22abf76f85364e8d899029cad661f03fe8fefe17
[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 *para_play*
178
179 A command line audio player.
180
181 *para_gui*
182
183 Curses-based gui that presents status information obtained in a curses
184 window. Appearance can be customized via themes. para_gui provides
185 key-bindings for the most common server commands and new key-bindings
186 can be added easily.
187
188
189 *para_fade*
190
191 An alarm clock and volume-fader for OSS and ALSA.
192
193 -----------
194 Quick start
195 -----------
196
197 This chapter lists the REFERENCE(Requirements, necessary software)
198 that must be installed to compile the paraslash package, describes
199 how to REFERENCE(Installation, compile and install) the paraslash
200 source code and the steps that have to be performed in order to
201 REFERENCE(Quick start, set up) a typical server and client.
202
203 Requirements
204 ~~~~~~~~~~~~
205
206 In any case you'll need
207
208 - XREFERENCE(http://systemlinux.org/~maan/osl/, libosl).
209 The _object storage layer_ library is used by para_server. To
210 clone the source code repository, execute
211
212 git clone git://git.tuebingen.mpg.de/osl
213
214 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gcc, gcc) or
215 XREFERENCE(http://clang.llvm.org, clang). All gcc versions
216 >= 3.3 are currently supported. Clang version 1.1 or newer
217 should work as well.
218
219 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/make, gnu make) is
220 also shipped with the disto. On BSD systems the gnu make
221 executable is often called gmake.
222
223 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/bash, bash). Some
224 scripts which run during compilation require the EMPH(Bourne
225 again shell). It is most likely already installed.
226
227 - XREFERENCE(http://www.openssl.org/, openssl) or
228 XREFERENCE(ftp://ftp.gnupg.org/gcrypt/libgcrypt/, libgcrypt).
229 At least one of these two libraries is needed as the backend
230 for cryptographic routines on both the server and the client
231 side. Both openssl and libgcrypt are usually shipped with the
232 distro, but you might have to install the development package
233 (libssl-dev or libgcrypt-dev on debian systems) as well.
234
235 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gengetopt/, gengetopt)
236 is needed to generate the C code for the command line parsers
237 of all paraslash executables.
238
239 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/help2man, help2man)
240 is used to create the man pages.
241
242 Optional:
243
244 - XREFERENCE(http://www.underbit.com/products/mad/, libmad).
245 To compile in MP3 support for paraslash, the development
246 package must be installed. It is called libmad0-dev on
247 debian-based systems. Note that libmad is not necessary on
248 the server side, i.e. for sending MP3 files.
249
250 - XREFERENCE(http://www.underbit.com/products/mad/,
251 libid3tag). For version-2 ID3 tag support, you'll need
252 the libid3tag development package libid3tag0-dev. Without
253 libid3tag, only version one tags are recognized.
254
255 - XREFERENCE(http://www.xiph.org/downloads/, ogg vorbis).
256 For ogg vorbis streams you'll need libogg, libvorbis,
257 libvorbisfile. The corresponding Debian packages are called
258 libogg-dev and libvorbis-dev.
259
260 - XREFERENCE(http://www.audiocoding.com/, libfaad). For aac
261 files (m4a) you'll need libfaad (libfaad-dev).
262
263 - XREFERENCE(http://www.speex.org/, speex). In order to stream
264 or decode speex files, libspeex (libspeex-dev) is required.
265
266 - XREFERENCE(http://flac.sourceforge.net/, flac). To stream
267 or decode files encoded with the _Free Lossless Audio Codec_,
268 libFLAC (libFLAC-dev) must be installed.
269
270 - XREFERENCE(http://www.mega-nerd.com/SRC/index.html,
271 libsamplerate). The resample filter will only be compiled if
272 this library is installed. Debian package: libsamplerate-dev.
273
274 - XREFERENCE(ftp://ftp.alsa-project.org/pub/lib/, alsa-lib). On
275 Linux, you'll need to have ALSA's development package
276 libasound2-dev installed.
277
278 - XREFERENCE(http://downloads.xiph.org/releases/ao/,
279 libao). Needed to build the ao writer (ESD, PulseAudio,...).
280 Debian package: libao-dev.
281
282 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/ncurses, curses). Needed
283 for para_gui. Debian package: libncurses-dev.
284
285 - XREFERENCE(http://cnswww.cns.cwru.edu/php/chet/readline/rltop.html,
286 GNU Readline). If this library (libreadline-dev) is installed,
287 para_client, para_audioc and para_play support interactive
288 sessions.
289
290 Installation
291 ~~~~~~~~~~~~
292
293 First make sure all non-optional packages listed in the section on
294 REFERENCE(Requirements, required software) are installed on your
295 system.
296
297 You don't need everything listed there. In particular, MP3, OGG/Vorbis,
298 OGG/Speex and AAC support are all optional. The configure script will
299 detect what is installed on your system and will only try to build
300 those executables that can be built with your setup.
301
302 Note that no special decoder library (not even the MP3 decoding library
303 libmad) is needed for para_server if you only want to stream MP3 or WMA
304 files. Also, it's fine to use para_server on a box without sound card.
305
306 Next, install the paraslash package on all machines, you'd like this
307 software to run on. If you compile from a released tarball, execute
308
309 (./configure && make) > /dev/null
310
311 When compiling from git or from snapshots downloaded via gitweb,
312 the above command will not work because the configure script is not
313 included in the git repository. In this case the following command
314 should be used instead:
315
316 ./autogen.sh
317
318 This runs autoconf to generate the configure script, then runs it as
319 above. Therefore you'll need autoconf for this to work.
320
321 There should be no errors but probably some warnings about missing
322 packages which usually implies that not all audio formats will be
323 supported. If headers or libs are installed at unusual locations you
324 might need to tell the configure script where to find them. Try
325
326 ./configure --help
327
328 to see a list of options. If the paraslash package was compiled
329 successfully, execute (optionally)
330
331 make test
332
333 to run the paraslash test suite. If all tests pass, execute as root
334
335 make install
336
337 to install executables under /usr/local/bin and the man pages under
338 /usr/local/man.
339
340 Configuration
341 ~~~~~~~~~~~~~
342
343 *Step 1*: Create a paraslash user
344
345 In order to control para_server at runtime you must create a paraslash
346 user. As authentication is based on the RSA crypto system you'll have
347 to create an RSA key pair. If you already have a user and an RSA key
348 pair, you may skip this step.
349
350 In this section we'll assume a typical setup: You would like to run
351 para_server on some host called server_host as user foo, and you want
352 to connect to para_server from another machine called client_host as
353 user bar.
354
355 As foo@server_host, create ~/.paraslash/server.users by typing the
356 following commands:
357
358 user=bar
359 target=~/.paraslash/server.users
360 key=~/.paraslash/id_rsa.pub.$user
361 perms=AFS_READ,AFS_WRITE,VSS_READ,VSS_WRITE
362 mkdir -p ~/.paraslash
363 echo "user $user $key $perms" >> $target
364
365 Next, change to the "bar" account on client_host and generate the
366 key pair with the commands
367
368 ssh-keygen -t rsa -b 2048
369 # hit enter twice to create a key with no passphrase
370
371 This generates the two files id_rsa and id_rsa.pub in ~/.ssh. Note
372 that paraslash can also read keys generated by the "openssl genrsa"
373 command. However, since keys created with ssh-keygen can also be used
374 for ssh, this method is recommended.
375
376 Note that para_server refuses to use a key if it is shorter than 2048
377 bits. In particular, the RSA keys of paraslash 0.3.x will not work
378 with version 0.4.x. Moreover, para_client refuses to use a (private)
379 key which is world-readable.
380
381 para_server only needs to know the public key of the key pair just
382 created. Copy this public key to server_host:
383
384 src=~/.ssh/id_rsa.pub
385 dest=.paraslash/id_rsa.pub.$LOGNAME
386 scp $src foo@server_host:$dest
387
388 Finally, tell para_client to connect to server_host:
389
390 conf=~/.paraslash/client.conf
391 echo 'hostname server_host' > $conf
392
393
394 *Step 2*: Start para_server
395
396 Before starting the server make sure you have write permissions to
397 the directory /var/paraslash that has been created during installation:
398
399 sudo chown $LOGNAME /var/paraslash
400
401 Alternatively, use the --afs_socket Option to specify a different
402 location for the AFS command socket.
403
404 For this first try, we'll use the info loglevel to make the output
405 of para_server more verbose.
406
407 para_server -l info
408
409 Now you can use para_client to connect to the server and issue
410 commands. Open a new shell as bar@client_host and try
411
412 para_client help
413 para_client si
414
415 to retrieve the list of available commands and some server info.
416 Don't proceed if this doesn't work.
417
418 *Step 3*: Create and populate the database
419
420 An empty database is created with
421
422 para_client init
423
424 This initializes a couple of empty tables under
425 ~/.paraslash/afs_database-0.4. You normally don't need to look at these
426 tables, but it's good to know that you can start from scratch with
427
428 rm -rf ~/.paraslash/afs_database-0.4
429
430 in case something went wrong.
431
432 Next, you need to add some audio files to that database so that
433 para_server knows about them. Choose an absolute path to a directory
434 containing some audio files and add them to the audio file table:
435
436 para_client add /my/mp3/dir
437
438 This might take a while, so it is a good idea to start with a directory
439 containing not too many files. Note that the table only contains data
440 about the audio files found, not the files themselves.
