same spirit
1870 as the git development itself, as described below.
1871
1872 The following text passage is based on "A note from the maintainer",
1873 written by Junio C Hamano, the maintainer of git.
1874
1875 There are four branches in the paraslash repository that track the
1876 source tree: "master", "maint", "next", and "pu".
1877
1878 The "master" branch is meant to contain what is well tested and
1879 ready to be used in a production setting. There could occasionally be
1880 minor breakages or brown paper bag bugs but they are not expected to
1881 be anything major, and more importantly quickly and easily fixable.
1882 Every now and then, a "feature release" is cut from the tip of this
1883 branch, named with three dotted decimal digits, like 0.4.2.
1884
1885 Whenever changes are about to be included that will eventually lead to
1886 a new major release (e.g. 0.5.0), a "maint" branch is forked off from
1887 "master" at that point. Obvious, safe and urgent fixes after the major
1888 release are applied to this branch and maintenance releases are cut
1889 from it. New features never go to this branch. This branch is also
1890 merged into "master" to propagate the fixes forward.
1891
1892 A trivial and safe enhancement goes directly on top of "master".
1893 New development does not usually happen on "master", however.
1894 Instead, a separate topic branch is forked from the tip of "master",
1895 and it first is tested in isolation; Usually there are a handful such
1896 topic branches that are running ahead of "master". The tip of these
1897 branches is not published in the public repository to keep the number
1898 of branches that downstream developers need to worry about low.
1899
1900 The quality of topic branches varies widely. Some of them start out as
1901 "good idea but obviously is broken in some areas" and then with some
1902 more work become "more or less done and can now be tested by wider
1903 audience". Luckily, most of them start out in the latter, better shape.
1904
1905 The "next" branch is to merge and test topic branches in the latter
1906 category. In general, this branch always contains the tip of "master".
1907 It might not be quite rock-solid production ready, but is expected to
1908 work more or less without major breakage. The maintainer usually uses
1909 the "next" version of paraslash for his own pleasure, so it cannot
1910 be _that_ broken. The "next" branch is where new and exciting things
1911 take place.
1912
1913 The two branches "master" and "maint" are never rewound, and "next"
1914 usually will not be either (this automatically means the topics that
1915 have been merged into "next" are usually not rebased, and you can find
1916 the tip of topic branches you are interested in from the output of
1917 "git log next"). You should be able to safely build on top of them.
1918
1919 However, at times "next" will be rebuilt from the tip of "master" to
1920 get rid of merge commits that will never be in "master". The commit
1921 that replaces "next" will usually have the identical tree, but it
1922 will have different ancestry from the tip of "master".
1923
1924 The "pu" (proposed updates) branch bundles the remainder of the
1925 topic branches. The "pu" branch, and topic branches that are only in
1926 "pu", are subject to rebasing in general. By the above definition
1927 of how "next" works, you can tell that this branch will contain quite
1928 experimental and obviously broken stuff.
1929
1930 When a topic that was in "pu" proves to be in testable shape, it
1931 graduates to "next". This is done with
1932
1933 git checkout next
1934 git merge that-topic-branch
1935
1936 Sometimes, an idea that looked promising turns out to be not so good
1937 and the topic can be dropped from "pu" in such a case.
1938
1939 A topic that is in "next" is expected to be polished to perfection
1940 before it is merged to "master". Similar to the above, this is
1941 done with
1942
1943 git checkout master
1944 git merge that-topic-branch
1945 git branch -d that-topic-branch
1946
1947 Note that being in "next" is not a guarantee to appear in the next
1948 release (being in "master" is such a guarantee, unless it is later
1949 found seriously broken and reverted), nor even in any future release.
1950
1951 Coding Style
1952 ~~~~~~~~~~~~
1953
1954 The preferred coding style for paraslash coincides more or less
1955 with the style of the Linux kernel. So rather than repeating what is
1956 written XREFERENCE(http://www.kernel.org/doc/Documentation/CodingStyle,
1957 there), here are the most important points.
1958
1959 - Burn the GNU coding standards.
1960 - Never use spaces for indentation.
1961 - Tabs are 8 characters, and thus indentations are also 8 characters.
1962 - Don't put multiple assignments on a single line.
1963 - Avoid tricky expressions.
1964 - Don't leave whitespace at the end of lines.
1965 - The limit on the length of lines is 80 columns.
1966 - Use K&R style for placing braces and spaces:
1967
1968 if (x is true) {
1969 we do y
1970 }
1971
1972 - Use a space after (most) keywords.
1973 - Do not add spaces around (inside) parenthesized expressions.
1974 - Use one space around (on each side of) most binary and ternary operators.
1975 - Do not use cute names like ThisVariableIsATemporaryCounter, call it tmp.
1976 - Mixed-case names are frowned upon.
1977 - Descriptive names for global variables are a must.
1978 - Avoid typedefs.
1979 - Functions should be short and sweet, and do just one thing.
1980 - The number of local variables shouldn't exceed 10.
1981 - Gotos are fine if they improve readability and reduce nesting.
1982 - Don't use C99-style "// ..." comments.
