Merge branch 't/configure_improvements'
[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
119 *para_audiod*
120
121 The local daemon that collects information from para_server.
122
123 It runs on the client side and connects to para_server. As soon as
124 para_server announces the availability of an audio stream, para_audiod
125 starts an appropriate receiver, any number of filters and a paraslash
126 writer to play the stream.
127
128 Moreover, para_audiod listens on a local socket and sends status
129 information about para_server and para_audiod to local clients on
130 request. Access via this local socket may be restricted by using Unix
131 socket credentials, if available.
132
133
134 *para_audioc*
135
136 The client program which talks to para_audiod. Used to control
137 para_audiod, to receive status info, or to grab the stream at any
138 point of the decoding process.
139
140 *para_recv*
141
142 A command line HTTP/DCCP/UDP stream grabber. The http mode is
143 compatible with arbitrary HTTP streaming sources (e.g. icecast).
144
145 *para_filter*
146
147 A filter program that reads from STDIN and writes to STDOUT.
148 Like para_recv, this is an atomic building block which can be used to
149 assemble higher-level audio receiving facilities. It combines several
150 different functionalities in one tool: decoders for multiple audio
151 formats and a number of processing filters, among these a normalizer
152 for audio volume.
153
154 *para_afh*
155
156 A small stand-alone program that prints tech info about the given
157 audio file to STDOUT. It can be instructed to print a "chunk table",
158 an array of offsets within the audio file or to write the content of
159 the audio file in complete chunks 'just in time'.
160
161 This allows third-party streaming software that is unaware of the
162 particular audio format to send complete frames in real time.
163
164 *para_write*
165
166 A modular audio stream writer. It supports a simple file writer
167 output plug-in and optional WAV/raw players for ALSA (Linux) and for
168 coreaudio (Mac OS). para_write can also be used as a stand-alone WAV
169 or raw audio player.
170
171
172 *para_gui*
173
174 Curses-based gui that presents status information obtained in a curses
175 window. Appearance can be customized via themes. para_gui provides
176 key-bindings for the most common server commands and new key-bindings
177 can be added easily.
178
179
180 *para_fade*
181
182 An (OSS-only) alarm clock and volume-fader.
183
184 -----------
185 Quick start
186 -----------
187
188 This chapter lists the REFERENCE(Requirements, necessary software)
189 that must be installed to compile the paraslash package, describes
190 how to REFERENCE(Installation, compile and install) the paraslash
191 source code and the steps that have to be performed in order to
192 REFERENCE(Quick start, set up) a typical server and client.
193
194 Requirements
195 ~~~~~~~~~~~~
196
197 In any case you'll need
198
199 - XREFERENCE(http://systemlinux.org/~maan/osl/, libosl).
200 The _object storage layer_ library is used by para_server. To
201 clone the source code repository, execute
202
203 git clone git://git.tuebingen.mpg.de/osl
204
205 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gcc, gcc). The
206 EMPH(gnu compiler collection) is usually shipped with the
207 distro. gcc-3.3 or newer is required.
208
209 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/make, gnu make) is
210 also shipped with the disto. On BSD systems the gnu make
211 executable is often called gmake.
212
213 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/bash, bash). Some
214 scripts which run during compilation require the EMPH(Bourne
215 again shell). It is most likely already installed.
216
217 - XREFERENCE(http://www.openssl.org/, openssl). The EMPH(Secure
218 Sockets Layer) library is needed for cryptographic routines
219 on both the server and the client side. It is usually shipped
220 with the distro, but you might have to install the "development
221 package" (called libssl-dev on debian systems) as well.
222
223 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/help2man, help2man)
224 is used to create the man pages.
225
226 Optional:
227
228 - XREFERENCE(http://www.underbit.com/products/mad/, libmad).
229 To compile in MP3 support for paraslash, the development
230 package must be installed. It is called libmad0-dev on
231 debian-based systems. Note that libmad is not necessary on
232 the server side, i.e. for sending MP3 files.
233
234 - XREFERENCE(http://www.underbit.com/products/mad/,
235 libid3tag). For version-2 ID3 tag support, you'll need
236 the libid3tag development package libid3tag0-dev. Without
237 libid3tag, only version one tags are recognized.
238
239 - XREFERENCE(http://www.xiph.org/downloads/, ogg vorbis).
240 For ogg vorbis streams you'll need libogg, libvorbis,
241 libvorbisfile. The corresponding Debian packages are called
242 libogg-dev and libvorbis-dev.
243
244 - XREFERENCE(http://www.audiocoding.com/, libfaad). For aac
245 files (m4a) you'll need libfaad (libfaad-dev).
246
247 - XREFERENCE(http://www.speex.org/, speex). In order to stream
248 or decode speex files, libspeex (libspeex-dev) is required.
249
250 - XREFERENCE(ftp://ftp.alsa-project.org/pub/lib/, alsa-lib). On
251 Linux, you'll need to have ALSA's development package
252 libasound2-dev installed.
253
254 Installation
255 ~~~~~~~~~~~~
256
257 First make sure all non-optional packages listed in the section on
258 REFERENCE(Requirements, required software) are installed on your
259 system.
260
261 You don't need everything listed there. In particular, MP3, OGG/Vorbis,
262 OGG/Speex and AAC support are all optional. The configure script will
263 detect what is installed on your system and will only try to build
264 those executables that can be built with your setup.
265
266 Note that no special decoder library (not even the MP3 decoding library
267 libmad) is needed for para_server if you only want to stream MP3 or WMA
268 files. Also, it's fine to use para_server on a box without sound card.
269
270 Next, install the paraslash package on all machines, you'd like this
271 software to run on:
272
273 (./configure && make) > /dev/null
274
275 There should be no errors but probably some warnings about missing
276 packages which usually implies that not all audio formats will be
277 supported. If headers or libs are installed at unusual locations you
278 might need to tell the configure script where to find them. Try
279
280 ./configure --help
281
282 to see a list of options. If the paraslash package was compiled
283 successfully, execute (optionally)
284
285 make test
286
287 to run the paraslash test suite. If all tests pass, execute as root
288
289 make install
290
291 to install executables under /usr/local/bin and the man pages under
292 /usr/local/man.
293
294 Configuration
295 ~~~~~~~~~~~~~
296
297 *Step 1*: Create a paraslash user
298
299 In order to control para_server at runtime you must create a paraslash
300 user. As authentication is based on the RSA crypto system you'll have
301 to create an RSA key pair. If you already have a user and an RSA key
302 pair, you may skip this step.
303
304 In this section we'll assume a typical setup: You would like to run
305 para_server on some host called server_host as user foo, and you want
306 to connect to para_server from another machine called client_host as
307 user bar.
308
309 As foo@server_host, create ~/.paraslash/server.users by typing the
310 following commands:
311
312 user=bar
313 target=~/.paraslash/server.users
314 key=~/.paraslash/key.pub.$user
315 perms=AFS_READ,AFS_WRITE,VSS_READ,VSS_WRITE
316 mkdir -p ~/.paraslash
317 echo "user $user $key $perms" >> $target
318
319 Next, change to the "bar" account on client_host and generate the
320 key pair with the commands
321
322 key=~/.paraslash/key.$LOGNAME
323 mkdir -p ~/.paraslash
324 (umask 077 && openssl genrsa -out $key 2048)
325
326 para_server only needs to know the public key of the key pair just
327 created. It can be extracted with
328
329 pubkey=~/.paraslash/key.pub.$LOGNAME
330 openssl rsa -in $key -pubout -out $pubkey
331
332 Copy the public key just created to server_host (you may skip this step
333 for a single-user setup, i.e. if foo=bar and server_host=client_host):
334
335 scp $pubkey foo@server_host:.paraslash/
336
337 Finally, tell para_client to connect to server_host:
338
339 conf=~/.paraslash/client.conf
340 echo 'hostname server_host' > $conf
341
342
343 *Step 2*: Start para_server
344
345 Before starting the server make sure you have write permissions to
346 the directory /var/paraslash that has been created during installation:
347
348 sudo chown $LOGNAME /var/paraslash
349
350 Alternatively, use the --afs_socket Option to specify a different
351 location for the AFS command socket.
352
353 For this first try, we'll use the info loglevel to make the output
354 of para_server more verbose.
355
356 para_server -l info
357
358 Now you can use para_client to connect to the server and issue
359 commands. Open a new shell as bar@client_host and try
360
361 para_client help
362 para_client si
363
364 to retrieve the list of available commands and some server info.
365 Don't proceed if this doesn't work.
366
367 *Step 3*: Create and populate the database
368
369 An empty database is created with
370
371 para_client init
372
373 This initializes a couple of empty tables under
374 ~/.paraslash/afs_database-0.4. You normally don't need to look at these
375 tables, but it's good to know that you can start from scratch with
376
377 rm -rf ~/.paraslash/afs_database-0.4
378
379 in case something went wrong.
380
381 Next, you need to add some audio files to that database so that
382 para_server knows about them. Choose an absolute path to a directory
383 containing some audio files and add them to the audio file table:
384
385 para_client add /my/mp3/dir
386
387 This might take a while, so it is a good idea to start with a directory
388 containing not too many files. Note that the table only contains data
389 about the audio files found, not the files themselves.
390
391 You may print the list of all known audio files with
392
393 para_client ls
394
395 *Step 4*: Configure para_audiod
396
397 para_audiod needs to create a "well-known" socket for the clients to
398 connect to. The default path for this socket is
399
400 /var/paraslash/audiod_socket.$HOSTNAME
401
402 In order to make this directory writable for para_audiod, execute
403 as bar@client_host
404
405 sudo chown $LOGNAME /var/paraslash
406
407
408 We will also have to tell para_audiod that it should receive the
409 audio stream from server_host:
410
411 para_audiod -l info -r 'mp3:http -i server_host'
412
413 You should now be able to listen to the audio stream once para_server
414 starts streaming. To activate streaming, execute
415
416 para_client play
417
418 Since no playlist has been specified yet, the "dummy" mode which
419 selects all known audio files is activated automatically. See the
420 section on the REFERENCE(The audio file selector, audio file selector)
421 for how to use playlists and moods to specify which files should be
422 streamed in which order.