441
442 You may print the list of all known audio files with
443
444 para_client ls
445
446 *Step 4*: Configure para_audiod
447
448 para_audiod needs to create a "well-known" socket for the clients to
449 connect to. The default path for this socket is
450
451 /var/paraslash/audiod_socket.$HOSTNAME
452
453 In order to make this directory writable for para_audiod, execute
454 as bar@client_host
455
456 sudo chown $LOGNAME /var/paraslash
457
458
459 We will also have to tell para_audiod that it should receive the
460 audio stream from server_host via http:
461
462 para_audiod -l info -r '.:http -i server_host'
463
464 You should now be able to listen to the audio stream once para_server
465 starts streaming. To activate streaming, execute
466
467 para_client play
468
469 Since no playlist has been specified yet, the "dummy" mode which
470 selects all known audio files is activated automatically. See the
471 section on the REFERENCE(The audio file selector, audio file selector)
472 for how to use playlists and moods to specify which files should be
473 streamed in which order.
474
475 *Troubleshooting*
476
477 It did not work? To find out why, try to receive, decode and play the
478 stream manually using para_recv, para_filter and para_write as follows.
479
480 For simplicity we assume that you're running Linux/ALSA and that only
481 MP3 files have been added to the database.
482
483 para_recv -r 'http -i server_host' > file.mp3
484 # (interrupt with CTRL+C after a few seconds)
485 ls -l file.mp3 # should not be empty
486 para_filter -f mp3dec -f wav < file.mp3 > file.wav
487 ls -l file.wav # should be much bigger than file.mp3
488 para_write -w alsa < file.wav
489
490 Double check what is logged by para_server and use the --loglevel
491 option of para_recv, para_filter and para_write to increase verbosity.
492
493 ---------------
494 User management
495 ---------------
496
497 para_server uses a challenge-response mechanism to authenticate
498 requests from incoming connections, similar to ssh's public key
499 authentication method. Authenticated connections are encrypted using
500 the RC4 stream cipher.
501
502 In this chapter we briefly describe RSA and RC4 and sketch the
503 REFERENCE(Client-server authentication, authentication handshake)
504 between para_client and para_server. User management is discussed
505 in the section on REFERENCE(The user_list file, the user_list file).
506 These sections are all about communication between the client and the
507 server. Connecting para_audiod is a different matter and is described
508 in a REFERENCE(Connecting para_audiod, separate section).
509
510
511
512 RSA and RC4
513 ~~~~~~~~~~~
514
515 RSA is an asymmetric block cipher which is used in many applications,
516 including ssh and gpg. An RSA key consists in fact of two keys,
517 called the public key and the private key. A message can be encrypted
518 with either key and only the counterpart of that key can decrypt
519 the message. While RSA can be used for both signing and encrypting
520 a message, paraslash uses RSA only for the latter purpose. The
521 RSA public key encryption and signatures algorithms are defined in
522 detail in RFC 2437.
523
524 RC4 is a stream cipher, i.e. the input is XORed with a pseudo-random
525 key stream to produce the output. Decryption uses the same function
526 calls as encryption. While RC4 supports variable key lengths,
527 paraslash uses a fixed length of 256 bits, which is considered a
528 strong encryption by today's standards. Since the same key must never
529 be used twice, a different, randomly-generated key is used for every
530 new connection.
531
532 Client-server authentication
533 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
534
535 The authentication handshake between para_client and para_server goes
536 as follows:
537
538 - para_client connects to para_server and sends an
539 authentication request for a user. It does so by connecting
540 to TCP port 2990 of the server host. This port is called the
541 para_server _control port_.
542
543 - para_server accepts the connection and forks a child process
544 which handles the incoming request. The parent process keeps
545 listening on the control port while the child process (also
546 called para_server below) continues as follows.
547
548 - para_server loads the RSA public key of that user, fills a
549 fixed-length buffer with random bytes, encrypts that buffer
550 using the public key and sends the encrypted buffer to the
551 client. The first part of the buffer is the challenge which
552 is used for authentication while the second part is the RC4
553 session key.
554
555 - para_client receives the encrypted buffer and decrypts it
556 with the user's private key, thereby obtaining the challenge
557 buffer and the session key. It sends the SHA1 hash value of
558 the challenge back to para_server and stores the session key
559 for further use.
560
561 - para_server also computes the SHA1 hash of the challenge
562 and compares it against what was sent back by the client.
563
564 - If the two hashes do not match, the authentication has
565 failed and para_server closes the connection.
566
567 - Otherwise the user is considered authenticated and the client
568 is allowed to proceed by sending a command to be executed. From
569 this point on the communication is encrypted using the RC4
570 stream cipher with the session key known to both peers.
571
572 paraslash relies on the quality of the pseudo-random bytes provided
573 by the crypto library (openssl or libgcrypt), on the security of
574 the implementation of the RSA and RC4 crypto routines and on the
575 infeasibility to invert the SHA1 function.
576
577 Neither para_server or para_client create RSA keys on their own. This
578 has to be done once for each user as sketched in REFERENCE(Quick start,
579 Quick start) and discussed in more detail REFERENCE(The user_list
580 file, below).
581
582 The user_list file
583 ~~~~~~~~~~~~~~~~~~
584
585 At startup para_server reads the user list file which contains one
586 line per user. The default location of the user list file may be
587 changed with the --user_list option.
588
589 There should be at least one user in this file. Each user must have
590 an RSA key pair. The public part of the key is needed by para_server
591 while the private key is needed by para_client. Each line of the
592 user list file must be of the form
593
594 user <username> <key> <perms>
595
596 where _username_ is an arbitrary string (usually the user's login
597 name), _key_ is the full path to that user's public RSA key, and
598 _perms_ is a comma-separated list of zero or more of the following
599 permission bits:
600
601 +---------------------------------------------------------+
602 | AFS_READ | read the contents of the databases |
603 +-----------+---------------------------------------------+
604 | AFS_WRITE | change database contents |
605 +-----------+---------------------------------------------+
606 | VSS_READ | obtain information about the current stream |
607 +-----------+---------------------------------------------+
608 | VSS_WRITE | change the current stream |
609 +---------------------------------------------------------+
610
611 The permission bits specify which commands the user is allowed to
612 execute. The output of
613
614 para_client help
615
616 contains in the third column the permissions needed to execute the
617 command.
618
619 It is possible to make para_server reread the user_list file by
620 executing the paraslash "hup" command or by sending SIGHUP to the
621 PID of para_server.
622
623
624 Connecting para_audiod
625 ~~~~~~~~~~~~~~~~~~~~~~
626
627 para_audiod listens on a Unix domain socket. Those sockets are
628 for local communication only, so only local users can connect to
629 para_audiod. The default is to let any user connect but this can be
630 restricted on platforms that support UNIX socket credentials which
631 allow para_audiod to obtain the Unix credentials of the connecting
632 process.
633
634 Use para_audiod's --user_allow option to allow connections only for
635 a limited set of users.
636
637 -----------------------
638 The audio file selector
639 -----------------------
640
641 paraslash comes with a sophisticated audio file selector (AFS),
642 whose main task is to determine which file to stream next, based on
643 information on the audio files stored in a database. It communicates
644 also with para_client whenever an AFS command is executed, for example
645 to answer a database query.
646
647 Besides the traditional playlists, AFS supports audio file selection
648 based on _moods_ which act as a filter that limits the set of all
649 known audio files to those which satisfy certain criteria. It also
650 maintains tables containing images (e.g. album cover art) and lyrics
651 that can be associated with one or more audio files.
652
653 AFS uses XREFERENCE(http://systemlinux.org/~maan/osl/, libosl), the
654 object storage layer library, as the backend library for storing
655 information on audio files, playlists, etc. This library offers
656 functionality similar to a relational database, but is much more
657 lightweight than a full database backend.
658
659 In this chapter we sketch the setup of the REFERENCE(The AFS process,
660 AFS process) during server startup and proceed with the description
661 of the REFERENCE(Database layout, layout) of the various database
662 tables. The section on REFERENCE(Playlists and moods, playlists
663 and moods) explains these two audio file selection mechanisms
664 in detail and contains pratical examples. The way REFERENCE(File
665 renames and content changes, file renames and content changes) are
666 detected is discussed briefly before the REFERENCE(Troubleshooting,
667 Troubleshooting) section concludes the chapter.
668
669 The AFS process
670 ~~~~~~~~~~~~~~~
671
672 On startup, para_server forks to create the AFS process which opens
673 the OSL database tables. The server process communicates with the
674 AFS process via pipes and shared memory. Usually, the AFS process
675 awakes only briefly whenever the current audio file changes. The AFS
676 process determines the next audio file, opens it, verifies it has
677 not been changed since it was added to the database and passes the
678 open file descriptor to the server process, along with audio file
679 meta-data such as file name, duration, audio format and so on. The
680 server process then starts to stream the audio file.
681
682 The AFS process also accepts connections from local clients via
683 a well-known socket. However, only child processes of para_server
684 may connect through this socket. All server commands that have the
685 AFS_READ or AFS_WRITE permission bits use this mechanism to query or
686 change the database.
687
688 Database layout
689 ~~~~~~~~~~~~~~~
690
691 *The audio file table*
692
693 This is the most important and usually also the largest table of the
694 AFS database. It contains the information needed to stream each audio
695 file. In particular the following data is stored for each audio file.
696
697 - SHA1 hash value of the audio file contents. This is computed
698 once when the file is added to the database. Whenever AFS
699 selects this audio file for streaming the hash value is
700 recomputed and checked against the value stored in the
701 database to detect content changes.
702
703 - The time when this audio file was last played.
704
705 - The number of times the file has been played so far.
706
707 - The attribute bitmask.
708
709 - The image id which describes the image associated with this
710 audio file.
711
712 - The lyrics id which describes the lyrics associated with
713 this audio file.
714
715 - The audio format id (MP3, OGG, ...).
716
717 - An amplification value that can be used by the amplification
718 filter to pre-amplify the decoded audio stream.
719
720 - The chunk table. It describes the location and the timing
721 of the building blocks of the audio file. This is used by
722 para_server to send chunks of the file at appropriate times.
723
724 - The duration of the audio file.
