1 dnl To generate the html version, execute
2 dnl m4 web/manual.m4 | grutatxt --toc
4 define(`LOCAL_LINK_NAME', `translit(`$1', `A-Z
6 define(`REMOVE_NEWLINE', `translit(`$1',`
9 define(`REFERENCE', ./``#''`LOCAL_LINK_NAME($1)' (`REMOVE_NEWLINE($2)'))
10 define(`XREFERENCE', `$1' (`REMOVE_NEWLINE($2)'))
11 define(`EMPH', ``_''`REMOVE_NEWLINE($1)'``_'')
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
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.
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.
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.
47 Typically, these two daemons run on different hosts but a local setup
50 A simplified picture of a typical setup is as follows
53 server_host client_host
54 ~~~~~~~~~~~ ~~~~~~~~~~~
56 +-----------+ audio stream +-----------+
57 |para_server| -----------------------------> |para_audiod|
58 +-----------+ +-----------+
69 | connect +-----------+
70 +-------------------------------------- |para_client|
75 The paraslash executables
76 ~~~~~~~~~~~~~~~~~~~~~~~~~
80 para_server streams binary audio data (MP3, ...) over local and/or
81 remote networks. It listens on a TCP port and accepts commands such
82 as play, stop, pause, next from authenticated clients. There are
83 many more commands though, see the man page of para_server for a
84 description of all commands.
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).
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
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.
100 It is also possible to store images (album covers) and lyrics in the
101 database and associate these to the corresponding audio files.
103 The section on the REFERENCE(The audio file selector, audio file
104 selector) discusses this topic.
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.
114 All connections between para_server and para_client are encrypted
115 with a symmetric session key. For each user of paraslash you must
116 create a public/secret RSA key pair for authentication.
118 If para_client is started without non-option arguments, an interactive
119 session (shell) is started. Command history and command completion are
120 supported through libreadline.
124 The local daemon that collects information from para_server.
126 It runs on the client side and connects to para_server. As soon as
127 para_server announces the availability of an audio stream, para_audiod
128 starts an appropriate receiver, any number of filters and a paraslash
129 writer to play the stream.
131 Moreover, para_audiod listens on a local socket and sends status
132 information about para_server and para_audiod to local clients on
133 request. Access via this local socket may be restricted by using Unix
134 socket credentials, if available.
139 The client program which talks to para_audiod. Used to control
140 para_audiod, to receive status info, or to grab the stream at any
141 point of the decoding process. Like para_client, para_audioc supports
142 interactive sessions on systems with libreadline.
146 A command line HTTP/DCCP/UDP stream grabber. The http mode is
147 compatible with arbitrary HTTP streaming sources (e.g. icecast).
148 In addition to the three network streaming modes, para_recv can also
149 operate in local (afh) mode. In this mode it writes the content of
150 an audio file on the local file system in complete chunks to stdout,
151 optionally 'just in time'. This allows to cut an audio file without
152 first decoding it, and it enables third-party software which is unaware
153 of the particular audio format to send complete frames in real time.
157 A filter program that reads from STDIN and writes to STDOUT.
158 Like para_recv, this is an atomic building block which can be used to
159 assemble higher-level audio receiving facilities. It combines several
160 different functionalities in one tool: decoders for multiple audio
161 formats and a number of processing filters, among these a normalizer
166 A small stand-alone program that prints tech info about the given
167 audio file to STDOUT. It can be instructed to print a "chunk table",
168 an array of offsets within the audio file.
172 A modular audio stream writer. It supports a simple file writer
173 output plug-in and optional WAV/raw players for ALSA (Linux) and for
174 coreaudio (Mac OS). para_write can also be used as a stand-alone WAV
179 A command line audio player.
183 Curses-based gui that presents status information obtained in a curses
184 window. Appearance can be customized via themes. para_gui provides
185 key-bindings for the most common server commands and new key-bindings
191 An alarm clock and volume-fader for OSS and ALSA.
197 This chapter lists the REFERENCE(Requirements, necessary software)
198 that must be installed to compile the paraslash package, describes
199 how to REFERENCE(Installation, compile and install) the paraslash
200 source code and the steps that have to be performed in order to
201 REFERENCE(Quick start, set up) a typical server and client.
207 git clone git://git.tuebingen.mpg.de/osl
208 cd osl && make && sudo make install && sudo ldconfig
209 sudo apt-get install autoconf libssl-dev help2man gengetopt \
210 libmad0-dev libid3tag0-dev libasound2-dev libvorbis-dev \
211 libfaad-dev libspeex-dev libFLAC-dev libsamplerate-dev \
212 libasound2-dev libao-dev libreadline-dev libncurses-dev \
215 Detailed description: In any case you'll need
217 - XREFERENCE(http://people.tuebingen.mpg.de/maan/osl/, libosl).
218 The _object storage layer_ library is used by para_server. To
219 clone the source code repository, execute
221 git clone git://git.tuebingen.mpg.de/osl
223 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gcc, gcc) or
224 XREFERENCE(http://clang.llvm.org, clang). All gcc versions
225 >= 3.3 are currently supported. Clang version 1.1 or newer
228 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/make, gnu make) is
229 also shipped with the disto. On BSD systems the gnu make
230 executable is often called gmake.
232 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/bash, bash). Some
233 scripts which run during compilation require the EMPH(Bourne
234 again shell). It is most likely already installed.
236 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/gengetopt/, gengetopt)
237 is needed to generate the C code for the command line parsers
238 of all paraslash executables.
240 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/help2man, help2man)
241 is used to create the man pages.
245 - XREFERENCE(http://www.openssl.org/, openssl) or
246 XREFERENCE(ftp://ftp.gnupg.org/gcrypt/libgcrypt/, libgcrypt).
247 At least one of these two libraries is needed as the backend
248 for cryptographic routines on both the server and the client
249 side. Both openssl and libgcrypt are usually shipped with the
250 distro, but you might have to install the development package
251 (libssl-dev or libgcrypt-dev on debian systems) as well.
253 - XREFERENCE(http://www.underbit.com/products/mad/, libmad).
254 To compile in MP3 support for paraslash, the development
255 package must be installed. It is called libmad0-dev on
256 debian-based systems. Note that libmad is not necessary on
257 the server side, i.e. for sending MP3 files.
259 - XREFERENCE(http://www.underbit.com/products/mad/,
260 libid3tag). For version-2 ID3 tag support, you'll need
261 the libid3tag development package libid3tag0-dev. Without
262 libid3tag, only version one tags are recognized.
264 - XREFERENCE(http://www.xiph.org/downloads/, ogg vorbis).
265 For ogg vorbis streams you'll need libogg, libvorbis,
266 libvorbisfile. The corresponding Debian packages are called
267 libogg-dev and libvorbis-dev.
269 - XREFERENCE(http://www.audiocoding.com/, libfaad). For aac
270 files (m4a) you'll need libfaad (libfaad-dev).
272 - XREFERENCE(http://www.speex.org/, speex). In order to stream
273 or decode speex files, libspeex (libspeex-dev) is required.
275 - XREFERENCE(http://flac.sourceforge.net/, flac). To stream
276 or decode files encoded with the _Free Lossless Audio Codec_,
277 libFLAC (libFLAC-dev) must be installed.
279 - XREFERENCE(http://www.mega-nerd.com/SRC/index.html,
280 libsamplerate). The resample filter will only be compiled if
281 this library is installed. Debian package: libsamplerate-dev.
283 - XREFERENCE(ftp://ftp.alsa-project.org/pub/lib/, alsa-lib). On
284 Linux, you'll need to have ALSA's development package
285 libasound2-dev installed.
287 - XREFERENCE(http://downloads.xiph.org/releases/ao/,
288 libao). Needed to build the ao writer (ESD, PulseAudio,...).
289 Debian package: libao-dev.
291 - XREFERENCE(ftp://ftp.gnu.org/pub/gnu/ncurses, curses). Needed
292 for para_gui. Debian package: libncurses-dev.
294 - XREFERENCE(http://cnswww.cns.cwru.edu/php/chet/readline/rltop.html,
295 GNU Readline). If this library (libreadline-dev) is installed,
296 para_client, para_audioc and para_play support interactive
302 First make sure all non-optional packages listed in the section on
303 REFERENCE(Requirements, required software) are installed on your
306 You don't need everything listed there. In particular, MP3, OGG/Vorbis,
307 OGG/Speex and AAC support are all optional. The configure script will
308 detect what is installed on your system and will only try to build
309 those executables that can be built with your setup.