423
424 *Troubleshooting*
425
426 It did not work? To find out why, try to receive, decode and play the
427 stream manually using para_recv, para_filter and para_write as follows.
428
429 For simplicity we assume that you're running Linux/ALSA and that only
430 MP3 files have been added to the database.
431
432 para_recv -r 'http -i server_host' > file.mp3
433 # (interrupt with CTRL+C after a few seconds)
434 ls -l file.mp3 # should not be empty
435 para_filter -f mp3dec -f wav < file.mp3 > file.wav
436 ls -l file.wav # should be much bigger than file.mp3
437 para_write -w alsa < file.wav
438
439 Double check what is logged by para_server and use the --loglevel
440 option of para_recv, para_filter and para_write to increase verbosity.
441
442 ---------------
443 User management
444 ---------------
445
446 para_server uses a challenge-response mechanism to authenticate
447 requests from incoming connections, similar to ssh's public key
448 authentication method. Authenticated connections are encrypted using
449 the RC4 stream cipher.
450
451 In this chapter we briefly describe RSA and RC4 and sketch the
452 REFERENCE(Client-server authentication, authentication handshake)
453 between para_client and para_server. User management is discussed
454 in the section on REFERENCE(The user_list file, the user_list file).
455 These sections are all about communication between the client and the
456 server. Connecting para_audiod is a different matter and is described
457 in a REFERENCE(Connecting para_audiod, separate section).
458
459
460
461 RSA and RC4
462 ~~~~~~~~~~~
463
464 RSA is an asymmetric block cipher which is used in many applications,
465 including ssh and gpg. An RSA key consists in fact of two keys,
466 called the public key and the private key. A message can be encrypted
467 with either key and only the counterpart of that key can decrypt
468 the message. While RSA can be used for both signing and encrypting
469 a message, paraslash only uses RSA only for the latter purpose. The
470 RSA public key encryption and signatures algorithms are defined in
471 detail in RFC 2437.
472
473 RC4 is a stream cipher, i.e. the input is XORed with a pseudo-random
474 key stream to produce the output. Decryption uses the same function
475 calls as encryption. While RC4 supports variable key lengths,
476 paraslash uses a fixed length of 256 bits, which is considered a
477 strong encryption by today's standards. Since the same key must never
478 be used twice, a different, randomly-generated key is used for every
479 new connection.
480
481 Client-server authentication
482 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
483
484 The authentication handshake between para_client and para_server goes
485 as follows:
486
487 - para_client connects to para_server and sends an
488 authentication request for a user. It does so by connecting
489 to para_server, TCP 2990, the control port of para_server.
490
491 - para_server accepts the connection and forks a child process
492 which is supposed to handle the connection. The parent process
493 keeps listening on the control port while the child process
494 (also called para_server below) continues as follows.
495
496 - para_server loads the RSA public key of that user, fills a
497 fixed-length buffer with random bytes, encrypts that buffer
498 using the public key and sends the encrypted buffer to the
499 client. The first part of the buffer is the challenge which
500 is used for authentication while the second part is the RC4
501 session key.
502
503 - para_client receives the encrypted buffer and decrypts it
504 using the user's private key, thereby obtaining the challenge
505 buffer and the session key. It sends the SHA1 hash value of
506 the challenge back to para_server and stores the session key
507 for further use.
508
509 - para_server also computes the SHA1 hash of the challenge
510 and compares it against what was sent back by the client.
511
512 - If the two hashes do not match, the authentication has
513 failed and para_server closes the connection.
514
515 - Otherwise the user is considered authenticated and the client
516 is allowed to proceed by sending a command to be executed. From
517 this point on the communication is encrypted using the RC4
518 stream cipher with the session key known to both peers.
519
520 paraslash relies on the quality of openssl's cryptographically strong
521 pseudo-random bytes, on the security of the implementation of the
522 openssl RSA and RC4 crypto routines and on the infeasibility to invert
523 the SHA1 function.
524
525 Neither para_server or para_client create RSA keys on their own. This
526 has to be done once for each user as sketched in REFERENCE(Quick start,
527 Quick start) and discussed in more detail REFERENCE(The user_list
528 file, below).
529
530 The user_list file
531 ~~~~~~~~~~~~~~~~~~
532
533 At startup para_server reads the user list file which must contain
534 one line per user. The default location of the user list file may be
535 changed with the --user_list option.
536
537 There should be at least one user in this file. Each user must have
538 an RSA key pair. The public part of the key is needed by para_server
539 while the private key is needed by para_client. Each line of the
540 user list file must be of the form
541
542 user <username> <key> <perms>
543
544 where _username_ is an arbitrary string (usually the user's login
545 name), _key_ is the full path to that user's public RSA key, and
546 _perms_ is a comma-separated list of zero or more of the following
547 permission bits:
548
549 +---------------------------------------------------------+
550 | AFS_READ | read the contents of the databases |
551 +-----------+---------------------------------------------+
552 | AFS_WRITE | change database contents |
553 +-----------+---------------------------------------------+
554 | VSS_READ | obtain information about the current stream |
555 +-----------+---------------------------------------------+
556 | VSS_WRITE | change the current stream |
557 +---------------------------------------------------------+
558
559 The permission bits specify which commands the user is allowed to
560 execute. The output of
561
562 para_client help
563
564 contains in the third column the permissions needed to execute the
565 command.
566
567 A new RSA key can be created with
568
569 openssl genrsa -out <private_key> 2048
570
571 and the public part may be extracted with
572
573 openssl rsa -in <private_key> -pubout -out <public_key>
574
575 Note that para_server refuses to use a key if it is shorter than 2048
576 bits. In particular, the RSA keys of paraslash 0.3.x will not work
577 with version 0.4.x. Moreover, para_client refuses to use a (private)
578 key which is world-readable.
579
580 It is possible to make para_server reread the user_list file by
581 executing the paraslash "hup" command or by sending SIGHUP to the
582 PID of para_server.
583
584
585 Connecting para_audiod
586 ~~~~~~~~~~~~~~~~~~~~~~
587
588 para_audiod listens on a Unix domain socket. Those sockets are
589 for local communication only, so only local users can connect to
590 para_audiod. The default is to let any user connect but this can be
591 restricted on platforms that support UNIX socket credentials which
592 allow para_audiod to obtain the Unix credentials of the connecting
593 process.
594
595 Use para_audiod's --user_allow option to allow connections only for
596 a limited set of users.
597
598 -----------------------
599 The audio file selector
600 -----------------------
601
602 paraslash comes with a sophisticated audio file selector (AFS),
603 whose main task is to determine which file to stream next, based on
604 information on the audio files stored in a database. It communicates
605 also with para_client whenever an AFS command is executed, for example
606 to answer a database query.
607
608 Besides the traditional playlists, AFS supports audio file selection
609 based on _moods_ which act as a filter that limits the set of all
610 known audio files to those which satisfy certain criteria. It also
611 maintains tables containing images (e.g. album cover art) and lyrics
612 that can be associated with one or more audio files.
613
614 AFS uses libosl, the object storage layer, as the backend library
615 for storing information on audio files, playlists, etc. This library
616 offers functionality similar to a relational database, but is much
617 more lightweight than a full database backend.
618
619 In this chapter we sketch the setup of the REFERENCE(The AFS process,
620 AFS process) during server startup and proceed with the description
621 of the REFERENCE(Database layout, layout) of the various database
622 tables. The section on REFERENCE(Playlists and moods, playlists
623 and moods) explains these two audio file selection mechanisms
624 in detail and contains pratical examples. The way REFERENCE(File
625 renames and content changes, file renames and content changes) are
626 detected is discussed briefly before the REFERENCE(Troubleshooting,
627 Troubleshooting) section which concludes the chapter.
628
629 The AFS process
630 ~~~~~~~~~~~~~~~
631
632 On startup, para_server forks to create the AFS process which opens
633 the OSL database tables. The server process communicates with the
634 AFS process via pipes and shared memory. Usually, the AFS process
635 awakes only briefly whenever the current audio file changes. The AFS
636 process determines the next audio file, opens it, verifies it has
637 not been changed since it was added to the database and passes the
638 open file descriptor to the server process, along with audio file
639 meta-data such as file name, duration, audio format and so on. The
640 server process then starts to stream the audio file.
641
642 The AFS process also accepts connections from local clients via
643 a well-known socket. However, only child processes of para_server
644 may connect through this socket. All server commands that have the
645 AFS_READ or AFS_WRITE permission bits use this mechanism to query or
646 change the database.
647
648 Database layout
649 ~~~~~~~~~~~~~~~
650
651 *The audio file table*
652
653 This is the most important and usually also the largest table of the
654 AFS database. It contains the information needed to stream each audio
655 file. In particular the following data is stored for each audio file.
656
657 - SHA1 hash value of the audio file contents. This is computed
658 once when the file is added to the database. Whenever AFS
659 selects this audio file for streaming the hash value is
660 recomputed and checked against the value stored in the
661 database to detect content changes.
662
663 - The time when this audio file was last played.
664
665 - The number of times the file has been played so far.
666
667 - The attribute bitmask.
668
669 - The image id which describes the image associated with this
670 audio file.
671
672 - The lyrics id which describes the lyrics associated with
673 this audio file.
674
675 - The audio format id (MP3, OGG, ...).
676
677 - An amplification value that can be used by the amplification
678 filter to pre-amplify the decoded audio stream.
