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