311 Note that no special decoder library (not even the MP3 decoding library
312 libmad) is needed for para_server if you only want to stream MP3 or WMA
313 files. Also, it's fine to use para_server on a box without sound card.
315 Next, install the paraslash package on all machines, you'd like this
316 software to run on. If you compile from a released tarball, execute
318 (./configure && make) > /dev/null
320 When compiling from git or from snapshots downloaded via gitweb,
321 the above command will not work because the configure script is not
322 included in the git repository. In this case the following command
323 should be used instead:
327 This runs autoconf to generate the configure script, then runs it as
328 above. Therefore you'll need autoconf for this to work.
330 There should be no errors but probably some warnings about missing
331 packages which usually implies that not all audio formats will be
332 supported. If headers or libs are installed at unusual locations you
333 might need to tell the configure script where to find them. Try
337 to see a list of options. If the paraslash package was compiled
338 successfully, execute (optionally)
342 to run the paraslash test suite. If all tests pass, execute as root
346 to install executables under /usr/local/bin and the man pages under
352 *Step 1*: Create a paraslash user
354 In order to control para_server at runtime you must create a paraslash
355 user. As authentication is based on the RSA crypto system you'll have
356 to create an RSA key pair. If you already have a user and an RSA key
357 pair, you may skip this step.
359 In this section we'll assume a typical setup: You would like to run
360 para_server on some host called server_host as user foo, and you want
361 to connect to para_server from another machine called client_host as
364 As foo@server_host, create ~/.paraslash/server.users by typing the
368 target=~/.paraslash/server.users
369 key=~/.paraslash/id_rsa.pub.$user
370 perms=AFS_READ,AFS_WRITE,VSS_READ,VSS_WRITE
371 mkdir -p ~/.paraslash
372 echo "user $user $key $perms" >> $target
374 Next, change to the "bar" account on client_host and generate the
375 key pair with the commands
377 ssh-keygen -t rsa -b 2048
378 # hit enter twice to create a key with no passphrase
380 This generates the two files id_rsa and id_rsa.pub in ~/.ssh.
381 Note that para_server refuses to use a key if it is shorter than
382 2048 bits. Moreover, para_client refuses to use a private key which
385 para_server only needs to know the public key of the key pair just
386 created. Copy this public key to server_host:
388 src=~/.ssh/id_rsa.pub
389 dest=.paraslash/id_rsa.pub.$LOGNAME
390 scp $src foo@server_host:$dest
392 Finally, tell para_client to connect to server_host:
394 conf=~/.paraslash/client.conf
395 echo 'hostname server_host' > $conf
398 *Step 2*: Start para_server
400 Before starting the server make sure you have write permissions to
401 the directory /var/paraslash that has been created during installation:
403 sudo chown $LOGNAME /var/paraslash
405 Alternatively, use the --afs-socket Option to specify a different
406 location for the AFS command socket.
408 For this first try, we'll use the info loglevel to make the output
409 of para_server more verbose.
413 Now you can use para_client to connect to the server and issue
414 commands. Open a new shell as bar@client_host and try
419 to retrieve the list of available commands and some server info.
420 Don't proceed if this doesn't work.
422 *Step 3*: Create and populate the database
424 An empty database is created with
428 This initializes a couple of empty tables under
429 ~/.paraslash/afs_database-0.4. You normally don't need to look at these
430 tables, but it's good to know that you can start from scratch with
432 rm -rf ~/.paraslash/afs_database-0.4
434 in case something went wrong.
436 Next, you need to add some audio files to that database so that
437 para_server knows about them. Choose an absolute path to a directory
438 containing some audio files and add them to the audio file table:
440 para_client add /my/mp3/dir
442 This might take a while, so it is a good idea to start with a directory
443 containing not too many files. Note that the table only contains data
444 about the audio files found, not the files themselves.
446 You may print the list of all known audio files with
450 *Step 4*: Configure para_audiod
452 para_audiod needs to create a "well-known" socket for the clients to
453 connect to. The default path for this socket is
455 /var/paraslash/audiod_socket.$HOSTNAME
457 In order to make this directory writable for para_audiod, execute
460 sudo chown $LOGNAME /var/paraslash
463 We will also have to tell para_audiod that it should receive the
464 audio stream from server_host via http:
466 para_audiod -l info -r '.:http -i server_host'
468 You should now be able to listen to the audio stream once para_server
469 starts streaming. To activate streaming, execute
473 Since no playlist has been specified yet, the "dummy" mode which
474 selects all known audio files is activated automatically. See the
475 section on the REFERENCE(The audio file selector, audio file selector)
476 for how to use playlists and moods to specify which files should be
477 streamed in which order.
481 It did not work? To find out why, try to receive, decode and play the
482 stream manually using para_recv, para_filter and para_write as follows.
484 For simplicity we assume that you're running Linux/ALSA and that only
485 MP3 files have been added to the database.
487 para_recv -r 'http -i server_host' > file.mp3
488 # (interrupt with CTRL+C after a few seconds)
489 ls -l file.mp3 # should not be empty
490 para_filter -f mp3dec -f wav < file.mp3 > file.wav
491 ls -l file.wav # should be much bigger than file.mp3
492 para_write -w alsa < file.wav
494 Double check what is logged by para_server and use the --loglevel
495 option of para_recv, para_filter and para_write to increase verbosity.
501 para_server uses a challenge-response mechanism to authenticate
502 requests from incoming connections, similar to ssh's public key
503 authentication method. Authenticated connections are encrypted using
504 a stream cipher, either RC4 or AES in integer counter mode.
506 In this chapter we briefly describe RSA, RC4 and AES, and sketch the
507 REFERENCE(Client-server authentication, authentication handshake)
508 between para_client and para_server. User management is discussed
509 in the section on REFERENCE(The user_list file, the user_list file).
510 These sections are all about communication between the client and the
511 server. Connecting para_audiod is a different matter and is described
512 in a REFERENCE(Connecting para_audiod, separate section).
519 RSA is an asymmetric block cipher which is used in many applications,
520 including ssh and gpg. An RSA key consists in fact of two keys,
521 called the public key and the private key. A message can be encrypted
522 with either key and only the counterpart of that key can decrypt
523 the message. While RSA can be used for both signing and encrypting
524 a message, paraslash uses RSA only for the latter purpose. The
525 RSA public key encryption and signatures algorithms are defined in
528 RC4 is a stream cipher, i.e. the input is XORed with a pseudo-random
529 key stream to produce the output. Decryption uses the same function
530 calls as encryption. While RC4 supports variable key lengths,
531 paraslash uses a fixed length of 256 bits, which is considered a
532 strong encryption by today's standards. Since the same key must never
533 be used twice, a different, randomly-generated key is used for every
536 AES, the advanced encryption standard, is a well-known symmetric block
537 cipher, i.e. a transformation operating on fixed-length blocks which
538 is determined by a single key for both encryption and decryption. Any
539 block cipher can be turned into a stream cipher by generating
540 a pseudo-random key stream by encrypting successive values of a
541 counter. The AES_CTR128 stream cipher used in paraslash is obtained
542 in this way from the AES block cipher with a 128 bit block size.
545 Client-server authentication
546 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
548 The authentication handshake between para_client and para_server goes
551 - para_client connects to para_server and sends an
552 authentication request for a user. It does so by connecting
553 to TCP port 2990 of the server host. This port is called the
554 para_server _control port_.
556 - para_server accepts the connection and forks a child process
557 which handles the incoming request. The parent process keeps
558 listening on the control port while the child process (also
559 called para_server below) continues as follows.
561 - para_server loads the RSA public key of that user, fills a
562 fixed-length buffer with random bytes, encrypts that buffer
563 using the public key and sends the encrypted buffer to the
564 client. The first part of the buffer is the challenge which
565 is used for authentication while the second part is the
568 - para_client receives the encrypted buffer and decrypts it
569 with the user's private key, thereby obtaining the challenge
570 buffer and the session key. It sends the SHA1 hash value of
571 the challenge back to para_server and stores the session key
574 - para_server also computes the SHA1 hash of the challenge
575 and compares it against what was sent back by the client.