679
680 - The chunk table. It describes the location and the timing
681 of the building blocks of the audio file. This is used by
682 para_server to send chunks of the file at appropriate times.
683
684 - The duration of the audio file.
685
686 - Tag information contained in the audio file (ID3 tags,
687 Vorbis comments, ...).
688
689 - The number of channels
690
691 - The encoding bitrate.
692
693 - The sampling frequency.
694
695 To add or refresh the data contained in the audio file table, the _add_
696 command is used. It takes the full path of either an audio file or a
697 directory. In the latter case, the directory is traversed recursively
698 and all files which are recognized as valid audio files are added to
699 the database.
700
701 *The attribute table*
702
703 The attribute table contains two columns, _name_ and _bitnum_. An
704 attribute is simply a name for a certain bit number in the attribute
705 bitmask of the audio file table.
706
707 Each of the 64 bits of the attribute bitmask can be set for each
708 audio file individually. Hence up to 64 different attributes may be
709 defined. For example, "pop", "rock", "blues", "jazz", "instrumental",
710 "german_lyrics", "speech", whatever. You are free to choose as
711 many attributes as you like and there are no naming restrictions
712 for attributes.
713
714 A new attribute "test" is created by
715
716 para_client addatt test
717 and
718 para_client lsatt
719
720 lists all available attributes. You can set the "test" attribute for
721 an audio file by executing
722
723 para_client setatt test+ /path/to/the/audio/file
724
725 Similarly, the "test" bit can be removed from an audio file with
726
727 para_client setatt test- /path/to/the/audio/file
728
729 Instead of a path you may use a shell wildcard pattern. The attribute
730 is applied to all audio files matching that pattern:
731
732 para_client setatt test+ '/test/directory/*'
733
734 The command
735
736 para_client -- ls -lv
737
738 gives you a verbose listing of your audio files also showing which
739 attributes are set.
740
741 In case you wonder why the double-dash in the above command is needed:
742 It tells para_client to not interpret the options after the dashes. If
743 you find this annoying, just say
744
745 alias para='para_client --'
746
747 and be happy. In what follows we shall use this alias.
748
749 The "test" attribute can be dropped from the database with
750
751 para rmatt test
752
753 Read the output of
754
755 para help ls
756 para help setatt
757
758 for more information and a complete list of command line options to
759 these commands.
760
761 *Blob tables*
762
763 The image, lyrics, moods and playlists tables are all blob tables.
764 Blob tables consist of three columns each: The identifier which is
765 a positive non-negative number that is auto-incremented, the name
766 (an arbitrary string) and the content (the blob).
767
768 All blob tables support the same set of actions: cat, ls, mv, rm
769 and add. Of course, _add_ is used for adding new blobs to the table
770 while the other actions have the same meaning as the corresponding
771 Unix commands. The paraslash commands to perform these actions are
772 constructed as the concatenation of the table name and the action. For
773 example addimg, catimg, lsimg, mvimg, rmimg are the commands that
774 manipulate or query the image table.
775
776 The add variant of these commands is special as these commands read
777 the blob contents from stdin. To add an image to the image table the
778 command
779
780 para addimg image_name < file.jpg
781
782 can be used.
783
784 Note that the images and lyrics are not interpreted at all, and also
785 the playlist and the mood blobs are only investigated when the mood
786 or playlist is activated by using the select command.
787
788 *The score table*
789
790 Unlike all other tables the contents of the score table remain in
791 memory and are never stored on disk. The score table contains two
792 columns: The SHA1 hash value (of an audio file) and its current
793 score.
794
795 However, only those files which are admissible for the current mood
796 or playlist are contained in the score table. The audio file selector
797 always chooses the row with the highest score as the file to stream
798 next. While doing so, it computes the new score and updates the
799 last_played and the num_played fields in the audio file table.
800
801 The score table is recomputed by the select command which loads a
802 new mood or playlist.
803
804 Playlists and moods
805 ~~~~~~~~~~~~~~~~~~~
806
807 Playlists and moods offer two different ways of specifying the set of
808 admissible files. A playlist in itself describes a set of admissible
809 files. A mood, in contrast, describes the set of admissible files in
810 terms of attributes and other type of information available in the
811 audio file table. As an example, a mood can define a filename pattern,
812 which is then matched against the names of audio files in the table.
813
814 Selecting a mood or playlist means the generation of a ranking
815 (a score table) for the set of admissible files. Audio files are
816 then selected on a highest-score-first basis. The score table is
817 recomputed at the moment the mood or playlist is selected.
818
819 *Playlists*
820
821 Playlists are accommodated in the playlist table of the afs database,
822 using the aforementioned blob format for tables. A new filelist is
823 created using the addpl command, by specifying the full (absolute)
824 paths of all desired audio files, separated by newlines. For example
825
826 find /my/mp3/dir -name "*.mp3" | para addpl my_playlist
827
828 If _my_playlist_ already exists it is overwritten. To activate the
829 new playlist, execute
830
831 para select p/my_playlist
832
833 The audio file selector will assign scores to each entry of the list,
834 in descending order so that files will be selected in order. If a
835 file could not be opened for streaming, its entry is removed from
836 the score table (but not from the playlist).
837
838 *Moods*
839
840 A mood consists of a unique name and its *mood definition*, which is
841 a set of *mood lines* containing expressions in terms of attributes
842 and other data contained in the database.
843
844 At any time, at most one mood can be *active* which means that
845 para_server is going to select only files from that subset of
846 admissible files.
847
848 So in order to create a mood definition one has to write a set of
849 mood lines. Mood lines come in three flavours: Accept lines, deny
850 lines and score lines.
851
852 The general syntax of the three types of mood lines is
853
854
855 accept [with score <score>] [if] [not] <mood_method> [options]
856 deny [with score <score>] [if] [not] <mood_method> [options]
857 score <score> [if] [not] <mood_method> [options]
858
859
860 Here <score> is either an integer or the string "random" which assigns
861 a random score to all matching files. The score value changes the
862 order in which admissible files are going to be selected, but is of
863 minor importance for this introduction.
864
865 So we concentrate on the first two forms, i.e. accept and deny
866 lines. As usual, everything in square brackets is optional, i.e.
867 accept/deny lines take the following form when ignoring scores:
868
869 accept [if] [not] <mood_method> [options]
870
871 and analogously for the deny case. The "if" keyword is only syntactic
872 sugar and has no function. The "not" keyword just inverts the result,
873 so the essence of a mood line is the mood method part and the options
874 following thereafter.
875
876 A *mood method* is realized as a function which takes an audio file
877 and computes a number from the data contained in the database.
878 If this number is non-negative, we say the file *matches* the mood
879 method. The file matches the full mood line if it either
880
881 - matches the mood method and the "not" keyword is not given,
882 or
883 - does not match the mood method, but the "not" keyword is given.
884
885 The set of admissible files for the whole mood is now defined as those
886 files which match at least one accept mood line, but no deny mood line.
887 More formally, an audio file F is admissible if and only if
888
889 (F ~ AL1 or F ~ AL2...) and not (F ~ DL1 or F ~ DN2 ...)
890
891 where AL1, AL2... are the accept lines, DL1, DL2... are the deny
892 lines and "~" means "matches".
893
894 The cases where no mood lines of accept/deny type are defined need
895 special treatment:
896
897 - Neither accept nor deny lines: This treats all files as
898 admissible (in fact, that is the definition of the dummy mood
899 which is activated automatically if no moods are available).
900
901 - Only accept lines: A file is admissible iff it matches at
902 least one accept line:
903
904 F ~ AL1 or F ~ AL2 or ...
905
906 - Only deny lines: A file is admissible iff it matches no
907 deny line:
908
909 not (F ~ DL1 or F ~ DN2 ...)
910
911
912
913 *List of mood_methods*
914
915 no_attributes_set
916
917 Takes no arguments and matches an audio file if and only if no
918 attributes are set.
919
920 is_set <attribute_name>
921
922 Takes the name of an attribute and matches iff that attribute is set.
923
924 path_matches <pattern>
925
926 Takes a filename pattern and matches iff the path of the audio file
927 matches the pattern.
928
929 artist_matches <pattern>
930 album_matches <pattern>
931 title_matches <pattern>
932 comment_matches <pattern>
933
934 Takes an extended regular expression and matches iff the text of the
935 corresponding tag of the audio file matches the pattern. If the tag
936 is not set, the empty string is matched against the pattern.
937
938 year ~ <num>
939 bitrate ~ <num>
940 frequency ~ <num>
941 channels ~ <num>
942 num_played ~ <num>
943
944 Takes a comparator ~ of the set {<, =, <=, >, >=, !=} and a number
945 <num>. Matches an audio file iff the condition <val> ~ <num> is
946 satisfied where val is the corresponding value of the audio file
947 (value of the year tag, bitrate in kbit/s, frequency in Hz, channel
948 count, play count).
949
950 The year tag is special as its value is undefined if the audio file
951 has no year tag or the content of the year tag is not a number. Such
952 audio files never match. Another difference is the special treatment
953 if the year tag is a two-digit number. In this case either 1900 or
954 2000 are added to the tag value depending on whether the number is
955 greater than 2000 plus the current year.