725
726 - Tag information contained in the audio file (ID3 tags,
727 Vorbis comments, ...).
728
729 - The number of channels
730
731 - The encoding bitrate.
732
733 - The sampling frequency.
734
735 To add or refresh the data contained in the audio file table, the _add_
736 command is used. It takes the full path of either an audio file or a
737 directory. In the latter case, the directory is traversed recursively
738 and all files which are recognized as valid audio files are added to
739 the database.
740
741 *The attribute table*
742
743 The attribute table contains two columns, _name_ and _bitnum_. An
744 attribute is simply a name for a certain bit number in the attribute
745 bitmask of the audio file table.
746
747 Each of the 64 bits of the attribute bitmask can be set for each
748 audio file individually. Hence up to 64 different attributes may be
749 defined. For example, "pop", "rock", "blues", "jazz", "instrumental",
750 "german_lyrics", "speech", whatever. You are free to choose as
751 many attributes as you like and there are no naming restrictions
752 for attributes.
753
754 A new attribute "test" is created by
755
756 para_client addatt test
757 and
758 para_client lsatt
759
760 lists all available attributes. You can set the "test" attribute for
761 an audio file by executing
762
763 para_client setatt test+ /path/to/the/audio/file
764
765 Similarly, the "test" bit can be removed from an audio file with
766
767 para_client setatt test- /path/to/the/audio/file
768
769 Instead of a path you may use a shell wildcard pattern. The attribute
770 is applied to all audio files matching this pattern:
771
772 para_client setatt test+ '/test/directory/*'
773
774 The command
775
776 para_client -- ls -lv
777
778 gives you a verbose listing of your audio files also showing which
779 attributes are set.
780
781 In case you wonder why the double-dash in the above command is needed:
782 It tells para_client to not interpret the options after the dashes. If
783 you find this annoying, just say
784
785 alias para='para_client --'
786
787 and be happy. In what follows we shall use this alias.
788
789 The "test" attribute can be dropped from the database with
790
791 para rmatt test
792
793 Read the output of
794
795 para help ls
796 para help setatt
797
798 for more information and a complete list of command line options to
799 these commands.
800
801 *Blob tables*
802
803 The image, lyrics, moods and playlists tables are all blob tables.
804 Blob tables consist of three columns each: The identifier which is
805 a positive non-negative number that is auto-incremented, the name
806 (an arbitrary string) and the content (the blob).
807
808 All blob tables support the same set of actions: cat, ls, mv, rm
809 and add. Of course, _add_ is used for adding new blobs to the table
810 while the other actions have the same meaning as the corresponding
811 Unix commands. The paraslash commands to perform these actions are
812 constructed as the concatenation of the table name and the action. For
813 example addimg, catimg, lsimg, mvimg, rmimg are the commands that
814 manipulate or query the image table.
815
816 The add variant of these commands is special as these commands read
817 the blob contents from stdin. To add an image to the image table the
818 command
819
820 para addimg image_name < file.jpg
821
822 can be used.
823
824 Note that the images and lyrics are not interpreted at all, and also
825 the playlist and the mood blobs are only investigated when the mood
826 or playlist is activated with the select command.
827
828 *The score table*
829
830 Unlike all other tables the contents of the score table remain in
831 memory and are never stored on disk. The score table contains two
832 columns: The SHA1 hash value (of an audio file) and its current
833 score.
834
835 However, only those files which are admissible for the current mood
836 or playlist are contained in the score table. The audio file selector
837 always chooses the row with the highest score as the file to stream
838 next. While doing so, it computes the new score and updates the
839 last_played and the num_played fields in the audio file table.
840
841 The score table is recomputed by the select command which loads a
842 mood or playlist. Audio files are chosen for streaming from the rows
843 of the score table on a highest-score-first basis.
844
845
846 Playlists and moods
847 ~~~~~~~~~~~~~~~~~~~
848
849 Playlists and moods offer two different ways of specifying the set of
850 admissible files. A playlist in itself describes a set of admissible
851 files. A mood, in contrast, describes the set of admissible files in
852 terms of attributes and other type of information available in the
853 audio file table. As an example, a mood can define a filename pattern,
854 which is then matched against the names of audio files in the table.
855
856 *Playlists*
857
858 Playlists are accommodated in the playlist table of the afs database,
859 using the aforementioned blob format for tables. A new playlist is
860 created with the addpl command by specifying the full (absolute)
861 paths of all desired audio files, separated by newlines. Example:
862
863 find /my/mp3/dir -name "*.mp3" | para addpl my_playlist
864
865 If _my_playlist_ already exists it is overwritten. To activate the
866 new playlist, execute
867
868 para select p/my_playlist
869
870 The audio file selector will assign scores to each entry of the list,
871 in descending order so that files will be selected in order. If a
872 file could not be opened for streaming, its entry is removed from
873 the score table (but not from the playlist).
874
875 *Moods*
876
877 A mood consists of a unique name and its *mood definition*, which is
878 a set of *mood lines* containing expressions in terms of attributes
879 and other data contained in the database.
880
881 At any time at most one mood can be *active* which means that
882 para_server is going to select only files from that subset of
883 admissible files.
884
885 So in order to create a mood definition one has to write a set of
886 mood lines. Mood lines come in three flavours: Accept lines, deny
887 lines and score lines.
888
889 The general syntax of the three types of mood lines is
890
891
892 accept [with score <score>] [if] [not] <mood_method> [options]
893 deny [with score <score>] [if] [not] <mood_method> [options]
894 score <score> [if] [not] <mood_method> [options]
895
896
897 Here <score> is either an integer or the string "random" which assigns
898 a random score to all matching files. The score value changes the
899 order in which admissible files are going to be selected, but is of
900 minor importance for this introduction.
901
902 So we concentrate on the first two forms, i.e. accept and deny
903 lines. As usual, everything in square brackets is optional, i.e.
904 accept/deny lines take the following form when ignoring scores:
905
906 accept [if] [not] <mood_method> [options]
907
908 and analogously for the deny case. The "if" keyword is only syntactic
909 sugar and has no function. The "not" keyword just inverts the result,
910 so the essence of a mood line is the mood method part and the options
911 following thereafter.
912
913 A *mood method* is realized as a function which takes an audio file
914 and computes a number from the data contained in the database.
915 If this number is non-negative, we say the file *matches* the mood
916 method. The file matches the full mood line if it either
917
918 - matches the mood method and the "not" keyword is not given,
919 or
920 - does not match the mood method, but the "not" keyword is given.
921
922 The set of admissible files for the whole mood is now defined as those
923 files which match at least one accept mood line, but no deny mood line.
924 More formally, an audio file F is admissible if and only if
925
926 (F ~ AL1 or F ~ AL2...) and not (F ~ DL1 or F ~ DN2 ...)
927
928 where AL1, AL2... are the accept lines, DL1, DL2... are the deny
929 lines and "~" means "matches".
930
931 The cases where no mood lines of accept/deny type are defined need
932 special treatment:
933
934 - Neither accept nor deny lines: This treats all files as
935 admissible (in fact, that is the definition of the dummy mood
936 which is activated automatically if no moods are available).
937
938 - Only accept lines: A file is admissible iff it matches at
939 least one accept line:
940
941 F ~ AL1 or F ~ AL2 or ...
942
943 - Only deny lines: A file is admissible iff it matches no
944 deny line:
945
946 not (F ~ DL1 or F ~ DN2 ...)
947
948
949
950 *List of mood_methods*
951
952 no_attributes_set
953
954 Takes no arguments and matches an audio file if and only if no
955 attributes are set.
956
957 is_set <attribute_name>
958
959 Takes the name of an attribute and matches iff that attribute is set.
960
961 path_matches <pattern>
962
963 Takes a filename pattern and matches iff the path of the audio file
964 matches the pattern.
965
966 artist_matches <pattern>
967 album_matches <pattern>
968 title_matches <pattern>
969 comment_matches <pattern>
970
971 Takes an extended regular expression and matches iff the text of the
972 corresponding tag of the audio file matches the pattern. If the tag
973 is not set, the empty string is matched against the pattern.
974
975 year ~ <num>
976 bitrate ~ <num>
977 frequency ~ <num>
978 channels ~ <num>
979 num_played ~ <num>
980
981 Takes a comparator ~ of the set {<, =, <=, >, >=, !=} and a number
982 <num>. Matches an audio file iff the condition <val> ~ <num> is
983 satisfied where val is the corresponding value of the audio file
984 (value of the year tag, bitrate in kbit/s, frequency in Hz, channel
985 count, play count).
986
987 The year tag is special as its value is undefined if the audio file
988 has no year tag or the content of the year tag is not a number. Such
989 audio files never match. Another difference is the special treatment
990 if the year tag is a two-digit number. In this case either 1900 or
991 2000 is added to the tag value, depending on whether the number is
992 greater than 2000 plus the current year.
993
994
995 *Mood usage*
996
997 To create a new mood called "my_mood", write its definition into
998 some temporary file, say "tmpfile", and add it to the mood table
999 by executing
1000
1001 para addmood my_mood < tmpfile
1002
1003 If the mood definition is really short, you may just pipe it to the
1004 client instead of using temporary files. Like this:
1005
1006 echo "$MOOD_DEFINITION" | para addmood my_mood
1007
1008 There is no need to keep the temporary file since you can always use
1009 the catmood command to get it back:
1010
1011 para catmood my_mood
1012
1013 A mood can be activated by executing
1014
1015 para select m/my_mood
1016
1017 Once active, the list of admissible files is shown by the ls command
1018 if the "-a" switch is given:
1019
1020 para ls -a
1021
1022
1023 *Example mood definition*
1024
1025 Suppose you have defined attributes "punk" and "rock" and want to define
1026 a mood containing only Punk-Rock songs. That is, an audio file should be
1027 admissible if and only if both attributes are set. Since
1028
1029 punk and rock
1030
1031 is obviously the same as
1032
1033 not (not punk or not rock)
1034
1035 (de Morgan's rule), a mood definition that selects only Punk-Rock
1036 songs is
1037
1038 deny if not is_set punk
1039 deny if not is_set rock
1040
1041
1042
1043 File renames and content changes
1044 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1045
1046 Since the audio file selector knows the SHA1 of each audio file that
1047 has been added to the afs database, it recognizes if the content of
1048 a file has changed, e.g. because an ID3 tag was added or modified.