577 - If the two hashes do not match, the authentication has
578 failed and para_server closes the connection.
580 - Otherwise the user is considered authenticated and the client
581 is allowed to proceed by sending a command to be executed. From
582 this point on the communication is encrypted using the stream
583 cipher with the session key known to both peers.
585 paraslash relies on the quality of the pseudo-random bytes provided
586 by the crypto library (openssl or libgcrypt), on the security of
587 the implementation of the RSA, RC4 and AES crypto routines and on the
588 infeasibility to invert the SHA1 function.
590 Neither para_server or para_client create RSA keys on their own. This
591 has to be done once for each user as sketched in REFERENCE(Quick start,
592 Quick start) and discussed in more detail REFERENCE(The user_list
598 At startup para_server reads the user list file which contains one
599 line per user. The default location of the user list file may be
600 changed with the --user-list option.
602 There should be at least one user in this file. Each user must have
603 an RSA key pair. The public part of the key is needed by para_server
604 while the private key is needed by para_client. Each line of the
605 user list file must be of the form
607 user <username> <key> <perms>
609 where _username_ is an arbitrary string (usually the user's login
610 name), _key_ is the full path to that user's public RSA key, and
611 _perms_ is a comma-separated list of zero or more of the following
614 +---------------------------------------------------------+
615 | AFS_READ | read the contents of the databases |
616 +-----------+---------------------------------------------+
617 | AFS_WRITE | change database contents |
618 +-----------+---------------------------------------------+
619 | VSS_READ | obtain information about the current stream |
620 +-----------+---------------------------------------------+
621 | VSS_WRITE | change the current stream |
622 +---------------------------------------------------------+
624 The permission bits specify which commands the user is allowed to
625 execute. The output of
629 contains in the third column the permissions needed to execute the
632 It is possible to make para_server reread the user_list file by
633 executing the paraslash "hup" command or by sending SIGHUP to the
637 Connecting para_audiod
638 ~~~~~~~~~~~~~~~~~~~~~~
640 para_audiod listens on a Unix domain socket. Those sockets are
641 for local communication only, so only local users can connect to
642 para_audiod. The default is to let any user connect but this can be
643 restricted on platforms that support UNIX socket credentials which
644 allow para_audiod to obtain the Unix credentials of the connecting
647 Use para_audiod's --user-allow option to allow connections only for
648 a limited set of users.
650 -----------------------
651 The audio file selector
652 -----------------------
654 paraslash comes with a sophisticated audio file selector (AFS),
655 whose main task is to determine which file to stream next, based on
656 information on the audio files stored in a database. It communicates
657 also with para_client whenever an AFS command is executed, for example
658 to answer a database query.
660 Besides the traditional playlists, AFS supports audio file selection
661 based on _moods_ which act as a filter that limits the set of all
662 known audio files to those which satisfy certain criteria. It also
663 maintains tables containing images (e.g. album cover art) and lyrics
664 that can be associated with one or more audio files.
666 AFS uses XREFERENCE(http://people.tuebingen.mpg.de/maan/osl/, libosl), the
667 object storage layer library, as the backend library for storing
668 information on audio files, playlists, etc. This library offers
669 functionality similar to a relational database, but is much more
670 lightweight than a full database backend.
672 In this chapter we sketch the setup of the REFERENCE(The AFS process,
673 AFS process) during server startup and proceed with the description
674 of the REFERENCE(Database layout, layout) of the various database
675 tables. The section on REFERENCE(Playlists and moods, playlists
676 and moods) explains these two audio file selection mechanisms
677 in detail and contains pratical examples. The way REFERENCE(File
678 renames and content changes, file renames and content changes) are
679 detected is discussed briefly before the REFERENCE(Troubleshooting,
680 Troubleshooting) section concludes the chapter.
685 On startup, para_server forks to create the AFS process which opens
686 the OSL database tables. The server process communicates with the
687 AFS process via pipes and shared memory. Usually, the AFS process
688 awakes only briefly whenever the current audio file changes. The AFS
689 process determines the next audio file, opens it, verifies it has
690 not been changed since it was added to the database and passes the
691 open file descriptor to the server process, along with audio file
692 meta-data such as file name, duration, audio format and so on. The
693 server process then starts to stream the audio file.
695 The AFS process also accepts connections from local clients via
696 a well-known socket. However, only child processes of para_server
697 may connect through this socket. All server commands that have the
698 AFS_READ or AFS_WRITE permission bits use this mechanism to query or
704 *The audio file table*
706 This is the most important and usually also the largest table of the
707 AFS database. It contains the information needed to stream each audio
708 file. In particular the following data is stored for each audio file.
710 - SHA1 hash value of the audio file contents. This is computed
711 once when the file is added to the database. Whenever AFS
712 selects this audio file for streaming the hash value is
713 recomputed and checked against the value stored in the
714 database to detect content changes.
716 - The time when this audio file was last played.
718 - The number of times the file has been played so far.
720 - The attribute bitmask.
722 - The image id which describes the image associated with this
725 - The lyrics id which describes the lyrics associated with
728 - The audio format id (MP3, OGG, ...).
730 - An amplification value that can be used by the amplification
731 filter to pre-amplify the decoded audio stream.
733 - The chunk table. It describes the location and the timing
734 of the building blocks of the audio file. This is used by
735 para_server to send chunks of the file at appropriate times.
737 - The duration of the audio file.
739 - Tag information contained in the audio file (ID3 tags,
740 Vorbis comments, ...).
742 - The number of channels
744 - The encoding bitrate.
746 - The sampling frequency.
748 To add or refresh the data contained in the audio file table, the _add_
749 command is used. It takes the full path of either an audio file or a
750 directory. In the latter case, the directory is traversed recursively
751 and all files which are recognized as valid audio files are added to
754 *The attribute table*
756 The attribute table contains two columns, _name_ and _bitnum_. An
757 attribute is simply a name for a certain bit number in the attribute
758 bitmask of the audio file table.
760 Each of the 64 bits of the attribute bitmask can be set for each
761 audio file individually. Hence up to 64 different attributes may be
762 defined. For example, "pop", "rock", "blues", "jazz", "instrumental",
763 "german_lyrics", "speech", whatever. You are free to choose as
764 many attributes as you like and there are no naming restrictions
767 A new attribute "test" is created by
769 para_client addatt test
773 lists all available attributes. You can set the "test" attribute for
774 an audio file by executing
776 para_client setatt test+ /path/to/the/audio/file
778 Similarly, the "test" bit can be removed from an audio file with
780 para_client setatt test- /path/to/the/audio/file
782 Instead of a path you may use a shell wildcard pattern. The attribute
783 is applied to all audio files matching this pattern:
785 para_client setatt test+ '/test/directory/*'
789 para_client -- ls -lv
791 gives you a verbose listing of your audio files also showing which
794 In case you wonder why the double-dash in the above command is needed:
795 It tells para_client to not interpret the options after the dashes. If
796 you find this annoying, just say
798 alias para='para_client --'
800 and be happy. In what follows we shall use this alias.
802 The "test" attribute can be dropped from the database with
811 for more information and a complete list of command line options to
816 The image, lyrics, moods and playlists tables are all blob tables.
817 Blob tables consist of three columns each: The identifier which is
818 a positive non-negative number that is auto-incremented, the name
819 (an arbitrary string) and the content (the blob).
821 All blob tables support the same set of actions: cat, ls, mv, rm
822 and add. Of course, _add_ is used for adding new blobs to the table
823 while the other actions have the same meaning as the corresponding
824 Unix commands. The paraslash commands to perform these actions are
825 constructed as the concatenation of the table name and the action. For
826 example addimg, catimg, lsimg, mvimg, rmimg are the commands that
827 manipulate or query the image table.
829 The add variant of these commands is special as these commands read
830 the blob contents from stdin. To add an image to the image table the
833 para addimg image_name < file.jpg
837 Note that the images and lyrics are not interpreted at all, and also
838 the playlist and the mood blobs are only investigated when the mood
839 or playlist is activated with the select command.