956
957
958 *Mood usage*
959
960 To create a new mood called "my_mood", write its definition into
961 some temporary file, say "tmpfile", and add it to the mood table
962 by executing
963
964 para addmood my_mood < tmpfile
965
966 If the mood definition is really short, you may just pipe it to the
967 client instead of using temporary files. Like this:
968
969 echo "$MOOD_DEFINITION" | para addmood my_mood
970
971 There is no need to keep the temporary file since you can always use
972 the catmood command to get it back:
973
974 para catmood my_mood
975
976 A mood can be activated by executing
977
978 para select m/my_mood
979
980 Once active, the list of admissible files is shown by the ls command
981 if the "-a" switch is given:
982
983 para ls -a
984
985
986 *Example mood definition*
987
988 Suppose you have defined attributes "punk" and "rock" and want to define
989 a mood containing only Punk-Rock songs. That is, an audio file should be
990 admissible if and only if both attributes are set. Since
991
992 punk and rock
993
994 is obviously the same as
995
996 not (not punk or not rock)
997
998 (de Morgan's rule), a mood definition that selects only Punk-Rock
999 songs is
1000
1001 deny if not is_set punk
1002 deny if not is_set rock
1003
1004
1005
1006 File renames and content changes
1007 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1008
1009 Since the audio file selector knows the SHA1 of each audio file that
1010 has been added to the afs database, it recognizes if the content of
1011 a file has changed, e.g. because an ID3 tag was added or modified.
1012 Also, if a file has been renamed or moved to a different location,
1013 afs will detect that an entry with the same hash value already exists
1014 in the audio file table.
1015
1016 In both cases it is enough to just re-add the new file. In the
1017 first case (file content changed), the audio table is updated, while
1018 metadata such as the num_played and last_played fields, as well as
1019 the attributes, remain unchanged. In the other case, when the file
1020 is moved or renamed, only the path information is updated, all other
1021 data remains as before.
1022
1023 It is possible to change the behaviour of the add command by using the
1024 "-l" (lazy add) or the "-f" (force add) option.
1025
1026 Troubleshooting
1027 ~~~~~~~~~~~~~~~
1028
1029 Use the debug loglevel (option -l debug for most commands) to show
1030 debugging info. Almost all paraslash executables have a brief online
1031 help which is displayed by using the -h switch. The --detailed-help
1032 option prints the full help text.
1033
1034 If para_server crashed or was killed by SIGKILL (signal 9), it
1035 may refuse to start again because of "dirty osl tables". In this
1036 case you'll have to run the oslfsck program of libosl to fix your
1037 database. It might be necessary to use --force (even if your name
1038 isn't Luke). However, make sure para_server isn't running before
1039 executing oslfsck --force.
1040
1041 If you don't mind to recreate your database you can start
1042 from scratch by removing the entire database directory, i.e.
1043
1044 rm -rf ~/.paraslash/afs_database-0.4
1045
1046 Be aware that this removes all attribute definitions, all playlists
1047 and all mood definitions and requires to re-initialize the tables.
1048
1049 Although oslfsck fixes inconsistencies in database tables it doesn't
1050 care about the table contents. To check for invalid table contents, use
1051
1052 para_client check
1053
1054 This prints out references to missing audio files as well as invalid
1055 playlists and mood definitions.
1056
1057 ---------------------------------------
1058 Audio formats and audio format handlers
1059 ---------------------------------------
1060
1061 Audio formats
1062 ~~~~~~~~~~~~~
1063
1064 The following audio formats are supported by paraslash:
1065
1066 *MP3*
1067
1068 Mp3, MPEG-1 Audio Layer 3, is a common audio format for audio storage,
1069 designed as part of its MPEG-1 standard. An MP3 file is made up of
1070 multiple MP3 frames, which consist of a header and a data block. The
1071 size of an MP3 frame depends on the bit rate and on the number
1072 of channels. For a typical CD-audio file (sample rate of 44.1 kHz
1073 stereo), encoded with a bit rate of 128 kbit, an MP3 frame is about
1074 400 bytes large.
1075
1076 *OGG/Vorbis*
1077
1078 OGG is a standardized audio container format, while Vorbis is an
1079 open source codec for lossy audio compression. Since Vorbis is most
1080 commonly made available via the OGG container format, it is often
1081 referred to as OGG/Vorbis. The OGG container format divides data into
1082 chunks called OGG pages. A typical OGG page is about 4KB large. The
1083 Vorbis codec creates variable-bitrate (VBR) data, where the bitrate
1084 may vary considerably.
1085
1086 *OGG/Speex*
1087
1088 Speex is an open-source speech codec that is based on CELP (Code
1089 Excited Linear Prediction) coding. It is designed for voice
1090 over IP applications, has modest complexity and a small memory
1091 footprint. Wideband and narrowband (telephone quality) speech are
1092 supported. As for Vorbis audio, Speex bit-streams are often stored
1093 in OGG files.
1094
1095 *AAC*
1096
1097 Advanced Audio Coding (AAC) is a standardized, lossy compression
1098 and encoding scheme for digital audio which is the default audio
1099 format for Apple's iPhone, iPod, iTunes. Usually MPEG-4 is used as
1100 the container format and audio files encoded with AAC have the .m4a
1101 extension. A typical AAC frame is about 700 bytes large.
1102
1103 *WMA*
1104
1105 Windows Media Audio (WMA) is an audio data compression technology
1106 developed by Microsoft. A WMA file is usually encapsulated in the
1107 Advanced Systems Format (ASF) container format, which also specifies
1108 how meta data about the file is to be encoded. The bit stream of WMA
1109 is composed of superframes, each containing one or more frames of
1110 2048 samples. For 16 bit stereo a WMA superframe is about 8K large.
1111
1112 Meta data
1113 ~~~~~~~~~
1114
1115 Unfortunately, each audio format has its own conventions how meta
1116 data is added as tags to the audio file.
1117
1118 For MP3 files, ID3, version 1 and 2 are widely used. ID3 version 1
1119 is rather simple but also very limited as it supports only artist,
1120 title, album, year and comment tags. Each of these can only be at most
1121 32 characters long. ID3, version 2 is much more flexible but requires
1122 a separate library being installed for paraslash to support it.
1123
1124 Ogg vorbis files contain meta data as Vorbis comments, which are
1125 typically implemented as strings of the form "[TAG]=[VALUE]". Unlike
1126 ID3 version 1 tags, one may use whichever tags are appropriate for
1127 the content.
1128
1129 AAC files usually use the MPEG-4 container format for storing meta
1130 data while WMA files wrap meta data as special objects within the
1131 ASF container format.
1132
1133 paraslash only tracks the most common tags that are supported by
1134 all tag variants: artist, title, year, album, comment. When a file
1135 is added to the AFS database, the meta data of the file is extracted
1136 and stored in the audio file table.
1137
1138 Chunks and chunk tables
1139 ~~~~~~~~~~~~~~~~~~~~~~~
1140
1141 paraslash uses the word "chunk" as common term for the building blocks
1142 of an audio file. For MP3 files, a chunk is the same as an MP3 frame,
1143 while for OGG files a chunk is an OGG page, etc. Therefore the chunk
1144 size varies considerably between audio formats, from a few hundred
1145 bytes (MP3) up to 8K (WMA).
1146
1147 The chunk table contains the offsets within the audio file that
1148 correspond to the chunk boundaries of the file. Like the meta data,
1149 the chunk table is computed and stored in the database whenever an
1150 audio file is added.
1151
1152 The paraslash senders (see below) always send complete chunks. The
1153 granularity for seeking is therefore determined by the chunk size.
1154
1155 Audio format handlers
1156 ~~~~~~~~~~~~~~~~~~~~~
1157
1158 For each audio format paraslash contains an audio format handler whose
1159 first task is to tell whether a given file is a valid audio file of
1160 this type. If so, the audio file handler extracts some technical data
1161 (duration, sampling rate, number of channels etc.), computes the
1162 chunk table and reads the meta data.
1163
1164 The audio format handler code is linked into para_server and executed
1165 via the _add_ command. The same code is also available as a stand-alone
1166 tool, para_afh, which can be used to print the technical data, the
1167 chunk table and the meta data of a file. Furthermore, one can use
1168 para_afh to cut an audio file, i.e. to select some of its chunks to
1169 produce a new file containing only these chunks.
1170
1171 ----------
1172 Networking
1173 ----------
1174
1175 Paraslash uses different network connections for control and data.
1176 para_client communicates with para_server over a dedicated TCP control
1177 connection. To transport audio data, separate data connections are
1178 used. For these data connections, a variety of transports (UDP, DCCP,
1179 HTTP) can be chosen.
1180
1181 The chapter starts with the REFERENCE(The paraslash control
1182 service, control service), followed by a section on the various
1183 REFERENCE(Streaming protocols, streaming protocols) in which the data
1184 connections are described. The way audio file headers are embedded into
1185 the stream is discussed REFERENCE(Streams with headers and headerless
1186 streams, briefly) before the REFERENCE(Networking examples, example
1187 section) which illustrates typical commands for real-life scenarios.
1188
1189 Both IPv4 and IPv6 are supported.
1190
1191 The paraslash control service
1192 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1193
1194 para_server is controlled at runtime via the paraslash control
1195 connection. This connection is used for server commands (play, stop,
1196 ...) as well as for afs commands (ls, select, ...).
1197
1198 The server listens on a TCP port and accepts connections from clients
1199 that connect the open port. Each connection causes the server to fork
1200 off a client process which inherits the connection and deals with that
1201 client only. In this classical accept/fork approach the server process
1202 is unaffected if the child dies or goes crazy for whatever reason. In
1203 fact, the child process can not change address space of server process.
1204
1205 The section on REFERENCE(Client-server authentication, client-server
1206 authentication) above described the early connection establishment
1207 from the crypto point of view. Here it is described what happens
1208 after the connection (including crypto setup) has been established.
1209 There are four processes involved during command dispatch as sketched
1210 in the following diagram.