1049 Also, if a file has been renamed or moved to a different location,
1050 afs will detect that an entry with the same hash value already exists
1051 in the audio file table.
1052
1053 In both cases it is enough to just re-add the new file. In the
1054 first case (file content changed), the audio table is updated, while
1055 metadata such as the num_played and last_played fields, as well as
1056 the attributes, remain unchanged. In the other case, when the file
1057 is moved or renamed, only the path information is updated, all other
1058 data remains as before.
1059
1060 It is possible to change the behaviour of the add command by using the
1061 "-l" (lazy add) or the "-f" (force add) option.
1062
1063 Troubleshooting
1064 ~~~~~~~~~~~~~~~
1065
1066 Use the debug loglevel (option -l debug for most commands) to show
1067 debugging info. Almost all paraslash executables have a brief online
1068 help which is displayed by using the -h switch. The --detailed-help
1069 option prints the full help text.
1070
1071 If para_server crashed or was killed by SIGKILL (signal 9), it
1072 may refuse to start again because of "dirty osl tables". In this
1073 case you'll have to run the oslfsck program of libosl to fix your
1074 database. It might be necessary to use --force (even if your name
1075 isn't Luke). However, make sure para_server isn't running before
1076 executing oslfsck --force.
1077
1078 If you don't mind to recreate your database you can start
1079 from scratch by removing the entire database directory, i.e.
1080
1081 rm -rf ~/.paraslash/afs_database-0.4
1082
1083 Be aware that this removes all attribute definitions, all playlists
1084 and all mood definitions and requires to re-initialize the tables.
1085
1086 Although oslfsck fixes inconsistencies in database tables it doesn't
1087 care about the table contents. To check for invalid table contents, use
1088
1089 para_client check
1090
1091 This prints out references to missing audio files as well as invalid
1092 playlists and mood definitions.
1093
1094 ---------------------------------------
1095 Audio formats and audio format handlers
1096 ---------------------------------------
1097
1098 Audio formats
1099 ~~~~~~~~~~~~~
1100
1101 The following audio formats are supported by paraslash:
1102
1103 *MP3*
1104
1105 Mp3, MPEG-1 Audio Layer 3, is a common audio format for audio storage,
1106 designed as part of its MPEG-1 standard. An MP3 file is made up of
1107 multiple MP3 frames, which consist of a header and a data block. The
1108 size of an MP3 frame depends on the bit rate and on the number
1109 of channels. For a typical CD-audio file (sample rate of 44.1 kHz
1110 stereo), encoded with a bit rate of 128 kbit, an MP3 frame is about
1111 400 bytes large.
1112
1113 *OGG/Vorbis*
1114
1115 OGG is a standardized audio container format, while Vorbis is an
1116 open source codec for lossy audio compression. Since Vorbis is most
1117 commonly made available via the OGG container format, it is often
1118 referred to as OGG/Vorbis. The OGG container format divides data into
1119 chunks called OGG pages. A typical OGG page is about 4KB large. The
1120 Vorbis codec creates variable-bitrate (VBR) data, where the bitrate
1121 may vary considerably.
1122
1123 *OGG/Speex*
1124
1125 Speex is an open-source speech codec that is based on CELP (Code
1126 Excited Linear Prediction) coding. It is designed for voice
1127 over IP applications, has modest complexity and a small memory
1128 footprint. Wideband and narrowband (telephone quality) speech are
1129 supported. As for Vorbis audio, Speex bit-streams are often stored
1130 in OGG files.
1131
1132 *AAC*
1133
1134 Advanced Audio Coding (AAC) is a standardized, lossy compression
1135 and encoding scheme for digital audio which is the default audio
1136 format for Apple's iPhone, iPod, iTunes. Usually MPEG-4 is used as
1137 the container format and audio files encoded with AAC have the .m4a
1138 extension. A typical AAC frame is about 700 bytes large.
1139
1140 *WMA*
1141
1142 Windows Media Audio (WMA) is an audio data compression technology
1143 developed by Microsoft. A WMA file is usually encapsulated in the
1144 Advanced Systems Format (ASF) container format, which also specifies
1145 how meta data about the file is to be encoded. The bit stream of WMA
1146 is composed of superframes, each containing one or more frames of
1147 2048 samples. For 16 bit stereo a WMA superframe is about 8K large.
1148
1149 *FLAC*
1150
1151 The Free Lossless Audio Codec (FLAC) compresses audio without quality
1152 loss. It gives better compression ratios than a general purpose
1153 compressor like zip or bzip2 because FLAC is designed specifically
1154 for audio. A FLAC-encoded file consits of frames of varying size, up
1155 to 16K. Each frame starts with a header that contains all information
1156 necessary to decode the frame.
1157
1158 Meta data
1159 ~~~~~~~~~
1160
1161 Unfortunately, each audio format has its own conventions how meta
1162 data is added as tags to the audio file.
1163
1164 For MP3 files, ID3, version 1 and 2 are widely used. ID3 version 1
1165 is rather simple but also very limited as it supports only artist,
1166 title, album, year and comment tags. Each of these can only be at most
1167 32 characters long. ID3, version 2 is much more flexible but requires
1168 a separate library being installed for paraslash to support it.
1169
1170 Ogg vorbis, ogg speex and flac files contain meta data as Vorbis
1171 comments, which are typically implemented as strings of the form
1172 "[TAG]=[VALUE]". Unlike ID3 version 1 tags, one may use whichever
1173 tags are appropriate for the content.
1174
1175 AAC files usually use the MPEG-4 container format for storing meta
1176 data while WMA files wrap meta data as special objects within the
1177 ASF container format.
1178
1179 paraslash only tracks the most common tags that are supported by
1180 all tag variants: artist, title, year, album, comment. When a file
1181 is added to the AFS database, the meta data of the file is extracted
1182 and stored in the audio file table.
1183
1184 Chunks and chunk tables
1185 ~~~~~~~~~~~~~~~~~~~~~~~
1186
1187 paraslash uses the word "chunk" as common term for the building blocks
1188 of an audio file. For MP3 files, a chunk is the same as an MP3 frame,
1189 while for OGG files a chunk is an OGG page, etc. Therefore the chunk
1190 size varies considerably between audio formats, from a few hundred
1191 bytes (MP3) up to 16K (FLAC).
1192
1193 The chunk table contains the offsets within the audio file that
1194 correspond to the chunk boundaries of the file. Like the meta data,
1195 the chunk table is computed and stored in the database whenever an
1196 audio file is added.
1197
1198 The paraslash senders (see below) always send complete chunks. The
1199 granularity for seeking is therefore determined by the chunk size.
1200
1201 Audio format handlers
1202 ~~~~~~~~~~~~~~~~~~~~~
1203
1204 For each audio format paraslash contains an audio format handler whose
1205 first task is to tell whether a given file is a valid audio file of
1206 this type. If so, the audio file handler extracts some technical data
1207 (duration, sampling rate, number of channels etc.), computes the
1208 chunk table and reads the meta data.
1209
1210 The audio format handler code is linked into para_server and executed
1211 via the _add_ command. The same code is also available as a stand-alone
1212 tool, para_afh, which can be used to print the technical data, the
1213 chunk table and the meta data of a file. Furthermore, one can use
1214 para_afh to cut an audio file, i.e. to select some of its chunks to
1215 produce a new file containing only these chunks.
1216
1217 ----------
1218 Networking
1219 ----------
1220
1221 Paraslash uses different network connections for control and data.
1222 para_client communicates with para_server over a dedicated TCP control
1223 connection. To transport audio data, separate data connections are
1224 used. For these data connections, a variety of transports (UDP, DCCP,
1225 HTTP) can be chosen.
1226
1227 The chapter starts with the REFERENCE(The paraslash control
1228 service, control service), followed by a section on the various
1229 REFERENCE(Streaming protocols, streaming protocols) in which the data
1230 connections are described. The way audio file headers are embedded into
1231 the stream is discussed REFERENCE(Streams with headers and headerless
1232 streams, briefly) before the REFERENCE(Networking examples, example
1233 section) which illustrates typical commands for real-life scenarios.
1234
1235 Both IPv4 and IPv6 are supported.
1236
1237 The paraslash control service
1238 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1239
1240 para_server is controlled at runtime via the paraslash control
1241 connection. This connection is used for server commands (play, stop,
1242 ...) as well as for afs commands (ls, select, ...).
1243
1244 The server listens on a TCP port and accepts connections from clients
1245 that connect the open port. Each connection causes the server to fork
1246 off a client process which inherits the connection and deals with that
1247 client only. In this classical accept/fork approach the server process
1248 is unaffected if the child dies or goes crazy for whatever reason. In
1249 fact, the child process can not change address space of server process.
1250
1251 The section on REFERENCE(Client-server authentication, client-server
1252 authentication) above described the early connection establishment
1253 from the crypto point of view. Here it is described what happens
1254 after the connection (including crypto setup) has been established.
1255 There are four processes involved during command dispatch as sketched
1256 in the following diagram.