843 Unlike all other tables the contents of the score table remain in
844 memory and are never stored on disk. The score table contains two
845 columns: The SHA1 hash value (of an audio file) and its current
848 However, only those files which are admissible for the current mood
849 or playlist are contained in the score table. The audio file selector
850 always chooses the row with the highest score as the file to stream
851 next. While doing so, it computes the new score and updates the
852 last_played and the num_played fields in the audio file table.
854 The score table is recomputed by the select command which loads a
855 mood or playlist. Audio files are chosen for streaming from the rows
856 of the score table on a highest-score-first basis.
862 Playlists and moods offer two different ways of specifying the set of
863 admissible files. A playlist in itself describes a set of admissible
864 files. A mood, in contrast, describes the set of admissible files in
865 terms of attributes and other type of information available in the
866 audio file table. As an example, a mood can define a filename pattern,
867 which is then matched against the names of audio files in the table.
871 Playlists are accommodated in the playlist table of the afs database,
872 using the aforementioned blob format for tables. A new playlist is
873 created with the addpl command by specifying the full (absolute)
874 paths of all desired audio files, separated by newlines. Example:
876 find /my/mp3/dir -name "*.mp3" | para addpl my_playlist
878 If _my_playlist_ already exists it is overwritten. To activate the
879 new playlist, execute
881 para select p/my_playlist
883 The audio file selector will assign scores to each entry of the list,
884 in descending order so that files will be selected in order. If a
885 file could not be opened for streaming, its entry is removed from
886 the score table (but not from the playlist).
890 A mood consists of a unique name and its *mood definition*, which is
891 a set of *mood lines* containing expressions in terms of attributes
892 and other data contained in the database.
894 At any time at most one mood can be *active* which means that
895 para_server is going to select only files from that subset of
898 So in order to create a mood definition one has to write a set of
899 mood lines. Mood lines come in three flavours: Accept lines, deny
900 lines and score lines.
902 The general syntax of the three types of mood lines is
905 accept [with score <score>] [if] [not] <mood_method> [options]
906 deny [with score <score>] [if] [not] <mood_method> [options]
907 score <score> [if] [not] <mood_method> [options]
910 Here <score> is either an integer or the string "random" which assigns
911 a random score to all matching files. The score value changes the
912 order in which admissible files are going to be selected, but is of
913 minor importance for this introduction.
915 So we concentrate on the first two forms, i.e. accept and deny
916 lines. As usual, everything in square brackets is optional, i.e.
917 accept/deny lines take the following form when ignoring scores:
919 accept [if] [not] <mood_method> [options]
921 and analogously for the deny case. The "if" keyword is only syntactic
922 sugar and has no function. The "not" keyword just inverts the result,
923 so the essence of a mood line is the mood method part and the options
924 following thereafter.
926 A *mood method* is realized as a function which takes an audio file
927 and computes a number from the data contained in the database.
928 If this number is non-negative, we say the file *matches* the mood
929 method. The file matches the full mood line if it either
931 - matches the mood method and the "not" keyword is not given,
933 - does not match the mood method, but the "not" keyword is given.
935 The set of admissible files for the whole mood is now defined as those
936 files which match at least one accept mood line, but no deny mood line.
937 More formally, an audio file F is admissible if and only if
939 (F ~ AL1 or F ~ AL2...) and not (F ~ DL1 or F ~ DN2 ...)
941 where AL1, AL2... are the accept lines, DL1, DL2... are the deny
942 lines and "~" means "matches".
944 The cases where no mood lines of accept/deny type are defined need
947 - Neither accept nor deny lines: This treats all files as
948 admissible (in fact, that is the definition of the dummy mood
949 which is activated automatically if no moods are available).
951 - Only accept lines: A file is admissible iff it matches at
952 least one accept line:
954 F ~ AL1 or F ~ AL2 or ...
956 - Only deny lines: A file is admissible iff it matches no
959 not (F ~ DL1 or F ~ DN2 ...)
963 *List of mood_methods*
967 Takes no arguments and matches an audio file if and only if no
970 is_set <attribute_name>
972 Takes the name of an attribute and matches iff that attribute is set.
974 path_matches <pattern>
976 Takes a filename pattern and matches iff the path of the audio file
979 artist_matches <pattern>
980 album_matches <pattern>
981 title_matches <pattern>
982 comment_matches <pattern>
984 Takes an extended regular expression and matches iff the text of the
985 corresponding tag of the audio file matches the pattern. If the tag
986 is not set, the empty string is matched against the pattern.
994 Takes a comparator ~ of the set {<, =, <=, >, >=, !=} and a number
995 <num>. Matches an audio file iff the condition <val> ~ <num> is
996 satisfied where val is the corresponding value of the audio file
997 (value of the year tag, bitrate in kbit/s, frequency in Hz, channel
1000 The year tag is special as its value is undefined if the audio file
1001 has no year tag or the content of the year tag is not a number. Such
1002 audio files never match. Another difference is the special treatment
1003 if the year tag is a two-digit number. In this case either 1900 or
1004 2000 is added to the tag value, depending on whether the number is
1005 greater than 2000 plus the current year.
1010 To create a new mood called "my_mood", write its definition into
1011 some temporary file, say "tmpfile", and add it to the mood table
1014 para addmood my_mood < tmpfile
1016 If the mood definition is really short, you may just pipe it to the
1017 client instead of using temporary files. Like this:
1019 echo "$MOOD_DEFINITION" | para addmood my_mood
1021 There is no need to keep the temporary file since you can always use
1022 the catmood command to get it back:
1024 para catmood my_mood
1026 A mood can be activated by executing
1028 para select m/my_mood
1030 Once active, the list of admissible files is shown by the ls command
1031 if the "-a" switch is given:
1036 *Example mood definition*
1038 Suppose you have defined attributes "punk" and "rock" and want to define
1039 a mood containing only Punk-Rock songs. That is, an audio file should be
1040 admissible if and only if both attributes are set. Since
1044 is obviously the same as
1046 not (not punk or not rock)
1048 (de Morgan's rule), a mood definition that selects only Punk-Rock
1051 deny if not is_set punk
1052 deny if not is_set rock
1056 File renames and content changes
1057 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1059 Since the audio file selector knows the SHA1 of each audio file that
1060 has been added to the afs database, it recognizes if the content of
1061 a file has changed, e.g. because an ID3 tag was added or modified.
1062 Also, if a file has been renamed or moved to a different location,
1063 afs will detect that an entry with the same hash value already exists
1064 in the audio file table.
1066 In both cases it is enough to just re-add the new file. In the
1067 first case (file content changed), the audio table is updated, while
1068 metadata such as the num_played and last_played fields, as well as
1069 the attributes, remain unchanged. In the other case, when the file
1070 is moved or renamed, only the path information is updated, all other
1071 data remains as before.
1073 It is possible to change the behaviour of the add command by using the
1074 "-l" (lazy add) or the "-f" (force add) option.
1079 Use the debug loglevel (-l debug) to show debugging info. All paraslash
1080 executables have a brief online help which is displayed when -h is
1081 given. The --detailed-help option prints the full help text.
1083 If para_server crashed or was killed by SIGKILL (signal 9), it
1084 may refuse to start again because of "dirty osl tables". In this
1085 case you'll have to run the oslfsck program of libosl to fix your
1088 oslfsck -fd ~/.paraslash/afs_database-0.4
1090 However, make sure para_server isn't running before executing oslfsck.
1092 If you don't mind to recreate your database you can start
1093 from scratch by removing the entire database directory, i.e.
1095 rm -rf ~/.paraslash/afs_database-0.4
1097 Be aware that this removes all attribute definitions, all playlists
1098 and all mood definitions and requires to re-initialize the tables.
1100 Although oslfsck fixes inconsistencies in database tables it doesn't
1101 care about the table contents. To check for invalid table contents, use
1105 This prints out references to missing audio files as well as invalid
1106 playlists and mood definitions.
