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