1983 - Names of macros defining constants and labels in enums are capitalized.
1984 - Enums are preferred when defining several related constants.
1985 - Always use the paraslash wrappers for allocating memory.
1986 - If the name of a function is an action or an imperative.
1987 command, the function should return an error-code integer
1988 (<0 means error, >=0 means success). If the name is a
1989 predicate, the function should return a "succeeded" boolean.
1990
1991
1992 Doxygen
1993 ~~~~~~~
1994
1995 Doxygen is a documentation system for various programming
1996 languages. The paraslash project uses Doxygen for generating the API
1997 reference on the web pages, but good source code documentation is
1998 also beneficial to people trying to understand the code structure
1999 and the interactions between the various source files.
2000
2001 It is more illustrative to look at the source code for examples than
2002 to describe the conventions for documenting the source in this manual,
2003 so we only describe which parts of the code need doxygen comments,
2004 but leave out details on documentation conventions.
2005
2006 As a rule, only the public part of the C source is documented with
2007 Doxygen. This includes structures, defines and enumerations in header
2008 files as well as public (non-static) C functions. These should be
2009 documented completely. For example each parameter and the return
2010 value of a public function should get a descriptive comment.
2011
2012 No doxygen comments are necessary for static functions and for
2013 structures and enumerations in C files (which are used only within
2014 this file). This does not mean, however, that those entities need
2015 no documentation at all. Instead, common sense should be applied to
2016 document what is not obvious from reading the code.
2017
2018 --------
2019 Appendix
2020 --------
2021
2022 Network protocols
2023 ~~~~~~~~~~~~~~~~~
2024
2025 *IP*. The _Internet Protocol_ is the primary networking protocol
2026 used for the Internet. All protocols described below use IP as the
2027 underlying layer. Both the prevalent IPv4 and the next-generation
2028 IPv6 variant are being deployed actively worldwide.
2029
2030 *Connection-oriented and connectionless protocols*. Connectionless
2031 protocols differ from connection-oriented ones in that state
2032 associated with the sending/receiving endpoints is treated
2033 implicitly. Connectionless protocols maintain no internal knowledge
2034 about the state of the connection. Hence they are not capable of
2035 reacting to state changes, such as sudden loss or congestion on the
2036 connection medium. Connection-oriented protocols, in contrast, make
2037 this knowledge explicit. The connection is established only after
2038 a bidirectional handshake which requires both endpoints to agree
2039 on the state of the connection, and may also involve negotiating
2040 specific parameters for the particular connection. Maintaining an
2041 up-to-date internal state of the connection also in general means
2042 that the sending endpoints perform congestion control, adapting to
2043 qualitative changes of the connection medium.
2044
2045 *Reliability*. In IP networking, packets can be lost, duplicated,
2046 or delivered out of order, and different network protocols handle
2047 these problems in different ways. We call a transport-layer protocol
2048 _reliable_, if it turns the unreliable IP delivery into an ordered,
2049 duplicate- and loss-free delivery of packets. Sequence numbers
2050 are used to discard duplicates and re-arrange packets delivered
2051 out-of-order. Retransmission is used to guarantee loss-free
2052 delivery. Unreliable protocols, in contrast, do not guarantee ordering
2053 or data integrity.
2054
2055 *Classification*. With these definitions the protocols which are used
2056 by paraslash for steaming audio data may be classified as follows.
2057
2058 - HTTP/TCP: connection-oriented, reliable,
2059 - UDP: connectionless, unreliable,
2060 - DCCP: connection-oriented, unreliable.
2061
2062 Below we give a short descriptions of these protocols.
2063
2064 *TCP*. The _Transmission Control Protocol_ provides reliable,
2065 ordered delivery of a stream and a classic window-based congestion
2066 control. In contrast to UDP and DCCP (see below), TCP does not have
2067 record-oriented or datagram-based syntax, i.e. it provides a stream
2068 which is unaware and independent of any record (packet) boundaries.
2069 TCP is used extensively by many application layers. Besides HTTP (the
2070 Hypertext Transfer Protocol), also FTP (the File Transfer protocol),
2071 SMTP (Simple Mail Transfer Protocol), SSH (Secure Shell) all sit on
2072 top of TCP.
2073
2074 *UDP*. The _User Datagram Protocol_ is the simplest transport-layer
2075 protocol, built as a thin layer directly on top of IP. For this reason,
2076 it offers the same best-effort service as IP itself, i.e. there is no
2077 detection of duplicate or reordered packets. Being a connectionless
2078 protocol, only minimal internal state about the connection is
2079 maintained, which means that there is no protection against packet
2080 loss or network congestion. Error checking and correction (if at all)
2081 are performed in the application.
2082
2083 *DCCP*. The _Datagram Congestion Control Protocol_ combines the
2084 connection-oriented state maintenance known from TCP with the
2085 unreliable, datagram-based transport of UDP. This means that it
2086 is capable of reacting to changes in the connection by performing
2087 congestion control, offering multiple alternative approaches. But it
2088 is bound to datagram boundaries (the maximum packet size supported
2089 by a medium), and like UDP it lacks retransmission to protect
2090 against loss. Due to the use of sequence numbers, it is however
2091 able to react to loss (interpreted as a congestion indication) and
2092 to ignore out-of-order and duplicate packets. Unlike TCP it allows
2093 to negotiate specific, binding features for a connection, such as
2094 the choice of congestion control: classic, window-based congestion
2095 control known from TCP is available as CCID-2, rate-based, "smooth"
2096 congestion control is offered as CCID-3.