1211
1212 <<
1213 <pre>
1214 server_host client_host
1215 ~~~~~~~~~~~ ~~~~~~~~~~~
1216
1217 +-----------+ connect +-----------+
1218 |para_server|<------------------------------ |para_client|
1219 +-----------+ +-----------+
1220 | ^
1221 | fork +---+ |
1222 +----------> |AFS| |
1223 | +---+ |
1224 | ^ |
1225 | | |
1226 | | connect (cookie) |
1227 | | |
1228 | | |
1229 | fork +-----+ inherited connection |
1230 +---------->|child|<--------------------------+
1231 +-----+
1232 </pre>
1233 >>
1234
1235 Note that the child process is not a child of the afs process,
1236 so communication of these two processes has to happen via local
1237 sockets. In order to avoid abuse of the local socket by unrelated
1238 processes, a magic cookie is created once at server startup time just
1239 before the server process forks off the AFS process. This cookie is
1240 known to the server, AFS and the child, but not to unrelated processes.
1241
1242 There are two different kinds of commands: First there are commands
1243 that cause the server to respond with some answer such as the list
1244 of all audio files. All but the addblob commands (addimg, addlyr,
1245 addpl, addmood) are of this kind. The addblob commands add contents
1246 to the database, so they need to transfer data the other way round,
1247 from the client to the server.
1248
1249 There is no knowledge about the server commands built into para_client,
1250 so it does not know about addblob commands. Instead, it inspects the
1251 first data package sent by the server for a magic string. If this
1252 string was found, it sends STDIN to the server, otherwise it dumps
1253 data from the server to STDOUT.
1254
1255 Streaming protocols
1256 ~~~~~~~~~~~~~~~~~~~
1257
1258 A network (audio) stream usually consists of one streaming source,
1259 the _sender_, and one or more _receivers_ which read data over the
1260 network from the streaming source.
1261
1262 Senders are thus part of para_server while receivers are part of
1263 para_audiod. Moreover, there is the stand-alone tool para_recv which
1264 can be used to manually download a stream, either from para_server
1265 or from a web-based audio streaming service.
1266
1267 The following three streaming protocols are supported by paraslash:
1268
1269 - HTTP. Recommended for public streams that can be played by
1270 any player like mpg123, xmms, itunes, winamp, etc. The HTTP
1271 sender is supported on all operating systems and all platforms.
1272
1273 - DCCP. Recommended for LAN streaming. DCCP is currently
1274 available only for Linux.
1275
1276 - UDP. Recommended for multicast LAN streaming.
1277
1278 See the Appendix on REFERENCE(Network protocols, network protocols)
1279 for brief descriptions of the various protocols relevant for network
1280 audio streaming with paraslash.
1281
1282 It is possible to activate more than one sender simultaneously.
1283 Senders can be controlled at run time and via config file and command
1284 line options.
1285
1286 Note that audio connections are _not_ encrypted. Transport or Internet
1287 layer encryption should be used if encrypted data connections are
1288 needed.
1289
1290 Since DCCP and TCP are both connection-oriented protocols, connection
1291 establishment/teardown and access control are very similar between
1292 these two streaming protocols. UDP is the most lightweight option,
1293 since in contrast to TCP/DCCP it is connectionless. It is also the
1294 only protocol supporting IP multicast.
1295
1296 The HTTP and the DCCP sender listen on a (TCP/DCCP) port waiting for
1297 clients to connect and establish a connection via some protocol-defined
1298 handshake mechanism. Both senders maintain two linked lists each:
1299 The list of all clients which are currently connected, and the list
1300 of access control entries which determines who is allowed to connect.
1301 IP-based access control may be configured through config file and
1302 command line options and via the "allow" and "deny" sender subcommands.
1303
1304 Upon receiving a GET request from the client, the HTTP sender sends
1305 back a status line and a message. The body of this message is the
1306 audio stream. This is common practice and is supported by many popular
1307 clients which can thus be used to play a stream offered by para_server.
1308 For DCCP things are a bit simpler: No messages are exchanged between
1309 the receiver and sender. The client simply connects and the sender
1310 starts to stream.
1311
1312 DCCP is an experimental protocol which offers a number of new features
1313 not available for TCP. Both ends can negotiate these features using
1314 a built-in negotiation mechanism. In contrast to TCP/HTTP, DCCP is
1315 datagram-based (no retransmissions) and thus should not be used over
1316 lossy media (e.g. WiFi networks). One useful feature offered by DCCP
1317 is access to a variety of different congestion-control mechanisms
1318 called CCIDs. Two different CCIDs are available per default on Linux:
1319
1320
1321 - _CCID 2_. A Congestion Control mechanism similar to that
1322 of TCP. The sender maintains a congestion window and halves
1323 this window in response to congestion.
1324
1325
1326 - _CCID-3_. Designed to be fair when competing for bandwidth.
1327 It has lower variation of throughput over time compared with
1328 TCP, which makes it suitable for streaming media.
1329
1330 Unlike the HTTP and DCCP senders, the UDP sender maintains only a
1331 single list, the _target list_. This list describes the set of clients
1332 to which the stream is sent. There is no list for access control and
1333 no "allow" and "deny" commands for the UDP sender. Instead, the "add"
1334 and "delete" commands can be used to modify the target list.
1335
1336 Since both UDP and DCCP offer an unreliable datagram-based transport,
1337 additional measures are necessary to guard against disruptions over
1338 networks that are lossy or which may be subject to interference (as
1339 is for instance the case with WiFi). Paraslash uses FEC (Forward
1340 Error Correction) to guard against packet losses and reordering. The
1341 stream is FEC-encoded before it is sent through the UDP socket and
1342 must be decoded accordingly on the receiver side.
1343
1344 The packet size and the amount of redundancy introduced by FEC can
1345 be configured via the FEC parameters which are dictated by server
1346 and may also be configured through the "sender" command. The FEC
1347 parameters are encoded in the header of each network packet, so no
1348 configuration is necessary on the receiver side. See the section on
1349 REFERENCE(Forward error correction, FEC) below.
1350
1351 Streams with headers and headerless streams
1352 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1353
1354 For OGG/Vorbis, OGG/Speex and wma streams, some of the information
1355 needed to decode the stream is only contained in the audio file
1356 header of the container format but not in each data chunk. Clients
1357 must be able to obtain this information in case streaming starts in
1358 the middle of the file or if para_audiod is started while para_server
1359 is already sending a stream.
1360
1361 This is accomplished in different ways, depending on the streaming
1362 protocol. For connection-oriented streams (HTTP, DCCP) the audio file
1363 header is sent prior to audio file data. This technique however does
1364 not work for the connectionless UDP transport. Hence the audio file
1365 header is periodically being embedded into the UDP audio data stream.
1366 By default, the header is resent after five seconds. The receiver has
1367 to wait until the next header arrives before it can start decoding
1368 the stream.
1369
1370 Examples
1371 ~~~~~~~~
1372
1373 The sender command of para_server allows to (de-)activate senders
1374 and to change the access permissions senders at runtime. The "si"
1375 (server info) command is used to list the streaming options of the
1376 currently running server as well as the various sender access lists.
1377
1378 -> Show client/target/access lists:
1379
1380 para_client si
1381
1382 -> Obtain general help for the sender command:
1383
1384 para_client help sender
1385
1386 -> Get help for a specific sender (contains further examples):
1387
1388 s=http # or dccp or udp
1389 para_client sender $s help
1390
1391 By default para_server activates both the HTTP and th DCCP sender on
1392 startup. This can be changed via command line options or para_server's
1393 config file.
1394
1395 -> List config file options for senders:
1396
1397 para_server -h
1398
1399 All senders share the "on" and "off" commands, so senders may be
1400 activated and deactivated independently of each other.
1401
1402 -> Switch off the http sender:
1403
1404 para_client sender http off
1405
1406 -> Receive a DCCP stream using CCID2 and write the output into a file:
1407
1408 host=foo.org; ccid=2; filename=bar
1409 para_recv --receiver "dccp --host $host --ccid $ccid" > $filename
1410
1411 Note the quotes around the arguments for the dccp receiver. Each
1412 receiver has its own set of command line options and its own command
1413 line parser, so arguments for the dccp receiver must be protected
1414 from being interpreted by para_recv.
1415
1416 -> Start UDP multicast, using the default multicast address:
1417
1418 para_client sender udp add 224.0.1.38
1419
1420 -> Receive FEC-encoded multicast stream and write the output into a file:
1421
1422 filename=foo
1423 para_recv -r udp > $filename
1424
1425 -> Add an UDP unicast for a client to the target list of the UDP sender:
1426
1427 t=client.foo.org
1428 para_client sender udp add $t
1429
1430 -> Receive this (FEC-encoded) unicast stream:
1431
1432 filename=foo
1433 para_recv -r 'udp -i 0.0.0.0' > $filename
1434
1435 -> Create a minimal config for para_audiod for HTTP streams:
1436
1437 c=$HOME/.paraslash/audiod.conf.min; s=server.foo.com
1438 formats="mp3 ogg aac wma" # remove what you do not have
1439 for f in $formats; do echo receiver \"$f:http -i $s\"; done > $c
1440 para_audiod --config $c
1441
1442 -------
1443 Filters
1444 -------
1445
1446 A paraslash filter is a module which transforms an input stream into
1447 an output stream. Filters are included in the para_audiod executable
1448 and in the stand-alone tool para_filter which usually contains the
1449 same modules.
1450
1451 While para_filter reads its input stream from STDIN and writes
1452 the output to STDOUT, the filter modules of para_audiod are always
1453 connected to a receiver which produces the input stream and a writer
1454 which absorbs the output stream.
1455
1456 Some filters depend on a specific library being installed and are
1457 not compiled in if this library was not found at compile time. To
1458 see the list of supported filters, run para_filter and para_audiod
1459 with the --help option. The output looks similar to the following:
1460
1461 Available filters:
1462 compress wav amp fecdec wmadec prebuffer oggdec aacdec mp3dec
1463
1464 Out of these filter modules, a chain of filters can be constructed,
1465 much in the way Unix pipes can be chained, and analogous to the use
1466 of modules in gstreamer: The output of the first filter becomes the
1467 input of the second filter. There is no limitation on the number of
1468 filters and the same filter may occur more than once.