1108 Similarly, para_audiod refuses to start if its socket file exists, since
1109 this indicates that another instance of para_audiod is running. After
1110 a crash a stale socket file might remain and you must run
1116 ---------------------------------------
1117 Audio formats and audio format handlers
1118 ---------------------------------------
1123 The following audio formats are supported by paraslash:
1127 Mp3, MPEG-1 Audio Layer 3, is a common audio format for audio storage,
1128 designed as part of its MPEG-1 standard. An MP3 file is made up of
1129 multiple MP3 frames, which consist of a header and a data block. The
1130 size of an MP3 frame depends on the bit rate and on the number
1131 of channels. For a typical CD-audio file (sample rate of 44.1 kHz
1132 stereo), encoded with a bit rate of 128 kbit, an MP3 frame is about
1137 OGG is a standardized audio container format, while Vorbis is an
1138 open source codec for lossy audio compression. Since Vorbis is most
1139 commonly made available via the OGG container format, it is often
1140 referred to as OGG/Vorbis. The OGG container format divides data into
1141 chunks called OGG pages. A typical OGG page is about 4KB large. The
1142 Vorbis codec creates variable-bitrate (VBR) data, where the bitrate
1143 may vary considerably.
1147 Speex is an open-source speech codec that is based on CELP (Code
1148 Excited Linear Prediction) coding. It is designed for voice
1149 over IP applications, has modest complexity and a small memory
1150 footprint. Wideband and narrowband (telephone quality) speech are
1151 supported. As for Vorbis audio, Speex bit-streams are often stored
1152 in OGG files. As of 2012 this codec is considered obsolete since the
1153 Oppus codec, described below, surpasses its performance in all areas.
1157 Opus is a lossy audio compression format standardized through RFC
1158 6716 in 2012. It combines the speech-oriented SILK codec and the
1159 low-latency CELT (Constrained Energy Lapped Transform) codec. Like
1160 OGG/Vorbis and OGG/Speex, Opus data is usually encapsulated in OGG
1161 containers. All known software patents which cover Opus are licensed
1162 under royalty-free terms.
1166 Advanced Audio Coding (AAC) is a standardized, lossy compression
1167 and encoding scheme for digital audio which is the default audio
1168 format for Apple's iPhone, iPod, iTunes. Usually MPEG-4 is used as
1169 the container format and audio files encoded with AAC have the .m4a
1170 extension. A typical AAC frame is about 700 bytes large.
1174 Windows Media Audio (WMA) is an audio data compression technology
1175 developed by Microsoft. A WMA file is usually encapsulated in the
1176 Advanced Systems Format (ASF) container format, which also specifies
1177 how meta data about the file is to be encoded. The bit stream of WMA
1178 is composed of superframes, each containing one or more frames of
1179 2048 samples. For 16 bit stereo a WMA superframe is about 8K large.
1183 The Free Lossless Audio Codec (FLAC) compresses audio without quality
1184 loss. It gives better compression ratios than a general purpose
1185 compressor like zip or bzip2 because FLAC is designed specifically
1186 for audio. A FLAC-encoded file consists of frames of varying size, up
1187 to 16K. Each frame starts with a header that contains all information
1188 necessary to decode the frame.
1193 Unfortunately, each audio format has its own conventions how meta
1194 data is added as tags to the audio file.
1196 For MP3 files, ID3, version 1 and 2 are widely used. ID3 version 1
1197 is rather simple but also very limited as it supports only artist,
1198 title, album, year and comment tags. Each of these can only be at most
1199 32 characters long. ID3, version 2 is much more flexible but requires
1200 a separate library being installed for paraslash to support it.
1202 Ogg vorbis, ogg speex and flac files contain meta data as Vorbis
1203 comments, which are typically implemented as strings of the form
1204 "[TAG]=[VALUE]". Unlike ID3 version 1 tags, one may use whichever
1205 tags are appropriate for the content.
1207 AAC files usually use the MPEG-4 container format for storing meta
1208 data while WMA files wrap meta data as special objects within the
1209 ASF container format.
1211 paraslash only tracks the most common tags that are supported by
1212 all tag variants: artist, title, year, album, comment. When a file
1213 is added to the AFS database, the meta data of the file is extracted
1214 and stored in the audio file table.
1216 Chunks and chunk tables
1217 ~~~~~~~~~~~~~~~~~~~~~~~
1219 paraslash uses the word "chunk" as common term for the building blocks
1220 of an audio file. For MP3 files, a chunk is the same as an MP3 frame,
1221 while for OGG files a chunk is an OGG page, etc. Therefore the chunk
1222 size varies considerably between audio formats, from a few hundred
1223 bytes (MP3) up to 16K (FLAC).
1225 The chunk table contains the offsets within the audio file that
1226 correspond to the chunk boundaries of the file. Like the meta data,
1227 the chunk table is computed and stored in the database whenever an
1228 audio file is added.
1230 The paraslash senders (see below) always send complete chunks. The
1231 granularity for seeking is therefore determined by the chunk size.
1233 Audio format handlers
1234 ~~~~~~~~~~~~~~~~~~~~~
1236 For each audio format paraslash contains an audio format handler whose
1237 first task is to tell whether a given file is a valid audio file of
1238 this type. If so, the audio file handler extracts some technical data
1239 (duration, sampling rate, number of channels etc.), computes the
1240 chunk table and reads the meta data.
1242 The audio format handler code is linked into para_server and executed
1243 via the _add_ command. The same code is also available as a stand-alone
1244 tool, para_afh, which prints the technical data, the chunk table
1245 and the meta data of a file. Moreover, all audio format handlers are
1246 combined in the afh receiver which is part of para_recv and para_play.
1252 Paraslash uses different network connections for control and data.
1253 para_client communicates with para_server over a dedicated TCP control
1254 connection. To transport audio data, separate data connections are
1255 used. For these data connections, a variety of transports (UDP, DCCP,
1256 HTTP) can be chosen.
1258 The chapter starts with the REFERENCE(The paraslash control
1259 service, control service), followed by a section on the various
1260 REFERENCE(Streaming protocols, streaming protocols) in which the data
1261 connections are described. The way audio file headers are embedded into
1262 the stream is discussed REFERENCE(Streams with headers and headerless
1263 streams, briefly) before the REFERENCE(Networking examples, example
1264 section) which illustrates typical commands for real-life scenarios.
1266 Both IPv4 and IPv6 are supported.
1268 The paraslash control service
1269 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1271 para_server is controlled at runtime via the paraslash control
1272 connection. This connection is used for server commands (play, stop,
1273 ...) as well as for afs commands (ls, select, ...).
1275 The server listens on a TCP port and accepts connections from clients
1276 that connect the open port. Each connection causes the server to fork
1277 off a client process which inherits the connection and deals with that
1278 client only. In this classical accept/fork approach the server process
1279 is unaffected if the child dies or goes crazy for whatever reason. In
1280 fact, the child process can not change address space of server process.
1282 The section on REFERENCE(Client-server authentication, client-server
1283 authentication) above described the early connection establishment
1284 from the crypto point of view. Here it is described what happens
1285 after the connection (including crypto setup) has been established.
1286 There are four processes involved during command dispatch as sketched
1287 in the following diagram.
1291 server_host client_host
1292 ~~~~~~~~~~~ ~~~~~~~~~~~
1294 +-----------+ connect +-----------+
1295 |para_server|<------------------------------ |para_client|
1296 +-----------+ +-----------+
1299 +----------> |AFS| |
1303 | | connect (cookie) |
1306 | fork +-----+ inherited connection |
1307 +---------->|child|<--------------------------+
1312 Note that the child process is not a child of the afs process,
1313 so communication of these two processes has to happen via local
1314 sockets. In order to avoid abuse of the local socket by unrelated
1315 processes, a magic cookie is created once at server startup time just
1316 before the server process forks off the AFS process. This cookie is
1317 known to the server, AFS and the child, but not to unrelated processes.
1319 There are two different kinds of commands: First there are commands
1320 that cause the server to respond with some answer such as the list
1321 of all audio files. All but the addblob commands (addimg, addlyr,
1322 addpl, addmood) are of this kind. The addblob commands add contents
1323 to the database, so they need to transfer data the other way round,
1324 from the client to the server.
1326 There is no knowledge about the server commands built into para_client,
1327 so it does not know about addblob commands. Instead, it inspects the
1328 first data package sent by the server for a magic string. If this
1329 string was found, it sends STDIN to the server, otherwise it dumps
1330 data from the server to STDOUT.
1335 A network (audio) stream usually consists of one streaming source,
1336 the _sender_, and one or more _receivers_ which read data over the
1337 network from the streaming source.