1257
1258 <<
1259 <pre>
1260 server_host client_host
1261 ~~~~~~~~~~~ ~~~~~~~~~~~
1262
1263 +-----------+ connect +-----------+
1264 |para_server|<------------------------------ |para_client|
1265 +-----------+ +-----------+
1266 | ^
1267 | fork +---+ |
1268 +----------> |AFS| |
1269 | +---+ |
1270 | ^ |
1271 | | |
1272 | | connect (cookie) |
1273 | | |
1274 | | |
1275 | fork +-----+ inherited connection |
1276 +---------->|child|<--------------------------+
1277 +-----+
1278 </pre>
1279 >>
1280
1281 Note that the child process is not a child of the afs process,
1282 so communication of these two processes has to happen via local
1283 sockets. In order to avoid abuse of the local socket by unrelated
1284 processes, a magic cookie is created once at server startup time just
1285 before the server process forks off the AFS process. This cookie is
1286 known to the server, AFS and the child, but not to unrelated processes.
1287
1288 There are two different kinds of commands: First there are commands
1289 that cause the server to respond with some answer such as the list
1290 of all audio files. All but the addblob commands (addimg, addlyr,
1291 addpl, addmood) are of this kind. The addblob commands add contents
1292 to the database, so they need to transfer data the other way round,
1293 from the client to the server.
1294
1295 There is no knowledge about the server commands built into para_client,
1296 so it does not know about addblob commands. Instead, it inspects the
1297 first data package sent by the server for a magic string. If this
1298 string was found, it sends STDIN to the server, otherwise it dumps
1299 data from the server to STDOUT.
1300
1301 Streaming protocols
1302 ~~~~~~~~~~~~~~~~~~~
1303
1304 A network (audio) stream usually consists of one streaming source,
1305 the _sender_, and one or more _receivers_ which read data over the
1306 network from the streaming source.
1307
1308 Senders are thus part of para_server while receivers are part of
1309 para_audiod. Moreover, there is the stand-alone tool para_recv which
1310 can be used to manually download a stream, either from para_server
1311 or from a web-based audio streaming service.
1312
1313 The following three streaming protocols are supported by paraslash:
1314
1315 - HTTP. Recommended for public streams that can be played by
1316 any player like mpg123, xmms, itunes, winamp, etc. The HTTP
1317 sender is supported on all operating systems and all platforms.
1318
1319 - DCCP. Recommended for LAN streaming. DCCP is currently
1320 available only for Linux.
1321
1322 - UDP. Recommended for multicast LAN streaming.
1323
1324 See the Appendix on REFERENCE(Network protocols, network protocols)
1325 for brief descriptions of the various protocols relevant for network
1326 audio streaming with paraslash.
1327
1328 It is possible to activate more than one sender simultaneously.
1329 Senders can be controlled at run time and via config file and command
1330 line options.
1331
1332 Note that audio connections are _not_ encrypted. Transport or Internet
1333 layer encryption should be used if encrypted data connections are
1334 needed.
1335
1336 Since DCCP and TCP are both connection-oriented protocols, connection
1337 establishment/teardown and access control are very similar between
1338 these two streaming protocols. UDP is the most lightweight option,
1339 since in contrast to TCP/DCCP it is connectionless. It is also the
1340 only protocol supporting IP multicast.
1341
1342 The HTTP and the DCCP sender listen on a (TCP/DCCP) port waiting for
1343 clients to connect and establish a connection via some protocol-defined
1344 handshake mechanism. Both senders maintain two linked lists each:
1345 The list of all clients which are currently connected, and the list
1346 of access control entries which determines who is allowed to connect.
1347 IP-based access control may be configured through config file and
1348 command line options and via the "allow" and "deny" sender subcommands.
1349
1350 Upon receiving a GET request from the client, the HTTP sender sends
1351 back a status line and a message. The body of this message is the
1352 audio stream. This is common practice and is supported by many popular
1353 clients which can thus be used to play a stream offered by para_server.
1354 For DCCP things are a bit simpler: No messages are exchanged between
1355 the receiver and sender. The client simply connects and the sender
1356 starts to stream.
1357
1358 DCCP is an experimental protocol which offers a number of new features
1359 not available for TCP. Both ends can negotiate these features using
1360 a built-in negotiation mechanism. In contrast to TCP/HTTP, DCCP is
1361 datagram-based (no retransmissions) and thus should not be used over
1362 lossy media (e.g. WiFi networks). One useful feature offered by DCCP
1363 is access to a variety of different congestion-control mechanisms
1364 called CCIDs. Two different CCIDs are available per default on Linux:
1365
1366
1367 - _CCID 2_. A Congestion Control mechanism similar to that
1368 of TCP. The sender maintains a congestion window and halves
1369 this window in response to congestion.
1370
1371
1372 - _CCID-3_. Designed to be fair when competing for bandwidth.
1373 It has lower variation of throughput over time compared with
1374 TCP, which makes it suitable for streaming media.
1375
1376 Unlike the HTTP and DCCP senders, the UDP sender maintains only a
1377 single list, the _target list_. This list describes the set of clients
1378 to which the stream is sent. There is no list for access control and
1379 no "allow" and "deny" commands for the UDP sender. Instead, the "add"
1380 and "delete" commands can be used to modify the target list.
1381
1382 Since both UDP and DCCP offer an unreliable datagram-based transport,
1383 additional measures are necessary to guard against disruptions over
1384 networks that are lossy or which may be subject to interference (as
1385 is for instance the case with WiFi). Paraslash uses FEC (Forward
1386 Error Correction) to guard against packet losses and reordering. The
1387 stream is FEC-encoded before it is sent through the UDP socket and
1388 must be decoded accordingly on the receiver side.
1389
1390 The packet size and the amount of redundancy introduced by FEC can
1391 be configured via the FEC parameters which are dictated by server
1392 and may also be configured through the "sender" command. The FEC
1393 parameters are encoded in the header of each network packet, so no
1394 configuration is necessary on the receiver side. See the section on
1395 REFERENCE(Forward error correction, FEC) below.
1396
1397 Streams with headers and headerless streams
1398 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1399
1400 For OGG/Vorbis, OGG/Speex and wma streams, some of the information
1401 needed to decode the stream is only contained in the audio file
1402 header of the container format but not in each data chunk. Clients
1403 must be able to obtain this information in case streaming starts in
1404 the middle of the file or if para_audiod is started while para_server
1405 is already sending a stream.
1406
1407 This is accomplished in different ways, depending on the streaming
1408 protocol. For connection-oriented streams (HTTP, DCCP) the audio file
1409 header is sent prior to audio file data. This technique however does
1410 not work for the connectionless UDP transport. Hence the audio file
1411 header is periodically being embedded into the UDP audio data stream.
1412 By default, the header is resent after five seconds. The receiver has
1413 to wait until the next header arrives before it can start decoding
1414 the stream.
1415
1416 Examples
1417 ~~~~~~~~
1418
1419 The sender command of para_server allows to (de-)activate senders
1420 and to change the access permissions senders at runtime. The "si"
1421 (server info) command is used to list the streaming options of the
1422 currently running server as well as the various sender access lists.
1423
1424 -> Show client/target/access lists:
1425
1426 para_client si
1427
1428 -> Obtain general help for the sender command:
1429
1430 para_client help sender
1431
1432 -> Get help for a specific sender (contains further examples):
1433
1434 s=http # or dccp or udp
1435 para_client sender $s help
1436
1437 By default para_server activates both the HTTP and th DCCP sender on
1438 startup. This can be changed via command line options or para_server's
1439 config file.
1440
1441 -> List config file options for senders:
1442
1443 para_server -h
1444
1445 All senders share the "on" and "off" commands, so senders may be
1446 activated and deactivated independently of each other.
1447
1448 -> Switch off the http sender:
1449
1450 para_client sender http off
1451
1452 -> Receive a DCCP stream using CCID2 and write the output into a file:
1453
1454 host=foo.org; ccid=2; filename=bar
1455 para_recv --receiver "dccp --host $host --ccid $ccid" > $filename
1456
1457 Note the quotes around the arguments for the dccp receiver. Each
1458 receiver has its own set of command line options and its own command
1459 line parser, so arguments for the dccp receiver must be protected
1460 from being interpreted by para_recv.
1461
1462 -> Start UDP multicast, using the default multicast address:
1463
1464 para_client sender udp add 224.0.1.38
1465
1466 -> Receive FEC-encoded multicast stream and write the output into a file:
1467
1468 filename=foo
1469 para_recv -r udp > $filename
1470
1471 -> Add an UDP unicast for a client to the target list of the UDP sender:
1472
1473 t=client.foo.org
1474 para_client sender udp add $t
1475
1476 -> Receive this (FEC-encoded) unicast stream:
1477
1478 filename=foo
1479 para_recv -r 'udp -i 0.0.0.0' > $filename
1480
1481 -> Create a minimal config for para_audiod for HTTP streams:
1482
1483 c=$HOME/.paraslash/audiod.conf.min; s=server.foo.com
1484 echo receiver \".:http -i $s\" > $c
1485 para_audiod --config $c
1486
1487 -------
1488 Filters
1489 -------
1490
1491 A paraslash filter is a module which transforms an input stream into
1492 an output stream. Filters are included in the para_audiod executable
1493 and in the stand-alone tool para_filter which usually contains the
1494 same modules.
1495
1496 While para_filter reads its input stream from STDIN and writes
1497 the output to STDOUT, the filter modules of para_audiod are always
1498 connected to a receiver which produces the input stream and a writer
1499 which absorbs the output stream.
1500
1501 Some filters depend on a specific library being installed and are
1502 not compiled in if this library was not found at compile time. To
1503 see the list of supported filters, run para_filter and para_audiod
1504 with the --help option. The output looks similar to the following:
1505
1506 Available filters:
1507 compress wav amp fecdec wmadec prebuffer oggdec aacdec mp3dec
1508
1509 Out of these filter modules, a chain of filters can be constructed,
1510 much in the way Unix pipes can be chained, and analogous to the use
1511 of modules in gstreamer: The output of the first filter becomes the
1512 input of the second filter. There is no limitation on the number of
1513 filters and the same filter may occur more than once.