1469
1470 Like receivers, each filter has its own command line options which
1471 must be quoted to protect them from the command line options of
1472 the driving application (para_audiod or para_filter). Example:
1473
1474 para_filter -f 'mp3dec --ignore-crc' -f 'compress --damp 1'
1475
1476 For para_audiod, each audio format has its own set of filters. The
1477 name of the audio format for which the filter should be applied is
1478 used as the prefix for the filter option. Example:
1479
1480 para_audiod -f 'mp3:prebuffer --duration 300'
1481
1482 Decoders
1483 ~~~~~~~~
1484
1485 For each supported audio format there is a corresponding filter
1486 which decodes audio data in this format to 16 bit PCM data which
1487 can be directly sent to the sound device or any other software that
1488 operates on undecoded PCM data (visualizers, equalizers etc.). Such
1489 filters are called _decoders_ in general, and xxxdec is the name of
1490 the paraslash decoder for the audio format xxx. For example, the mp3
1491 decoder filter is called mp3dec.
1492
1493 Note that the output of the decoder is about 10 times larger than
1494 its input. This means that filters that operate on the decoded audio
1495 stream have to deal with much more data than filters that transform
1496 the audio stream before it is fed to the decoder.
1497
1498 Paraslash relies on external libraries for most decoders, so these
1499 libraries must be installed for the decoder to be included in the
1500 para_filter and para_audiod executables. The oggdec filter depends
1501 on the libogg and libvorbis libraries for example.
1502
1503 Forward error correction
1504 ~~~~~~~~~~~~~~~~~~~~~~~~
1505
1506 As already mentioned REFERENCE(Streaming protocols, earlier),
1507 paraslash uses forward error correction (FEC) for the unreliable
1508 UDP transport. FEC is a technique which was invented already in
1509 1960 by Reed and Solomon and which is widely used for the parity
1510 calculations of storage devices (RAID arrays). It is based on the
1511 algebraic concept of finite fields, today called Galois fields, in
1512 honour of the mathematician Galois (1811-1832). The FEC implementation
1513 of paraslash is based on code by Luigi Rizzo.
1514
1515 Although the details require a sound knowledge of the underlying
1516 mathematics, the basic idea is not hard to understand: For positive
1517 integers k and n with k < n it is possible to compute for any k given
1518 data bytes d_1, ..., d_k the corresponding r := n -k parity bytes p_1,
1519 ..., p_r such that all data bytes can be reconstructed from *any*
1520 k bytes of the set
1521
1522 {d_1, ..., d_k, p_1, ..., p_r}.
1523
1524 FEC-encoding for unreliable network transports boils down to slicing
1525 the audio stream into groups of k suitably sized pieces called _slices_
1526 and computing the r corresponding parity slices. This step is performed
1527 in para_server which then sends both the data and the parity slices
1528 over the unreliable network connection. If the client was able
1529 to receive at least k of the n = k + r slices, it can reconstruct
1530 (FEC-decode) the original audio stream.
1531
1532 From these observations it is clear that there are three different
1533 FEC parameters: The slice size, the number of data slices k, and the
1534 total number of slices n. It is crucial to choose the slice size
1535 such that no fragmentation of network packets takes place because
1536 FEC only guards against losses and reodering but fails if slices are
1537 received partially.
1538
1539 FEC decoding in paralash is performed through the fecdec filter which
1540 usually is the first filter (there can be other filters before fecdec
1541 if these do not alter the audio stream).
1542
1543
1544 Volume adjustment (amp and compress)
1545 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1546
1547 The amp and the compress filter both adjust the volume of the audio
1548 stream. These filters operate on uncompressed audio samples. Hence
1549 they are usually placed directly after the decoding filter. Each
1550 sample is multiplied with a scaling factor (>= 1) which makes amp
1551 and compress quite expensive in terms of computing power.
1552
1553 *amp*
1554
1555 The amp filter amplifies the audio stream by a fixed scaling factor
1556 that must be known in advance. For para_audiod this factor is derived
1557 from the amplification field of the audio file's entry in the audio
1558 file table while para_filter uses the value given at the command line.
1559
1560 The optimal scaling factor F for an audio file is the largest real
1561 number F >= 1 such that after multiplication with F all samples still
1562 fit into the sample interval [-32768, 32767]. One can use para_filter
1563 in combination with the sox utility to compute F:
1564
1565 para_filter -f mp3dec -f wav < file.mp3 | sox -t wav - -e stat -v
1566
1567 The amplification value V which is stored in the audio file table,
1568 however, is an integer between 0 and 255 which is connected to F
1569 through the formula
1570
1571 V = (F - 1) * 64.
1572
1573 To store V in the audio file table, the command
1574
1575 para_client -- touch -a=V file.mp3
1576
1577 is used. The reader is encouraged to write a script that performs
1578 these computations :)
1579
1580 *compress*
1581
1582 Unlike the amplification filter, the compress filter adjusts the volume
1583 of the audio stream dynamically without prior knowledge about the peak
1584 value. It maintains the maximal volume of the last n samples of the
1585 audio stream and computes a suitable amplification factor based on that
1586 value and the various configuration options. It tries to chose this
1587 factor such that the adjusted volume meets the desired target level.
1588
1589 Note that it makes sense to combine amp and compress.
1590
1591 Misc filters (wav and prebuffer)
1592 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1593
1594 These filters are rather simple and do not modify the audio stream at
1595 all. The wav filter is only useful with para_filter and in connection
1596 with a decoder. It asks the decoder for the number of channels and the
1597 sample rate of the stream and adds a Microsoft wave header containing
1598 this information at the beginning. This allows to write wav files
1599 rather than raw PCM files (which do not contain any information about
1600 the number of channels and the sample rate).
1601
1602 The prebuffer filter simply delays the output until the given time has
1603 passed (starting from the time the first byte was available in its
1604 input queue) or until the given amount of data has accumulated. It
1605 is mainly useful for para_audiod if the standard parameters result
1606 in buffer underruns.
1607
1608 Both filters require almost no additional computing time, even when
1609 operating on uncompressed audio streams, since data buffers are simply
1610 "pushed down" rather than copied.
1611
1612 Examples
1613 ~~~~~~~~
1614
1615 -> Decode an mp3 file to wav format:
1616
1617 para_filter -f mp3dec -f wav < file.mp3 > file.wav
1618
1619 -> Amplify a raw audio file by a factor of 1.5:
1620
1621 para_filter -f amp --amp 32 < foo.raw > bar.raw
1622
1623 ------
1624 Output
1625 ------
1626
1627 Once an audio stream has been received and decoded to PCM format,
1628 it can be sent to a sound device for playback. This part is performed
1629 by paraslash _writers_ which are described in this chapter.
1630
1631 Writers
1632 ~~~~~~~
1633
1634 A paraslash writer acts as a data sink that consumes but does not
1635 produce audio data. Paraslash writers operate on the client side and
1636 are contained in para_audiod and in the stand-alone tool para_write.
1637
1638 The para_write program reads uncompressed audio data from STDIN. If
1639 this data starts with a wav header, sample rate, sample format and
1640 channel count are read from the header. Otherwise CD audio (44.1KHz
1641 16 bit little endian, stereo) is assumed but this can be overridden
1642 by command line options. para_audiod, on the other hand, obtains
1643 the sample rate and the number of channels from the decoder.
1644
1645 Like receivers and filters, each writer has an individual set of
1646 command line options, and for para_audiod writers can be configured
1647 per audio format separately. It is possible to activate more than
1648 one writer for the same stream simultaneously.
1649
1650 OS-dependent APIs
1651 ~~~~~~~~~~~~~~~~~
1652
1653 Unfortunately, the various flavours of Unix on which paraslash
1654 runs on have different APIs for opening a sound device and starting
1655 playback. Hence for each such API there is a paraslash writer that
1656 can play the audio stream via this API.
1657
1658 *ALSA*. The _Advanced Linux Sound Architecture_ is only available on
1659 Linux systems. Although there are several mid-layer APIs in use by
1660 the various Linux distributions (ESD, Jack, PulseAudio), paraslash
1661 currently supports only the low-level ALSA API which is not supposed
1662 to be change. ALSA is very feature-rich, in particular it supports
1663 software mixing via its DMIX plugin. ALSA is the default writer on
1664 Linux systems.
1665
1666 *OSS*. The _Open Sound System_ is the only API on *BSD Unixes and
1667 is also available on Linux systems, usually provided by ALSA as an
1668 emulation for backwards compatibility. This API is rather simple but
1669 also limited. For example only one application can open the device
1670 at any time. The OSS writer is activated by default on BSD Systems.
1671
1672 *OSX*. Mac OS X has yet another API called CoreAudio. The OSX writer
1673 for this API is only compiled in on such systems and is of course
1674 the default there.
1675
1676 *FILE*. The file writer allows to capture the audio stream and
1677 write the PCM data to a file on the file system rather than playing
1678 it through a sound device. It is supported on all platforms and is
1679 always compiled in.
1680
1681 Examples
1682 ~~~~~~~~
1683
1684 -> Use the OSS writer to play a wav file:
1685
1686 para_write --writer oss < file.wav
1687
1688 -> Enable ALSA software mixing for mp3 streams
1689
1690 para_audiod --writer 'mp3:alsa -d plug:swmix'
1691
1692
1693 ---
1694 Gui
1695 ---
1696
1697 para_gui executes an arbitrary command which is supposed to print
1698 status information to STDOUT. It then displays this information in
1699 a curses window. By default the command
1700
1701 para_audioc -- stat -p
1702
1703 is executed, but this can be customized via the --stat_cmd option. In
1704 particular it possible to use
1705
1706 para_client -- stat -p
1707
1708 to make para_gui work on systems on which para_audiod is not running.