1339 Senders are thus part of para_server while receivers are part of
1340 para_audiod. Moreover, there is the stand-alone tool para_recv which
1341 can be used to manually download a stream, either from para_server
1342 or from a web-based audio streaming service.
1344 The following three streaming protocols are supported by paraslash:
1346 - HTTP. Recommended for public streams that can be played by
1347 any player like mpg123, xmms, itunes, winamp, etc. The HTTP
1348 sender is supported on all operating systems and all platforms.
1350 - DCCP. Recommended for LAN streaming. DCCP is currently
1351 available only for Linux.
1353 - UDP. Recommended for multicast LAN streaming.
1355 See the Appendix on REFERENCE(Network protocols, network protocols)
1356 for brief descriptions of the various protocols relevant for network
1357 audio streaming with paraslash.
1359 It is possible to activate more than one sender simultaneously.
1360 Senders can be controlled at run time and via config file and command
1363 Note that audio connections are _not_ encrypted. Transport or Internet
1364 layer encryption should be used if encrypted data connections are
1367 Since DCCP and TCP are both connection-oriented protocols, connection
1368 establishment/teardown and access control are very similar between
1369 these two streaming protocols. UDP is the most lightweight option,
1370 since in contrast to TCP/DCCP it is connectionless. It is also the
1371 only protocol supporting IP multicast.
1373 The HTTP and the DCCP sender listen on a (TCP/DCCP) port waiting for
1374 clients to connect and establish a connection via some protocol-defined
1375 handshake mechanism. Both senders maintain two linked lists each:
1376 The list of all clients which are currently connected, and the list
1377 of access control entries which determines who is allowed to connect.
1378 IP-based access control may be configured through config file and
1379 command line options and via the "allow" and "deny" sender subcommands.
1381 Upon receiving a GET request from the client, the HTTP sender sends
1382 back a status line and a message. The body of this message is the
1383 audio stream. This is common practice and is supported by many popular
1384 clients which can thus be used to play a stream offered by para_server.
1385 For DCCP things are a bit simpler: No messages are exchanged between
1386 the receiver and sender. The client simply connects and the sender
1389 DCCP is an experimental protocol which offers a number of new features
1390 not available for TCP. Both ends can negotiate these features using
1391 a built-in negotiation mechanism. In contrast to TCP/HTTP, DCCP is
1392 datagram-based (no retransmissions) and thus should not be used over
1393 lossy media (e.g. WiFi networks). One useful feature offered by DCCP
1394 is access to a variety of different congestion-control mechanisms
1395 called CCIDs. Two different CCIDs are available per default on Linux:
1398 - _CCID 2_. A Congestion Control mechanism similar to that
1399 of TCP. The sender maintains a congestion window and halves
1400 this window in response to congestion.
1403 - _CCID-3_. Designed to be fair when competing for bandwidth.
1404 It has lower variation of throughput over time compared with
1405 TCP, which makes it suitable for streaming media.
1407 Unlike the HTTP and DCCP senders, the UDP sender maintains only a
1408 single list, the _target list_. This list describes the set of clients
1409 to which the stream is sent. There is no list for access control and
1410 no "allow" and "deny" commands for the UDP sender. Instead, the "add"
1411 and "delete" commands can be used to modify the target list.
1413 Since both UDP and DCCP offer an unreliable datagram-based transport,
1414 additional measures are necessary to guard against disruptions over
1415 networks that are lossy or which may be subject to interference (as
1416 is for instance the case with WiFi). Paraslash uses FEC (Forward
1417 Error Correction) to guard against packet losses and reordering. The
1418 stream is FEC-encoded before it is sent through the UDP socket and
1419 must be decoded accordingly on the receiver side.
1421 The packet size and the amount of redundancy introduced by FEC can
1422 be configured via the FEC parameters which are dictated by server
1423 and may also be configured through the "sender" command. The FEC
1424 parameters are encoded in the header of each network packet, so no
1425 configuration is necessary on the receiver side. See the section on
1426 REFERENCE(Forward error correction, FEC) below.
1428 Streams with headers and headerless streams
1429 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1431 For OGG/Vorbis, OGG/Speex and wma streams, some of the information
1432 needed to decode the stream is only contained in the audio file
1433 header of the container format but not in each data chunk. Clients
1434 must be able to obtain this information in case streaming starts in
1435 the middle of the file or if para_audiod is started while para_server
1436 is already sending a stream.
1438 This is accomplished in different ways, depending on the streaming
1439 protocol. For connection-oriented streams (HTTP, DCCP) the audio file
1440 header is sent prior to audio file data. This technique however does
1441 not work for the connectionless UDP transport. Hence the audio file
1442 header is periodically being embedded into the UDP audio data stream.
1443 By default, the header is resent after five seconds. The receiver has
1444 to wait until the next header arrives before it can start decoding
1450 The "si" (server info) command lists some information about the
1451 currently running server process.
1453 -> Show PIDs, number of connected clients, uptime, and more:
1457 The sender command of para_server prints information about senders,
1458 like the various access control lists, and it allows to (de-)activate
1459 senders and to change the access permissions at runtime.
1465 -> Obtain general help for the sender command:
1467 para_client help sender
1469 -> Get help for a specific sender (contains further examples):
1471 s=http # or dccp or udp
1472 para_client sender $s help
1474 -> Show status of the http sender
1476 para_client sender http status
1478 By default para_server activates both the HTTP and th DCCP sender on
1479 startup. This can be changed via command line options or para_server's
1482 -> List config file options for senders:
1486 All senders share the "on" and "off" commands, so senders may be
1487 activated and deactivated independently of each other.
1489 -> Switch off the http sender:
1491 para_client sender http off
1493 -> Receive a DCCP stream using CCID2 and write the output into a file:
1495 host=foo.org; ccid=2; filename=bar
1496 para_recv --receiver "dccp --host $host --ccid $ccid" > $filename
1498 Note the quotes around the arguments for the dccp receiver. Each
1499 receiver has its own set of command line options and its own command
1500 line parser, so arguments for the dccp receiver must be protected
1501 from being interpreted by para_recv.
1503 -> Start UDP multicast, using the default multicast address:
1505 para_client sender udp add 224.0.1.38
1507 -> Receive FEC-encoded multicast stream and write the output into a file:
1510 para_recv -r udp > $filename
1512 -> Add an UDP unicast for a client to the target list of the UDP sender:
1515 para_client sender udp add $t
1517 -> Receive this (FEC-encoded) unicast stream:
1520 para_recv -r 'udp -i 0.0.0.0' > $filename
1522 -> Create a minimal config for para_audiod for HTTP streams:
1524 c=$HOME/.paraslash/audiod.conf.min; s=server.foo.com
1525 echo receiver \".:http -i $s\" > $c
1526 para_audiod --config $c
1532 A paraslash filter is a module which transforms an input stream into
1533 an output stream. Filters are included in the para_audiod executable
1534 and in the stand-alone tool para_filter which usually contains the
1537 While para_filter reads its input stream from STDIN and writes
1538 the output to STDOUT, the filter modules of para_audiod are always
1539 connected to a receiver which produces the input stream and a writer
1540 which absorbs the output stream.
1542 Some filters depend on a specific library and are not compiled in
1543 if this library was not found at compile time. To see the list of
1544 supported filters, run para_filter and para_audiod with the --help
1545 option. The output looks similar to the following:
1548 compress wav amp fecdec wmadec prebuffer oggdec aacdec mp3dec
1550 Out of these filter modules, a chain of filters can be constructed,
1551 much in the way Unix pipes can be chained, and analogous to the use
1552 of modules in gstreamer: The output of the first filter becomes the
1553 input of the second filter. There is no limitation on the number of
1554 filters and the same filter may occur more than once.
1556 Like receivers, each filter has its own command line options which
1557 must be quoted to protect them from the command line options of
1558 the driving application (para_audiod or para_filter). Example:
1560 para_filter -f 'mp3dec --ignore-crc' -f 'compress --damp 1'
1562 For para_audiod, each audio format has its own set of filters. The
1563 name of the audio format for which the filter should be applied can
1564 be used as the prefix for the filter option. Example:
1566 para_audiod -f 'mp3:prebuffer --duration 300'
1568 The "mp3" prefix above is actually interpreted as a POSIX extended
1569 regular expression. Therefore
1571 para_audiod -f '.:prebuffer --duration 300'
1573 activates the prebuffer filter for all supported audio formats (because
1574 "." matches all audio formats) while
1576 para_audiod -f 'wma|ogg:prebuffer --duration 300'
1578 activates it only for wma and ogg streams.