1514
1515 Like receivers, each filter has its own command line options which
1516 must be quoted to protect them from the command line options of
1517 the driving application (para_audiod or para_filter). Example:
1518
1519 para_filter -f 'mp3dec --ignore-crc' -f 'compress --damp 1'
1520
1521 For para_audiod, each audio format has its own set of filters. The
1522 name of the audio format for which the filter should be applied can
1523 be used as the prefix for the filter option. Example:
1524
1525 para_audiod -f 'mp3:prebuffer --duration 300'
1526
1527 The "mp3" prefix above is actually interpreted as a POSIX extended
1528 regular expression. Therefore
1529
1530 para_audiod -f '.:prebuffer --duration 300'
1531
1532 activates the prebuffer filter for all supported audio formats (because
1533 "." matches all audio formats) while
1534
1535 para_audiod -f 'wma|ogg:prebuffer --duration 300'
1536
1537 activates it only for wma and ogg streams.
1538
1539 Decoders
1540 ~~~~~~~~
1541
1542 For each supported audio format there is a corresponding filter
1543 which decodes audio data in this format to 16 bit PCM data which
1544 can be directly sent to the sound device or any other software that
1545 operates on undecoded PCM data (visualizers, equalizers etc.). Such
1546 filters are called _decoders_ in general, and xxxdec is the name of
1547 the paraslash decoder for the audio format xxx. For example, the mp3
1548 decoder filter is called mp3dec.
1549
1550 Note that the output of the decoder is about 10 times larger than
1551 its input. This means that filters that operate on the decoded audio
1552 stream have to deal with much more data than filters that transform
1553 the audio stream before it is fed to the decoder.
1554
1555 Paraslash relies on external libraries for most decoders, so these
1556 libraries must be installed for the decoder to be included in the
1557 para_filter and para_audiod executables. The oggdec filter depends
1558 on the libogg and libvorbis libraries for example.
1559
1560 Forward error correction
1561 ~~~~~~~~~~~~~~~~~~~~~~~~
1562
1563 As already mentioned REFERENCE(Streaming protocols, earlier),
1564 paraslash uses forward error correction (FEC) for the unreliable UDP
1565 and DCCP transports. FEC is a technique which was invented already
1566 in 1960 by Reed and Solomon and which is widely used for the parity
1567 calculations of storage devices (RAID arrays). It is based on the
1568 algebraic concept of finite fields, today called Galois fields, in
1569 honour of the mathematician Galois (1811-1832). The FEC implementation
1570 of paraslash is based on code by Luigi Rizzo.
1571
1572 Although the details require a sound knowledge of the underlying
1573 mathematics, the basic idea is not hard to understand: For positive
1574 integers k and n with k < n it is possible to compute for any k given
1575 data bytes d_1, ..., d_k the corresponding r := n -k parity bytes p_1,
1576 ..., p_r such that all data bytes can be reconstructed from *any*
1577 k bytes of the set
1578
1579 {d_1, ..., d_k, p_1, ..., p_r}.
1580
1581 FEC-encoding for unreliable network transports boils down to slicing
1582 the audio stream into groups of k suitably sized pieces called _slices_
1583 and computing the r corresponding parity slices. This step is performed
1584 in para_server which then sends both the data and the parity slices
1585 over the unreliable network connection. If the client was able
1586 to receive at least k of the n = k + r slices, it can reconstruct
1587 (FEC-decode) the original audio stream.
1588
1589 From these observations it is clear that there are three different
1590 FEC parameters: The slice size, the number of data slices k, and the
1591 total number of slices n. It is crucial to choose the slice size
1592 such that no fragmentation of network packets takes place because
1593 FEC only guards against losses and reordering but fails if slices are
1594 received partially.
1595
1596 FEC decoding in paralash is performed through the fecdec filter which
1597 usually is the first filter (there can be other filters before fecdec
1598 if these do not alter the audio stream).
1599
1600
1601 Volume adjustment (amp and compress)
1602 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1603
1604 The amp and the compress filter both adjust the volume of the audio
1605 stream. These filters operate on uncompressed audio samples. Hence
1606 they are usually placed directly after the decoding filter. Each
1607 sample is multiplied with a scaling factor (>= 1) which makes amp
1608 and compress quite expensive in terms of computing power.
1609
1610 *amp*
1611
1612 The amp filter amplifies the audio stream by a fixed scaling factor
1613 that must be known in advance. For para_audiod this factor is derived
1614 from the amplification field of the audio file's entry in the audio
1615 file table while para_filter uses the value given at the command line.
1616
1617 The optimal scaling factor F for an audio file is the largest real
1618 number F >= 1 such that after multiplication with F all samples still
1619 fit into the sample interval [-32768, 32767]. One can use para_filter
1620 in combination with the sox utility to compute F:
1621
1622 para_filter -f mp3dec -f wav < file.mp3 | sox -t wav - -e stat -v
1623
1624 The amplification value V which is stored in the audio file table,
1625 however, is an integer between 0 and 255 which is connected to F
1626 through the formula
1627
1628 V = (F - 1) * 64.
1629
1630 To store V in the audio file table, the command
1631
1632 para_client -- touch -a=V file.mp3
1633
1634 is used. The reader is encouraged to write a script that performs
1635 these computations :)
1636
1637 *compress*
1638
1639 Unlike the amplification filter, the compress filter adjusts the volume
1640 of the audio stream dynamically without prior knowledge about the peak
1641 value. It maintains the maximal volume of the last n samples of the
1642 audio stream and computes a suitable amplification factor based on that
1643 value and the various configuration options. It tries to chose this
1644 factor such that the adjusted volume meets the desired target level.
1645
1646 Note that it makes sense to combine amp and compress.
1647
1648 Misc filters (wav and prebuffer)
1649 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1650
1651 These filters are rather simple and do not modify the audio stream at
1652 all. The wav filter is only useful with para_filter and in connection
1653 with a decoder. It asks the decoder for the number of channels and the
1654 sample rate of the stream and adds a Microsoft wave header containing
1655 this information at the beginning. This allows to write wav files
1656 rather than raw PCM files (which do not contain any information about
1657 the number of channels and the sample rate).
1658
1659 The prebuffer filter simply delays the output until the given time has
1660 passed (starting from the time the first byte was available in its
1661 input queue) or until the given amount of data has accumulated. It
1662 is mainly useful for para_audiod if the standard parameters result
1663 in buffer underruns.
1664
1665 Both filters require almost no additional computing time, even when
1666 operating on uncompressed audio streams, since data buffers are simply
1667 "pushed down" rather than copied.
1668
1669 Examples
1670 ~~~~~~~~
1671
1672 -> Decode an mp3 file to wav format:
1673
1674 para_filter -f mp3dec -f wav < file.mp3 > file.wav
1675
1676 -> Amplify a raw audio file by a factor of 1.5:
1677
1678 para_filter -f amp --amp 32 < foo.raw > bar.raw
1679
1680 ------
1681 Output
1682 ------
1683
1684 Once an audio stream has been received and decoded to PCM format,
1685 it can be sent to a sound device for playback. This part is performed
1686 by paraslash _writers_ which are described in this chapter.
1687
1688 Writers
1689 ~~~~~~~
1690
1691 A paraslash writer acts as a data sink that consumes but does not
1692 produce audio data. Paraslash writers operate on the client side and
1693 are contained in para_audiod and in the stand-alone tool para_write.
1694
1695 The para_write program reads uncompressed audio data from STDIN. If
1696 this data starts with a wav header, sample rate, sample format and
1697 channel count are read from the header. Otherwise CD audio (44.1KHz
1698 16 bit little endian, stereo) is assumed but this can be overridden
1699 by command line options. para_audiod, on the other hand, obtains
1700 the sample rate and the number of channels from the decoder.
1701
1702 Like receivers and filters, each writer has an individual set of
1703 command line options, and for para_audiod writers can be configured
1704 per audio format separately. It is possible to activate more than
1705 one writer for the same stream simultaneously.
1706
1707 OS-dependent APIs
1708 ~~~~~~~~~~~~~~~~~
1709
1710 Unfortunately, the various flavours of Unix on which paraslash
1711 runs on have different APIs for opening a sound device and starting
1712 playback. Hence for each such API there is a paraslash writer that
1713 can play the audio stream via this API.
1714
1715 *ALSA*. The _Advanced Linux Sound Architecture_ is only available on
1716 Linux systems. Although there are several mid-layer APIs in use by
1717 the various Linux distributions (ESD, Jack, PulseAudio), paraslash
1718 currently supports only the low-level ALSA API which is not supposed
1719 to be change. ALSA is very feature-rich, in particular it supports
1720 software mixing via its DMIX plugin. ALSA is the default writer on
1721 Linux systems.
1722
1723 *OSS*. The _Open Sound System_ is the only API on *BSD Unixes and
1724 is also available on Linux systems, usually provided by ALSA as an
1725 emulation for backwards compatibility. This API is rather simple but
1726 also limited. For example only one application can open the device
1727 at any time. The OSS writer is activated by default on BSD Systems.
1728
1729 *OSX*. Mac OS X has yet another API called CoreAudio. The OSX writer
1730 for this API is only compiled in on such systems and is of course
1731 the default there.
1732
1733 *FILE*. The file writer allows to capture the audio stream and
1734 write the PCM data to a file on the file system rather than playing
1735 it through a sound device. It is supported on all platforms and is
1736 always compiled in.
1737
1738 *AO*. _Libao_ is a cross-platform audio library which supports a wide
1739 variety of platforms including PulseAudio (gnome), ESD (Enlightened
1740 Sound Daemon), AIX, Solaris and IRIX. The ao writer plays audio
1741 through an output plugin of libao.