1709
1710 Key bindings
1711 ~~~~~~~~~~~~
1712
1713 It is possible to bind keys to arbitrary commands via custom
1714 key-bindings. Besides the internal keys which can not be changed (help,
1715 quit, loglevel, version...), the following flavours of key-bindings
1716 are supported:
1717
1718 - external: Shutdown curses before launching the given command.
1719 Useful for starting other ncurses programs from within
1720 para_gui, e.g. aumix or dialog scripts. Or, use the mbox
1721 output format to write a mailbox containing one mail for each
1722 (admissible) file the audio file selector knows about. Then
1723 start mutt from within para_gui to browse your collection!
1724
1725 - display: Launch the command and display its stdout in
1726 para_gui's bottom window.
1727
1728 - para: Like display, but start "para_client <specified
1729 command>" instead of "<specified command>".
1730
1731 The general form of a key binding is
1732
1733 key_map k:m:c
1734
1735 which maps key k to command c using mode m. Mode may be x, d or p
1736 for external, display and paraslash commands, respectively.
1737
1738 Themes
1739 ~~~~~~
1740
1741 Currently there are only two themes for para_gui. It is easy, however,
1742 to add more themes. To create a new theme one has to define the
1743 position, color and geometry for for each status item that should be
1744 shown by this theme. See gui_theme.c for examples.
1745
1746 The "." and "," keys are used to switch between themes.
1747
1748 Examples
1749 ~~~~~~~~
1750
1751 -> Show server info:
1752
1753 key_map "i:p:si"
1754
1755 -> Jump to the middle of the current audio file by pressing F5:
1756
1757 key_map "<F5>:p:jmp 50"
1758
1759 -> vi-like bindings for jumping around:
1760
1761 key_map "l:p:ff 10"
1762 key_map "h:p:ff 10-"
1763 key_map "w:p:ff 60"
1764 key_map "b:p:ff 60-"
1765
1766 -> Print the current date and time:
1767
1768 key_map "D:d:date"
1769
1770 -> Call other curses programs:
1771
1772 key_map "U:x:aumix"
1773 key_map "!:x:/bin/bash"
1774 key_map "^E:x:/bin/sh -c 'vi ~/.paraslash/gui.conf'"
1775
1776 -----------
1777 Development
1778 -----------
1779
1780 Tools
1781 ~~~~~
1782
1783 In order to compile the sources from the git repository (rather than
1784 from tar balls) and for contributing non-trivial changes to the
1785 paraslash project, some additional tools should be installed on a
1786 developer machine.
1787
1788 http://git.or.cz/ (git). As described in more detail REFERENCE(Git
1789 branches, below), the git source code management tool is used for
1790 paraslash development. It is necessary for cloning the git repository
1791 and for getting updates.
1792
1793 ftp://ftp.gnu.org/pub/gnu/gengetopt/ (gengetopt). The C code for
1794 the command line parsers of all paraslash executables is generated
1795 by gengetopt. The generated C files are shipped in the tarballs but
1796 are not contained in the git repository.
1797
1798 ftp://ftp.gnu.org/pub/gnu/m4/ (m4). Some input files for gengetopt
1799 are generated from templates by the m4 macro processor.
1800
1801 ftp://ftp.gnu.org/pub/gnu/autoconf/ (autoconf) GNU autoconf creates
1802 the configure file which is shipped in the tarballs but has to be
1803 generated when compiling from git.
1804
1805 http://www.triptico.com/software/grutatxt.html (grutatxt). The
1806 HTML version of this manual and some of the paraslash web pages are
1807 generated by the grutatxt plain text to HTML converter. If changes
1808 are made to these text files the grutatxt package must be installed
1809 to regenerate the HTML files.
1810
1811 http://www.stack.nl/~dimitri/doxygen/ (doxygen). The documentation
1812 of paraslash's C sources uses the doxygen documentation system. The
1813 conventions for documenting the source code is described in the
1814 REFERENCE(Doxygen, Doxygen section).
1815
1816 ftp://ftp.gnu.org/pub/gnu/global (global). This is used to generate
1817 browsable HTML from the C sources. It is needed by doxygen.
1818
1819 Git branches
1820 ~~~~~~~~~~~~
1821
1822 Paraslash has been developed using the git source code management
1823 tool since 2006. Development is organized roughly in the same spirit
1824 as the git development itself, as described below.
1825
1826 The following text passage is based on "A note from the maintainer",
1827 written by Junio C Hamano, the maintainer of git.
1828
1829 There are four branches in the paraslash repository that track the
1830 source tree: "master", "maint", "next", and "pu".
1831
1832 The "master" branch is meant to contain what is well tested and
1833 ready to be used in a production setting. There could occasionally be
1834 minor breakages or brown paper bag bugs but they are not expected to
1835 be anything major, and more importantly quickly and easily fixable.
1836 Every now and then, a "feature release" is cut from the tip of this
1837 branch, named with three dotted decimal digits, like 0.4.2.
1838
1839 Whenever changes are about to be included that will eventually lead to
1840 a new major release (e.g. 0.5.0), a "maint" branch is forked off from
1841 "master" at that point. Obvious, safe and urgent fixes after the major
1842 release are applied to this branch and maintenance releases are cut
1843 from it. New features never go to this branch. This branch is also
1844 merged into "master" to propagate the fixes forward.
1845
1846 A trivial and safe enhancement goes directly on top of "master".
1847 New development does not usually happen on "master", however.
1848 Instead, a separate topic branch is forked from the tip of "master",
1849 and it first is tested in isolation; Usually there are a handful such
1850 topic branches that are running ahead of "master". The tip of these
1851 branches is not published in the public repository to keep the number
1852 of branches that downstream developers need to worry about low.
1853
1854 The quality of topic branches varies widely. Some of them start out as
1855 "good idea but obviously is broken in some areas" and then with some
1856 more work become "more or less done and can now be tested by wider
1857 audience". Luckily, most of them start out in the latter, better shape.
1858
1859 The "next" branch is to merge and test topic branches in the latter
1860 category. In general, this branch always contains the tip of "master".
1861 It might not be quite rock-solid production ready, but is expected to
1862 work more or less without major breakage. The maintainer usually uses
1863 the "next" version of paraslash for his own pleasure, so it cannot
1864 be _that_ broken. The "next" branch is where new and exciting things
1865 take place.
1866
1867 The two branches "master" and "maint" are never rewound, and "next"
1868 usually will not be either (this automatically means the topics that
1869 have been merged into "next" are usually not rebased, and you can find
1870 the tip of topic branches you are interested in from the output of
1871 "git log next"). You should be able to safely build on top of them.
1872
1873 However, at times "next" will be rebuilt from the tip of "master" to
1874 get rid of merge commits that will never be in "master. The commit
1875 that replaces "next" will usually have the identical tree, but it
1876 will have different ancestry from the tip of "master".
1877
1878 The "pu" (proposed updates) branch bundles the remainder of the
1879 topic branches. The "pu" branch, and topic branches that are only in
1880 "pu", are subject to rebasing in general. By the above definition
1881 of how "next" works, you can tell that this branch will contain quite
1882 experimental and obviously broken stuff.
1883
1884 When a topic that was in "pu" proves to be in testable shape, it
1885 graduates to "next". This is done with
1886
1887 git checkout next
1888 git merge that-topic-branch
1889
1890 Sometimes, an idea that looked promising turns out to be not so good
1891 and the topic can be dropped from "pu" in such a case.
1892
1893 A topic that is in "next" is expected to be polished to perfection
1894 before it is merged to "master". Similar to the above, this is
1895 done with
1896
1897 git checkout master
1898 git merge that-topic-branch
1899 git branch -d that-topic-branch
1900
1901 Note that being in "next" is not a guarantee to appear in the next
1902 release (being in "master" is such a guarantee, unless it is later
1903 found seriously broken and reverted), nor even in any future release.
1904
1905 Coding Style
1906 ~~~~~~~~~~~~
1907
1908 The preferred coding style for paraslash coincides more or less
1909 with the style of the Linux kernel. So rather than repeating what is
1910 written XREFERENCE(http://www.kernel.org/doc/Documentation/CodingStyle,
1911 there), here are the most important points.
1912
1913 - Burn the GNU coding standards.
1914 - Never use spaces for indentation.
1915 - Tabs are 8 characters, and thus indentations are also 8 characters.
1916 - Don't put multiple assignments on a single line.
1917 - Avoid tricky expressions.
1918 - Don't leave whitespace at the end of lines.
1919 - The limit on the length of lines is 80 columns.
1920 - Use K&R style for placing braces and spaces:
1921
1922 if (x is true) {
1923 we do y
1924 }
1925
1926 - Use a space after (most) keywords.
1927 - Do not add spaces around (inside) parenthesized expressions.
1928 - Use one space around (on each side of) most binary and ternary operators.
1929 - Do not use cute names like ThisVariableIsATemporaryCounter, call it tmp.
1930 - Mixed-case names are frowned upon.
1931 - Descriptive names for global variables are a must.
1932 - Avoid typedefs.
1933 - Functions should be short and sweet, and do just one thing.
1934 - The number of local variables shouldn't exceed 10.
1935 - Gotos are fine if they improve readability and reduce nesting.
1936 - Don't use C99-style "// ..." comments.
1937 - Names of macros defining constants and labels in enums are capitalized.
1938 - Enums are preferred when defining several related constants.
1939 - Always use the paraslash wrappers for allocating memory.
1940 - If the name of a function is an action or an imperative.
1941 command, the function should return an error-code integer
1942 (<0 means error, >=0 means success). If the name is a
1943 predicate, the function should return a "succeeded" boolean.