1583 For each supported audio format there is a corresponding filter
1584 which decodes audio data in this format to 16 bit PCM data which
1585 can be directly sent to the sound device or any other software that
1586 operates on undecoded PCM data (visualizers, equalizers etc.). Such
1587 filters are called _decoders_ in general, and xxxdec is the name of
1588 the paraslash decoder for the audio format xxx. For example, the mp3
1589 decoder filter is called mp3dec.
1591 Note that the output of the decoder is about 10 times larger than
1592 its input. This means that filters that operate on the decoded audio
1593 stream have to deal with much more data than filters that transform
1594 the audio stream before it is fed to the decoder.
1596 Paraslash relies on external libraries for most decoders, so these
1597 libraries must be installed for the decoder to be included in the
1598 para_filter and para_audiod executables. The oggdec filter depends
1599 on the libogg and libvorbis libraries for example.
1601 Forward error correction
1602 ~~~~~~~~~~~~~~~~~~~~~~~~
1604 As already mentioned REFERENCE(Streaming protocols, earlier),
1605 paraslash uses forward error correction (FEC) for the unreliable UDP
1606 and DCCP transports. FEC is a technique which was invented already
1607 in 1960 by Reed and Solomon and which is widely used for the parity
1608 calculations of storage devices (RAID arrays). It is based on the
1609 algebraic concept of finite fields, today called Galois fields, in
1610 honour of the mathematician Galois (1811-1832). The FEC implementation
1611 of paraslash is based on code by Luigi Rizzo.
1613 Although the details require a sound knowledge of the underlying
1614 mathematics, the basic idea is not hard to understand: For positive
1615 integers k and n with k < n it is possible to compute for any k given
1616 data bytes d_1, ..., d_k the corresponding r := n -k parity bytes p_1,
1617 ..., p_r such that all data bytes can be reconstructed from *any*
1620 {d_1, ..., d_k, p_1, ..., p_r}.
1622 FEC-encoding for unreliable network transports boils down to slicing
1623 the audio stream into groups of k suitably sized pieces called _slices_
1624 and computing the r corresponding parity slices. This step is performed
1625 in para_server which then sends both the data and the parity slices
1626 over the unreliable network connection. If the client was able
1627 to receive at least k of the n = k + r slices, it can reconstruct
1628 (FEC-decode) the original audio stream.
1630 From these observations it is clear that there are three different
1631 FEC parameters: The slice size, the number of data slices k, and the
1632 total number of slices n. It is crucial to choose the slice size
1633 such that no fragmentation of network packets takes place because
1634 FEC only guards against losses and reordering but fails if slices are
1637 FEC decoding in paralash is performed through the fecdec filter which
1638 usually is the first filter (there can be other filters before fecdec
1639 if these do not alter the audio stream).
1642 Volume adjustment (amp and compress)
1643 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1645 The amp and the compress filter both adjust the volume of the audio
1646 stream. These filters operate on uncompressed audio samples. Hence
1647 they are usually placed directly after the decoding filter. Each
1648 sample is multiplied with a scaling factor (>= 1) which makes amp
1649 and compress quite expensive in terms of computing power.
1653 The amp filter amplifies the audio stream by a fixed scaling factor
1654 that must be known in advance. For para_audiod this factor is derived
1655 from the amplification field of the audio file's entry in the audio
1656 file table while para_filter uses the value given at the command line.
1658 The optimal scaling factor F for an audio file is the largest real
1659 number F >= 1 such that after multiplication with F all samples still
1660 fit into the sample interval [-32768, 32767]. One can use para_filter
1661 in combination with the sox utility to compute F:
1663 para_filter -f mp3dec -f wav < file.mp3 | sox -t wav - -e stat -v
1665 The amplification value V which is stored in the audio file table,
1666 however, is an integer between 0 and 255 which is connected to F
1671 To store V in the audio file table, the command
1673 para_client -- touch -a=V file.mp3
1675 is used. The reader is encouraged to write a script that performs
1676 these computations :)
1680 Unlike the amplification filter, the compress filter adjusts the volume
1681 of the audio stream dynamically without prior knowledge about the peak
1682 value. It maintains the maximal volume of the last n samples of the
1683 audio stream and computes a suitable amplification factor based on that
1684 value and the various configuration options. It tries to chose this
1685 factor such that the adjusted volume meets the desired target level.
1687 Note that it makes sense to combine amp and compress.
1689 Misc filters (wav and prebuffer)
1690 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1692 These filters are rather simple and do not modify the audio stream at
1693 all. The wav filter is only useful with para_filter and in connection
1694 with a decoder. It asks the decoder for the number of channels and the
1695 sample rate of the stream and adds a Microsoft wave header containing
1696 this information at the beginning. This allows to write wav files
1697 rather than raw PCM files (which do not contain any information about
1698 the number of channels and the sample rate).
1700 The prebuffer filter simply delays the output until the given time has
1701 passed (starting from the time the first byte was available in its
1702 input queue) or until the given amount of data has accumulated. It
1703 is mainly useful for para_audiod if the standard parameters result
1704 in buffer underruns.
1706 Both filters require almost no additional computing time, even when
1707 operating on uncompressed audio streams, since data buffers are simply
1708 "pushed down" rather than copied.
1713 -> Decode an mp3 file to wav format:
1715 para_filter -f mp3dec -f wav < file.mp3 > file.wav
1717 -> Amplify a raw audio file by a factor of 1.5:
1719 para_filter -f amp --amp 32 < foo.raw > bar.raw
1725 Once an audio stream has been received and decoded to PCM format,
1726 it can be sent to a sound device for playback. This part is performed
1727 by paraslash _writers_ which are described in this chapter.
1732 A paraslash writer acts as a data sink that consumes but does not
1733 produce audio data. Paraslash writers operate on the client side and
1734 are contained in para_audiod and in the stand-alone tool para_write.
1736 The para_write program reads uncompressed audio data from STDIN. If
1737 this data starts with a wav header, sample rate, sample format and
1738 channel count are read from the header. Otherwise CD audio (44.1KHz
1739 16 bit little endian, stereo) is assumed but this can be overridden
1740 by command line options. para_audiod, on the other hand, obtains
1741 the sample rate and the number of channels from the decoder.
1743 Like receivers and filters, each writer has an individual set of
1744 command line options, and for para_audiod writers can be configured
1745 per audio format separately. It is possible to activate more than
1746 one writer for the same stream simultaneously.
1751 Unfortunately, the various flavours of Unix on which paraslash
1752 runs on have different APIs for opening a sound device and starting
1753 playback. Hence for each such API there is a paraslash writer that
1754 can play the audio stream via this API.
1756 *ALSA*. The _Advanced Linux Sound Architecture_ is only available on
1757 Linux systems. Although there are several mid-layer APIs in use by
1758 the various Linux distributions (ESD, Jack, PulseAudio), paraslash
1759 currently supports only the low-level ALSA API which is not supposed
1760 to be change. ALSA is very feature-rich, in particular it supports
1761 software mixing via its DMIX plugin. ALSA is the default writer on
1764 *OSS*. The _Open Sound System_ is the only API on *BSD Unixes and
1765 is also available on Linux systems, usually provided by ALSA as an
1766 emulation for backwards compatibility. This API is rather simple but
1767 also limited. For example only one application can open the device
1768 at any time. The OSS writer is activated by default on BSD Systems.
1770 *OSX*. Mac OS X has yet another API called CoreAudio. The OSX writer
1771 for this API is only compiled in on such systems and is of course
1774 *FILE*. The file writer allows to capture the audio stream and
1775 write the PCM data to a file on the file system rather than playing
1776 it through a sound device. It is supported on all platforms and is
1779 *AO*. _Libao_ is a cross-platform audio library which supports a wide
1780 variety of platforms including PulseAudio (gnome), ESD (Enlightened
1781 Sound Daemon), AIX, Solaris and IRIX. The ao writer plays audio
1782 through an output plugin of libao.