2097
2098 *HTTP*. _The Hypertext Transfer Protocol_ is an application layer
2099 protocol on top of TCP. It is spoken by web servers and is most often
2100 used for web services. However, as can be seen by the many Internet
2101 radio stations and YouTube/Flash videos, http is by far not limited to
2102 the delivery of web pages only. Being a simple request/response based
2103 protocol, the semantics of the protocol also allow the delivery of
2104 multimedia content, such as audio over http.
2105
2106 *Multicast*. IP multicast is not really a protocol but a technique
2107 for one-to-many communication over an IP network. The challenge is to
2108 deliver information to a group of destinations simultaneously using
2109 the most efficient strategy to send the messages over each link of
2110 the network only once. This has benefits for streaming multimedia:
2111 the standard one-to-one unicast offered by TCP/DCCP means that
2112 n clients listening to the same stream also consume n-times the
2113 resources, whereas multicast requires to send the stream just once,
2114 irrespective of the number of receivers. Since it would be costly to
2115 maintain state for each listening receiver, multicast often implies
2116 connectionless transport, which is the reason that it is currently
2117 only available via UDP.
2118
2119 License
2120 ~~~~~~~
2121
2122 Paraslash is licensed under the GPL, version 2. Most of the code
2123 base has been written from scratch, and those parts are GPL V2
2124 throughout. Notable exceptions are FEC and the WMA decoder. See the
2125 corresponding source files for licencing details for these parts. Some
2126 code sniplets of several other third party software packages have
2127 been incorporated into the paraslash sources, for example log message
2128 coloring was taken from the git sources. These third party software
2129 packages are all published under the GPL or some other license
2130 compatible to the GPL.
2131
2132 Acknowledgements
2133 ~~~~~~~~~~~~~~~~
2134
2135 Many thanks to Gerrit Renker who read an early draft of this manual
2136 and contributed significant improvements.
2137
2138 ----------
2139 References
2140 ----------
2141
2142 Articles
2143 ~~~~~~~~
2144 - Reed, Irving S.; Solomon, Gustave (1960),
2145 XREFERENCE(http://kom.aau.dk/~heb/kurser/NOTER/KOFA01.PDF,
2146 Polynomial Codes over Certain Finite Fields), Journal of the
2147 Society for Industrial and Applied Mathematics (SIAM) 8 (2):
2148 300-304, doi:10.1137/0108018)
2149
2150 RFCs
2151 ~~~~
2152
2153 - XREFERENCE(http://www.ietf.org/rfc/rfc768.txt, RFC 768) (1980):
2154 User Datagram Protocol
2155 - XREFERENCE(http://www.ietf.org/rfc/rfc791.txt, RFC 791) (1981):
2156 Internet Protocol
2157 - XREFERENCE(http://www.ietf.org/rfc/rfc2437.txt, RFC 2437) (1998):
2158 RSA Cryptography Specifications
2159 - XREFERENCE(http://www.ietf.org/rfc/rfc4340.txt, RFC 4340)
2160 (2006): Datagram Congestion Control Protocol (DCCP)
2161 - XREFERENCE(http://www.ietf.org/rfc/rfc4341.txt, RFC 4341) (2006):
2162 Congestion Control ID 2: TCP-like Congestion Control
2163 - XREFERENCE(http://www.ietf.org/rfc/rfc4342.txt, RFC 4342) (2006):
2164 Congestion Control ID 3: TCP-Friendly Rate Control (TFRC)
2165
2166 Application web pages
2167 ~~~~~~~~~~~~~~~~~~~~~
2168
2169 - XREFERENCE(http://paraslash.systemlinux.org/, paraslash)
2170 - XREFERENCE(http://xmms2.org/wiki/Main_Page, xmms)
2171 - XREFERENCE(http://www.mpg123.de/, mpg123)
2172 - XREFERENCE(http://gstreamer.freedesktop.org/, gstreamer)
2173 - XREFERENCE(http://www.icecast.org/, icecast)
2174 - XREFERENCE(http://beesbuzz.biz/code/audiocompress.php, Audio Compress)
2175
2176 External documentation
2177 ~~~~~~~~~~~~~~~~~~~~~~
2178
2179 - XREFERENCE(http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf,
2180 H. Peter Anvin: The mathematics of Raid6)
2181 - XREFERENCE(http://info.iet.unipi.it/~luigi/fec_ccr.ps.gz,
2182 Luigi Rizzo: Effective Erasure Codes for reliable Computer
2183 Communication Protocols)
2184
2185 Code
2186 ~~~~
2187 - XREFERENCE(http://info.iet.unipi.it/~luigi/vdm.tar.gz,
2188 Original FEC implementation by Luigi Rizzo)
2189