1944
1945
1946 Doxygen
1947 ~~~~~~~
1948
1949 Doxygen is a documentation system for various programming
1950 languages. The paraslash project uses Doxygen for generating the API
1951 reference on the web pages, but good source code documentation is
1952 also beneficial to people trying to understand the code structure
1953 and the interactions between the various source files.
1954
1955 It is more illustrative to look at the source code for examples than
1956 to describe the conventions for documenting the source in this manual,
1957 so we only describe which parts of the code need doxygen comments,
1958 but leave out details on documentation conventions.
1959
1960 As a rule, only the public part of the C source is documented with
1961 Doxygen. This includes structures, defines and enumerations in header
1962 files as well as public (non-static) C functions. These should be
1963 documented completely. For example each parameter and the return
1964 value of a public function should get a descriptive comment.
1965
1966 No doxygen comments are necessary for static functions and for
1967 structures and enumerations in C files (which are used only within
1968 this file). This does not mean, however, that those entities need
1969 no documentation at all. Instead, common sense should be applied to
1970 document what is not obvious from reading the code.
1971
1972 --------
1973 Appendix
1974 --------
1975
1976 Network protocols
1977 ~~~~~~~~~~~~~~~~~
1978
1979 *IP*. The _Internet Protocol_ is the primary networking protocol
1980 used for the Internet. All protocols described below use IP as the
1981 underlying layer. Both the prevalent IPv4 and the next-generation
1982 IPv6 variant are being deployed actively worldwide.
1983
1984 *Connection-oriented and connectionless protocols*. Connectionless
1985 protocols differ from connection-oriented ones in that state
1986 associated with the sending/receiving endpoints is treated
1987 implicitly. Connectionless protocols maintain no internal knowledge
1988 about the state of the connection. Hence they are not capable of
1989 reacting to state changes, such as sudden loss or congestion on the
1990 connection medium. Connection-oriented protocols, in contrast, make
1991 this knowledge explicit. The connection is established only after
1992 a bidirectional handshake which requires both endpoints to agree
1993 on the state of the connection, and may also involve negotiating
1994 specific parameters for the particular connection. Maintaining an
1995 up-to-date internal state of the connection also in general means
1996 that the sending endpoints perform congestion control, adapting to
1997 qualitative changes of the connection medium.
1998
1999 *Reliability*. In IP networking, packets can be lost, duplicated,
2000 or delivered out of order, and different network protocols handle
2001 these problems in different ways. We call a transport-layer protocol
2002 _reliable_, if it turns the unreliable IP delivery into an ordered,
2003 duplicate- and loss-free delivery of packets. Sequence numbers
2004 are used to discard duplicates and re-arrange packets delivered
2005 out-of-order. Retransmission is used to guarantee loss-free
2006 delivery. Unreliable protocols, in contrast, do not guarantee ordering
2007 or data integrity.
2008
2009 *Classification*. With these definitions the protocols which are used
2010 by paraslash for steaming audio data may be classified as follows.
2011
2012 - HTTP/TCP: connection-oriented, reliable,
2013 - UDP: connectionless, unreliable,
2014 - DCCP: connection-oriented, unreliable.
2015
2016 Below we give a short descriptions of these protocols.
2017
2018 *TCP*. The _Transmission Control Protocol_ provides reliable,
2019 ordered delivery of a stream and a classic window-based congestion
2020 control. In contrast to UDP and DCCP (see below), TCP does not have
2021 record-oriented or datagram-based syntax, i.e. it provides a stream
2022 which is unaware and independent of any record (packet) boundaries.
2023 TCP is used extensively by many application layers. Besides HTTP (the
2024 Hypertext Transfer Protocol), also FTP (the File Transfer protocol),
2025 SMTP (Simple Mail Transfer Protocol), SSH (Secure Shell) all sit on
2026 top of TCP.
2027
2028 *UDP*. The _User Datagram Protocol_ is the simplest transport-layer
2029 protocol, built as a thin layer directly on top of IP. For this reason,
2030 it offers the same best-effort service as IP itself, i.e. there is no
2031 detection of duplicate or reordered packets. Being a connectionless
2032 protocol, only minimal internal state about the connection is
2033 maintained, which means that there is no protection against packet
2034 loss or network congestion. Error checking and correction (if at all)
2035 are performed in the application.'
2036
2037 *DCCP*. The _Datagram Congestion Control Protocol_ combines the
2038 connection-oriented state maintenance known from TCP with the
2039 unreliable, datagram-based transport of UDP. This means that it
2040 is capable of reacting to changes in the connection by performing
2041 congestion control, offering multiple alternative approaches. But it
2042 is bound to datagram boundaries (the maximum packet size supported
2043 by a medium), and like UDP it lacks retransmission to protect
2044 against loss. Due to the use of sequence numbers, it is however
2045 able to react to loss (interpreted as a congestion indication) and
2046 to ignore out-of-order and duplicate packets. Unlike TCP it allows
2047 to negotiate specific, binding features for a connection, such as
2048 the choice of congestion control: classic, window-based congestion
2049 control known from TCP is available as CCID-2, rate-based, "smooth"
2050 congestion control is offered as CCID-3.
2051
2052 *HTTP*. _The Hypertext Transfer Protocol_ is an application layer
2053 protocol on top of TCP. It is spoken by web servers and is most often
2054 used for web services. However, as can be seen by the many Internet
2055 radio stations and YouTube/Flash videos, http is by far not limited to
2056 the delivery of web pages only. Being a simple request/response based
2057 protocol, the semantics of the protocol also allow the delivery of
2058 multimedia content, such as audio over http.
2059
2060 *Multicast*. IP multicast is not really a protocol but a technique
2061 for one-to-many communication over an IP network. The challenge is to
2062 deliver information to a group of destinations simultaneously using
2063 the most efficient strategy to send the messages over each link of
2064 the network only once. This has benefits for streaming multimedia:
2065 the standard one-to-one unicast offered by TCP/DCCP means that
2066 n clients listening to the same stream also consume n-times the
2067 resources, whereas multicast requires to send the stream just once,
2068 irrespective of the number of receivers. Since it would be costly to
2069 maintain state for each listening receiver, multicast often implies
2070 connectionless transport, which is the reason that it is currently
2071 only available via UDP.
2072
2073 License
2074 ~~~~~~~
2075
2076 Paraslash is licensed under the GPL, version 2. Most of the code
2077 base has been written from scratch, and those parts are GPL V2
2078 throughout. Notable exceptions are FEC and the WMA decoder. See the
2079 corresponding source files for licencing details for these parts. Some
2080 code sniplets of several other third party software packages have
2081 been incorporated into the paraslash sources, for example log message
2082 coloring was taken from the git sources. These third party software
2083 packages are all published under the GPL or some other license
2084 compatible to the GPL.
2085
2086 Acknowledgements
2087 ~~~~~~~~~~~~~~~~
2088
2089 Many thanks to Gerrit Renker who read an early draft of this manual
2090 and contributed significant improvements.
2091
2092 ----------
2093 References
2094 ----------
2095
2096 Articles
2097 ~~~~~~~~
2098 - Reed, Irving S.; Solomon, Gustave (1960),
2099 XREFERENCE(http://kom.aau.dk/~heb/kurser/NOTER/KOFA01.PDF,
2100 Polynomial Codes over Certain Finite Fields), Journal of the
2101 Society for Industrial and Applied Mathematics (SIAM) 8 (2):
2102 300-304, doi:10.1137/0108018)
2103
2104 RFCs
2105 ~~~~
2106
2107 - XREFERENCE(http://www.ietf.org/rfc/rfc768.txt, RFC 768) (1980):
2108 User Datagram Protocol
2109 - XREFERENCE(http://www.ietf.org/rfc/rfc791.txt, RFC 791) (1981):
2110 Internet Protocol
2111 - XREFERENCE(http://www.ietf.org/rfc/rfc2437.txt, RFC 2437) (1998):
2112 RSA Cryptography Specifications
2113 - XREFERENCE(http://www.ietf.org/rfc/rfc4340.txt, RFC 4340)
2114 (2006): Datagram Congestion Control Protocol (DCCP)
2115 - XREFERENCE(http://www.ietf.org/rfc/rfc4341.txt, RFC 4341) (2006):
2116 Congestion Control ID 2: TCP-like Congestion Control
2117 - XREFERENCE(http://www.ietf.org/rfc/rfc4342.txt, RFC 4342) (2006):
2118 Congestion Control ID 3: TCP-Friendly Rate Control (TFRC)
2119
2120 Application web pages
2121 ~~~~~~~~~~~~~~~~~~~~~
2122
2123 - XREFERENCE(http://paraslash.systemlinux.org/, paraslash)
2124 - XREFERENCE(http://xmms2.org/wiki/Main_Page, xmms)
2125 - XREFERENCE(http://www.mpg123.de/, mpg123)
2126 - XREFERENCE(http://gstreamer.freedesktop.org/, gstreamer)
2127 - XREFERENCE(http://www.icecast.org/, icecast)
2128 - XREFERENCE(http://beesbuzz.biz/code/audiocompress.php, Audio Compress)
2129
2130 External documentation
2131 ~~~~~~~~~~~~~~~~~~~~~~
2132
2133 - XREFERENCE(http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf,
2134 H. Peter Anvin: The mathematics of Raid6)
2135 - XREFERENCE(http://info.iet.unipi.it/~luigi/fec_ccr.ps.gz,
2136 Luigi Rizzo: Effective Erasure Codes for reliable Computer
2137 Communication Protocols)
2138
2139 Code
2140 ~~~~
2141 - XREFERENCE(http://info.iet.unipi.it/~luigi/vdm.tar.gz,
2142 Original FEC implementation by Luigi Rizzo)
2143