1742
1743 Examples
1744 ~~~~~~~~
1745
1746 -> Use the OSS writer to play a wav file:
1747
1748 para_write --writer oss < file.wav
1749
1750 -> Enable ALSA software mixing for mp3 streams
1751
1752 para_audiod --writer 'mp3:alsa -d plug:swmix'
1753
1754
1755 ---
1756 Gui
1757 ---
1758
1759 para_gui executes an arbitrary command which is supposed to print
1760 status information to STDOUT. It then displays this information in
1761 a curses window. By default the command
1762
1763 para_audioc -- stat -p
1764
1765 is executed, but this can be customized via the --stat_cmd option. In
1766 particular it possible to use
1767
1768 para_client -- stat -p
1769
1770 to make para_gui work on systems on which para_audiod is not running.
1771
1772 Key bindings
1773 ~~~~~~~~~~~~
1774
1775 It is possible to bind keys to arbitrary commands via custom
1776 key-bindings. Besides the internal keys which can not be changed (help,
1777 quit, loglevel, version...), the following flavours of key-bindings
1778 are supported:
1779
1780 - external: Shutdown curses before launching the given command.
1781 Useful for starting other ncurses programs from within
1782 para_gui, e.g. aumix or dialog scripts. Or, use the mbox
1783 output format to write a mailbox containing one mail for each
1784 (admissible) file the audio file selector knows about. Then
1785 start mutt from within para_gui to browse your collection!
1786
1787 - display: Launch the command and display its stdout in
1788 para_gui's bottom window.
1789
1790 - para: Like display, but start "para_client <specified
1791 command>" instead of "<specified command>".
1792
1793 The general form of a key binding is
1794
1795 key_map k:m:c
1796
1797 which maps key k to command c using mode m. Mode may be x, d or p
1798 for external, display and paraslash commands, respectively.
1799
1800 Themes
1801 ~~~~~~
1802
1803 Currently there are only two themes for para_gui. It is easy, however,
1804 to add more themes. To create a new theme one has to define the
1805 position, color and geometry for for each status item that should be
1806 shown by this theme. See gui_theme.c for examples.
1807
1808 The "." and "," keys are used to switch between themes.
1809
1810 Examples
1811 ~~~~~~~~
1812
1813 -> Show server info:
1814
1815 key_map "i:p:si"
1816
1817 -> Jump to the middle of the current audio file by pressing F5:
1818
1819 key_map "<F5>:p:jmp 50"
1820
1821 -> vi-like bindings for jumping around:
1822
1823 key_map "l:p:ff 10"
1824 key_map "h:p:ff 10-"
1825 key_map "w:p:ff 60"
1826 key_map "b:p:ff 60-"
1827
1828 -> Print the current date and time:
1829
1830 key_map "D:d:date"
1831
1832 -> Call other curses programs:
1833
1834 key_map "U:x:aumix"
1835 key_map "!:x:/bin/bash"
1836 key_map "^E:x:/bin/sh -c 'vi ~/.paraslash/gui.conf'"
1837
1838 -----------
1839 Development
1840 -----------
1841
1842 Tools
1843 ~~~~~
1844
1845 In order to compile the sources from the git repository (rather than
1846 from tar balls) and for contributing non-trivial changes to the
1847 paraslash project, some additional tools should be installed on a
1848 developer machine.
1849
1850 http://git.or.cz/ (git). As described in more detail REFERENCE(Git
1851 branches, below), the git source code management tool is used for
1852 paraslash development. It is necessary for cloning the git repository
1853 and for getting updates.
1854
1855 ftp://ftp.gnu.org/pub/gnu/m4/ (m4). Some input files for gengetopt
1856 are generated from templates by the m4 macro processor.
1857
1858 ftp://ftp.gnu.org/pub/gnu/autoconf/ (autoconf) GNU autoconf creates
1859 the configure file which is shipped in the tarballs but has to be
1860 generated when compiling from git.
1861
1862 http://www.triptico.com/software/grutatxt.html (grutatxt). The
1863 HTML version of this manual and some of the paraslash web pages are
1864 generated by the grutatxt plain text to HTML converter. If changes
1865 are made to these text files the grutatxt package must be installed
1866 to regenerate the HTML files.
1867
1868 http://www.stack.nl/~dimitri/doxygen/ (doxygen). The documentation
1869 of paraslash's C sources uses the doxygen documentation system. The
1870 conventions for documenting the source code is described in the
1871 REFERENCE(Doxygen, Doxygen section).
1872
1873 ftp://ftp.gnu.org/pub/gnu/global (global). This is used to generate
1874 browsable HTML from the C sources. It is needed by doxygen.
1875
1876 Git branches
1877 ~~~~~~~~~~~~
1878
1879 Paraslash has been developed using the git source code management
1880 tool since 2006. Development is organized roughly in the same spirit
1881 as the git development itself, as described below.
1882
1883 The following text passage is based on "A note from the maintainer",
1884 written by Junio C Hamano, the maintainer of git.
1885
1886 There are four branches in the paraslash repository that track the
1887 source tree: "master", "maint", "next", and "pu".
1888
1889 The "master" branch is meant to contain what is well tested and
1890 ready to be used in a production setting. There could occasionally be
1891 minor breakages or brown paper bag bugs but they are not expected to
1892 be anything major, and more importantly quickly and easily fixable.
1893 Every now and then, a "feature release" is cut from the tip of this
1894 branch, named with three dotted decimal digits, like 0.4.2.
1895
1896 Whenever changes are about to be included that will eventually lead to
1897 a new major release (e.g. 0.5.0), a "maint" branch is forked off from
1898 "master" at that point. Obvious, safe and urgent fixes after the major
1899 release are applied to this branch and maintenance releases are cut
1900 from it. New features never go to this branch. This branch is also
1901 merged into "master" to propagate the fixes forward.
1902
1903 A trivial and safe enhancement goes directly on top of "master".
1904 New development does not usually happen on "master", however.
1905 Instead, a separate topic branch is forked from the tip of "master",
1906 and it first is tested in isolation; Usually there are a handful such
1907 topic branches that are running ahead of "master". The tip of these
1908 branches is not published in the public repository to keep the number
1909 of branches that downstream developers need to worry about low.
1910
1911 The quality of topic branches varies widely. Some of them start out as
1912 "good idea but obviously is broken in some areas" and then with some
1913 more work become "more or less done and can now be tested by wider
1914 audience". Luckily, most of them start out in the latter, better shape.
1915
1916 The "next" branch is to merge and test topic branches in the latter
1917 category. In general, this branch always contains the tip of "master".
1918 It might not be quite rock-solid production ready, but is expected to
1919 work more or less without major breakage. The maintainer usually uses
1920 the "next" version of paraslash for his own pleasure, so it cannot
1921 be _that_ broken. The "next" branch is where new and exciting things
1922 take place.
1923
1924 The two branches "master" and "maint" are never rewound, and "next"
1925 usually will not be either (this automatically means the topics that
1926 have been merged into "next" are usually not rebased, and you can find
1927 the tip of topic branches you are interested in from the output of
1928 "git log next"). You should be able to safely build on top of them.
1929
1930 However, at times "next" will be rebuilt from the tip of "master" to
1931 get rid of merge commits that will never be in "master". The commit
1932 that replaces "next" will usually have the identical tree, but it
1933 will have different ancestry from the tip of "master".
1934
1935 The "pu" (proposed updates) branch bundles the remainder of the
1936 topic branches. The "pu" branch, and topic branches that are only in
1937 "pu", are subject to rebasing in general. By the above definition
1938 of how "next" works, you can tell that this branch will contain quite
1939 experimental and obviously broken stuff.
1940
1941 When a topic that was in "pu" proves to be in testable shape, it
1942 graduates to "next". This is done with
1943
1944 git checkout next
1945 git merge that-topic-branch
1946
1947 Sometimes, an idea that looked promising turns out to be not so good
1948 and the topic can be dropped from "pu" in such a case.
1949
1950 A topic that is in "next" is expected to be polished to perfection
1951 before it is merged to "master". Similar to the above, this is
1952 done with
1953
1954 git checkout master
1955 git merge that-topic-branch
1956 git branch -d that-topic-branch
1957
1958 Note that being in "next" is not a guarantee to appear in the next
1959 release (being in "master" is such a guarantee, unless it is later
1960 found seriously broken and reverted), nor even in any future release.
1961
1962 Coding Style
1963 ~~~~~~~~~~~~
1964
1965 The preferred coding style for paraslash coincides more or less
1966 with the style of the Linux kernel. So rather than repeating what is
1967 written XREFERENCE(http://www.kernel.org/doc/Documentation/CodingStyle,
1968 there), here are the most important points.
1969
1970 - Burn the GNU coding standards.
1971 - Never use spaces for indentation.
1972 - Tabs are 8 characters, and thus indentations are also 8 characters.
1973 - Don't put multiple assignments on a single line.
1974 - Avoid tricky expressions.
1975 - Don't leave whitespace at the end of lines.
1976 - The limit on the length of lines is 80 columns.
1977 - Use K&R style for placing braces and spaces:
1978
1979 if (x is true) {
1980 we do y
1981 }
1982
1983 - Use a space after (most) keywords.
1984 - Do not add spaces around (inside) parenthesized expressions.
1985 - Use one space around (on each side of) most binary and ternary operators.
1986 - Do not use cute names like ThisVariableIsATemporaryCounter, call it tmp.
1987 - Mixed-case names are frowned upon.
1988 - Descriptive names for global variables are a must.
1989 - Avoid typedefs.
1990 - Functions should be short and sweet, and do just one thing.
1991 - The number of local variables shouldn't exceed 10.
1992 - Gotos are fine if they improve readability and reduce nesting.
1993 - Don't use C99-style "// ..." comments.
1994 - Names of macros defining constants and labels in enums are capitalized.
1995 - Enums are preferred when defining several related constants.
1996 - Always use the paraslash wrappers for allocating memory.
1997 - If the name of a function is an action or an imperative.
1998 command, the function should return an error-code integer
1999 (<0 means error, >=0 means success). If the name is a
2000 predicate, the function should return a "succeeded" boolean.
2001
2002
2003 Doxygen
2004 ~~~~~~~
2005
2006 Doxygen is a documentation system for various programming
2007 languages. The paraslash project uses Doxygen for generating the API
2008 reference on the web pages, but good source code documentation is
2009 also beneficial to people trying to understand the code structure
2010 and the interactions between the various source files.
2011
2012 It is more illustrative to look at the source code for examples than
2013 to describe the conventions for documenting the source in this manual,
2014 so we only describe which parts of the code need doxygen comments,
2015 but leave out details on documentation conventions.
2016
2017 As a rule, only the public part of the C source is documented with
2018 Doxygen. This includes structures, defines and enumerations in header
2019 files as well as public (non-static) C functions. These should be
2020 documented completely. For example each parameter and the return
2021 value of a public function should get a descriptive comment.
2022
2023 No doxygen comments are necessary for static functions and for
2024 structures and enumerations in C files (which are used only within
2025 this file). This does not mean, however, that those entities need
2026 no documentation at all. Instead, common sense should be applied to
2027 document what is not obvious from reading the code.
2028
2029 --------
2030 Appendix
2031 --------
2032
2033 Network protocols
2034 ~~~~~~~~~~~~~~~~~
2035
2036 *IP*. The _Internet Protocol_ is the primary networking protocol
2037 used for the Internet. All protocols described below use IP as the
2038 underlying layer. Both the prevalent IPv4 and the next-generation
2039 IPv6 variant are being deployed actively worldwide.
2040
2041 *Connection-oriented and connectionless protocols*. Connectionless
2042 protocols differ from connection-oriented ones in that state
2043 associated with the sending/receiving endpoints is treated
2044 implicitly. Connectionless protocols maintain no internal knowledge
2045 about the state of the connection. Hence they are not capable of
2046 reacting to state changes, such as sudden loss or congestion on the
2047 connection medium. Connection-oriented protocols, in contrast, make
2048 this knowledge explicit. The connection is established only after
2049 a bidirectional handshake which requires both endpoints to agree
2050 on the state of the connection, and may also involve negotiating
2051 specific parameters for the particular connection. Maintaining an
2052 up-to-date internal state of the connection also in general means
2053 that the sending endpoints perform congestion control, adapting to
2054 qualitative changes of the connection medium.
2055
2056 *Reliability*. In IP networking, packets can be lost, duplicated,
2057 or delivered out of order, and different network protocols handle
2058 these problems in different ways. We call a transport-layer protocol
2059 _reliable_, if it turns the unreliable IP delivery into an ordered,
2060 duplicate- and loss-free delivery of packets. Sequence numbers
2061 are used to discard duplicates and re-arrange packets delivered
2062 out-of-order. Retransmission is used to guarantee loss-free
2063 delivery. Unreliable protocols, in contrast, do not guarantee ordering
2064 or data integrity.
2065
2066 *Classification*. With these definitions the protocols which are used
2067 by paraslash for steaming audio data may be classified as follows.
2068
2069 - HTTP/TCP: connection-oriented, reliable,
2070 - UDP: connectionless, unreliable,
2071 - DCCP: connection-oriented, unreliable.
2072
2073 Below we give a short descriptions of these protocols.
2074
2075 *TCP*. The _Transmission Control Protocol_ provides reliable,
2076 ordered delivery of a stream and a classic window-based congestion
2077 control. In contrast to UDP and DCCP (see below), TCP does not have
2078 record-oriented or datagram-based syntax, i.e. it provides a stream
2079 which is unaware and independent of any record (packet) boundaries.
2080 TCP is used extensively by many application layers. Besides HTTP (the
2081 Hypertext Transfer Protocol), also FTP (the File Transfer protocol),
2082 SMTP (Simple Mail Transfer Protocol), SSH (Secure Shell) all sit on
2083 top of TCP.
2084
2085 *UDP*. The _User Datagram Protocol_ is the simplest transport-layer
2086 protocol, built as a thin layer directly on top of IP. For this reason,
2087 it offers the same best-effort service as IP itself, i.e. there is no
2088 detection of duplicate or reordered packets. Being a connectionless
2089 protocol, only minimal internal state about the connection is
2090 maintained, which means that there is no protection against packet
2091 loss or network congestion. Error checking and correction (if at all)
2092 are performed in the application.
2093
2094 *DCCP*. The _Datagram Congestion Control Protocol_ combines the
2095 connection-oriented state maintenance known from TCP with the
2096 unreliable, datagram-based transport of UDP. This means that it
2097 is capable of reacting to changes in the connection by performing
2098 congestion control, offering multiple alternative approaches. But it
2099 is bound to datagram boundaries (the maximum packet size supported
2100 by a medium), and like UDP it lacks retransmission to protect
2101 against loss. Due to the use of sequence numbers, it is however
2102 able to react to loss (interpreted as a congestion indication) and
2103 to ignore out-of-order and duplicate packets. Unlike TCP it allows
2104 to negotiate specific, binding features for a connection, such as
2105 the choice of congestion control: classic, window-based congestion
2106 control known from TCP is available as CCID-2, rate-based, "smooth"
2107 congestion control is offered as CCID-3.
2108
2109 *HTTP*. _The Hypertext Transfer Protocol_ is an application layer
2110 protocol on top of TCP. It is spoken by web servers and is most often
2111 used for web services. However, as can be seen by the many Internet
2112 radio stations and YouTube/Flash videos, http is by far not limited to
2113 the delivery of web pages only. Being a simple request/response based
2114 protocol, the semantics of the protocol also allow the delivery of
2115 multimedia content, such as audio over http.
2116
2117 *Multicast*. IP multicast is not really a protocol but a technique
2118 for one-to-many communication over an IP network. The challenge is to
2119 deliver information to a group of destinations simultaneously using
2120 the most efficient strategy to send the messages over each link of
2121 the network only once. This has benefits for streaming multimedia:
2122 the standard one-to-one unicast offered by TCP/DCCP means that
2123 n clients listening to the same stream also consume n-times the
2124 resources, whereas multicast requires to send the stream just once,
2125 irrespective of the number of receivers. Since it would be costly to
2126 maintain state for each listening receiver, multicast often implies
2127 connectionless transport, which is the reason that it is currently
2128 only available via UDP.
2129
2130 License
2131 ~~~~~~~
2132
2133 Paraslash is licensed under the GPL, version 2. Most of the code
2134 base has been written from scratch, and those parts are GPL V2
2135 throughout. Notable exceptions are FEC and the WMA decoder. See the
2136 corresponding source files for licencing details for these parts. Some
2137 code sniplets of several other third party software packages have
2138 been incorporated into the paraslash sources, for example log message
2139 coloring was taken from the git sources. These third party software
2140 packages are all published under the GPL or some other license
2141 compatible to the GPL.
2142
2143 Acknowledgements
2144 ~~~~~~~~~~~~~~~~
2145
2146 Many thanks to Gerrit Renker who read an early draft of this manual
2147 and contributed significant improvements.
2148
2149 ----------
2150 References
2151 ----------
2152
2153 Articles
2154 ~~~~~~~~
2155 - Reed, Irving S.; Solomon, Gustave (1960),
2156 XREFERENCE(http://kom.aau.dk/~heb/kurser/NOTER/KOFA01.PDF,
2157 Polynomial Codes over Certain Finite Fields), Journal of the
2158 Society for Industrial and Applied Mathematics (SIAM) 8 (2):
2159 300-304, doi:10.1137/0108018)
2160
2161 RFCs
2162 ~~~~
2163
2164 - XREFERENCE(http://www.ietf.org/rfc/rfc768.txt, RFC 768) (1980):
2165 User Datagram Protocol
2166 - XREFERENCE(http://www.ietf.org/rfc/rfc791.txt, RFC 791) (1981):
2167 Internet Protocol
2168 - XREFERENCE(http://www.ietf.org/rfc/rfc2437.txt, RFC 2437) (1998):
2169 RSA Cryptography Specifications
2170 - XREFERENCE(http://www.ietf.org/rfc/rfc4340.txt, RFC 4340)
2171 (2006): Datagram Congestion Control Protocol (DCCP)
2172 - XREFERENCE(http://www.ietf.org/rfc/rfc4341.txt, RFC 4341) (2006):
2173 Congestion Control ID 2: TCP-like Congestion Control
2174 - XREFERENCE(http://www.ietf.org/rfc/rfc4342.txt, RFC 4342) (2006):
2175 Congestion Control ID 3: TCP-Friendly Rate Control (TFRC)
2176
2177 Application web pages
2178 ~~~~~~~~~~~~~~~~~~~~~
2179
2180 - XREFERENCE(http://paraslash.systemlinux.org/, paraslash)
2181 - XREFERENCE(http://xmms2.org/wiki/Main_Page, xmms)
2182 - XREFERENCE(http://www.mpg123.de/, mpg123)
2183 - XREFERENCE(http://gstreamer.freedesktop.org/, gstreamer)
2184 - XREFERENCE(http://www.icecast.org/, icecast)
2185 - XREFERENCE(http://beesbuzz.biz/code/audiocompress.php, Audio Compress)
2186
2187 External documentation
2188 ~~~~~~~~~~~~~~~~~~~~~~
2189
2190 - XREFERENCE(http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf,
2191 H. Peter Anvin: The mathematics of Raid6)
2192 - XREFERENCE(http://info.iet.unipi.it/~luigi/fec_ccr.ps.gz,
2193 Luigi Rizzo: Effective Erasure Codes for reliable Computer
2194 Communication Protocols)
2195
2196 Code
2197 ~~~~
2198 - XREFERENCE(http://info.iet.unipi.it/~luigi/vdm.tar.gz,
2199 Original FEC implementation by Luigi Rizzo)
2200