1787 -> Use the OSS writer to play a wav file:
1789 para_write --writer oss < file.wav
1791 -> Enable ALSA software mixing for mp3 streams
1793 para_audiod --writer 'mp3:alsa -d plug:swmix'
1800 para_gui executes an arbitrary command which is supposed to print
1801 status information to STDOUT. It then displays this information in
1802 a curses window. By default the command
1804 para_audioc -- stat -p
1806 is executed, but this can be customized via the --stat-cmd option. In
1807 particular it possible to use
1809 para_client -- stat -p
1811 to make para_gui work on systems on which para_audiod is not running.
1816 It is possible to bind keys to arbitrary commands via custom
1817 key-bindings. Besides the internal keys which can not be changed (help,
1818 quit, loglevel, version...), the following flavours of key-bindings
1821 - external: Shutdown curses before launching the given command.
1822 Useful for starting other ncurses programs from within
1823 para_gui, e.g. aumix or dialog scripts. Or, use the mbox
1824 output format to write a mailbox containing one mail for each
1825 (admissible) file the audio file selector knows about. Then
1826 start mutt from within para_gui to browse your collection!
1828 - display: Launch the command and display its stdout in
1829 para_gui's bottom window.
1831 - para: Like display, but start "para_client <specified
1832 command>" instead of "<specified command>".
1834 The general form of a key binding is
1838 which maps key k to command c using mode m. Mode may be x, d or p
1839 for external, display and paraslash commands, respectively.
1844 Currently there are only two themes for para_gui. It is easy, however,
1845 to add more themes. To create a new theme one has to define the
1846 position, color and geometry for for each status item that should be
1847 shown by this theme. See gui_theme.c for examples.
1849 The "." and "," keys are used to switch between themes.
1854 -> Show server info:
1858 -> Jump to the middle of the current audio file by pressing F5:
1860 key_map "<F5>:p:jmp 50"
1862 -> vi-like bindings for jumping around:
1865 key_map "h:p:ff 10-"
1867 key_map "b:p:ff 60-"
1869 -> Print the current date and time:
1873 -> Call other curses programs:
1876 key_map "!:x:/bin/bash"
1877 key_map "^E:x:/bin/sh -c 'vi ~/.paraslash/gui.conf'"
1886 In order to compile the sources from the git repository (rather than
1887 from tar balls) and for contributing non-trivial changes to the
1888 paraslash project, some additional tools should be installed on a
1891 http://git.or.cz/ (git). As described in more detail REFERENCE(Git
1892 branches, below), the git source code management tool is used for
1893 paraslash development. It is necessary for cloning the git repository
1894 and for getting updates.
1896 ftp://ftp.gnu.org/pub/gnu/m4/ (m4). Some input files for gengetopt
1897 are generated from templates by the m4 macro processor.
1899 ftp://ftp.gnu.org/pub/gnu/autoconf/ (autoconf) GNU autoconf creates
1900 the configure file which is shipped in the tarballs but has to be
1901 generated when compiling from git.
1903 http://www.triptico.com/software/grutatxt.html (grutatxt). The
1904 HTML version of this manual and some of the paraslash web pages are
1905 generated by the grutatxt plain text to HTML converter. If changes
1906 are made to these text files the grutatxt package must be installed
1907 to regenerate the HTML files.
1909 http://www.stack.nl/~dimitri/doxygen/ (doxygen). The documentation
1910 of paraslash's C sources uses the doxygen documentation system. The
1911 conventions for documenting the source code is described in the
1912 REFERENCE(Doxygen, Doxygen section).
1914 ftp://ftp.gnu.org/pub/gnu/global (global). This is used to generate
1915 browsable HTML from the C sources. It is needed by doxygen.
1920 Paraslash has been developed using the git source code management
1921 tool since 2006. Development is organized roughly in the same spirit
1922 as the git development itself, as described below.
1924 The following text passage is based on "A note from the maintainer",
1925 written by Junio C Hamano, the maintainer of git.
1927 There are four branches in the paraslash repository that track the
1928 source tree: "master", "maint", "next", and "pu".
1930 The "master" branch is meant to contain what is well tested and
1931 ready to be used in a production setting. There could occasionally be
1932 minor breakages or brown paper bag bugs but they are not expected to
1933 be anything major, and more importantly quickly and easily fixable.
1934 Every now and then, a "feature release" is cut from the tip of this
1935 branch, named with three dotted decimal digits, like 0.4.2.
1937 Whenever changes are about to be included that will eventually lead to
1938 a new major release (e.g. 0.5.0), a "maint" branch is forked off from
1939 "master" at that point. Obvious, safe and urgent fixes after the major
1940 release are applied to this branch and maintenance releases are cut
1941 from it. New features never go to this branch. This branch is also
1942 merged into "master" to propagate the fixes forward.
1944 A trivial and safe enhancement goes directly on top of "master".
1945 New development does not usually happen on "master", however.
1946 Instead, a separate topic branch is forked from the tip of "master",
1947 and it first is tested in isolation; Usually there are a handful such
1948 topic branches that are running ahead of "master". The tip of these
1949 branches is not published in the public repository to keep the number
1950 of branches that downstream developers need to worry about low.
1952 The quality of topic branches varies widely. Some of them start out as
1953 "good idea but obviously is broken in some areas" and then with some
1954 more work become "more or less done and can now be tested by wider
1955 audience". Luckily, most of them start out in the latter, better shape.
1957 The "next" branch is to merge and test topic branches in the latter
1958 category. In general, this branch always contains the tip of "master".
1959 It might not be quite rock-solid production ready, but is expected to
1960 work more or less without major breakage. The maintainer usually uses
1961 the "next" version of paraslash for his own pleasure, so it cannot
1962 be _that_ broken. The "next" branch is where new and exciting things
1965 The two branches "master" and "maint" are never rewound, and "next"
1966 usually will not be either (this automatically means the topics that
1967 have been merged into "next" are usually not rebased, and you can find
1968 the tip of topic branches you are interested in from the output of
1969 "git log next"). You should be able to safely build on top of them.
1971 However, at times "next" will be rebuilt from the tip of "master" to
1972 get rid of merge commits that will never be in "master". The commit
1973 that replaces "next" will usually have the identical tree, but it
1974 will have different ancestry from the tip of "master".
1976 The "pu" (proposed updates) branch bundles the remainder of the
1977 topic branches. The "pu" branch, and topic branches that are only in
1978 "pu", are subject to rebasing in general. By the above definition
1979 of how "next" works, you can tell that this branch will contain quite
1980 experimental and obviously broken stuff.
1982 When a topic that was in "pu" proves to be in testable shape, it
1983 graduates to "next". This is done with
1986 git merge that-topic-branch
1988 Sometimes, an idea that looked promising turns out to be not so good
1989 and the topic can be dropped from "pu" in such a case.
1991 A topic that is in "next" is expected to be polished to perfection
1992 before it is merged to "master". Similar to the above, this is
1996 git merge that-topic-branch
1997 git branch -d that-topic-branch
1999 Note that being in "next" is not a guarantee to appear in the next
2000 release (being in "master" is such a guarantee, unless it is later
2001 found seriously broken and reverted), nor even in any future release.
2006 The preferred coding style for paraslash coincides more or less
2007 with the style of the Linux kernel. So rather than repeating what is
2008 written XREFERENCE(http://www.kernel.org/doc/Documentation/CodingStyle,
2009 there), here are the most important points.
2011 - Burn the GNU coding standards.
2012 - Never use spaces for indentation.
2013 - Tabs are 8 characters, and thus indentations are also 8 characters.
2014 - Don't put multiple assignments on a single line.
2015 - Avoid tricky expressions.
2016 - Don't leave whitespace at the end of lines.
2017 - The limit on the length of lines is 80 columns.
2018 - Use K&R style for placing braces and spaces:
2024 - Use a space after (most) keywords.
2025 - Do not add spaces around (inside) parenthesized expressions.
2026 - Use one space around (on each side of) most binary and ternary operators.
2027 - Do not use cute names like ThisVariableIsATemporaryCounter, call it tmp.
2028 - Mixed-case names are frowned upon.
2029 - Descriptive names for global variables are a must.
2031 - Functions should be short and sweet, and do just one thing.
2032 - The number of local variables shouldn't exceed 10.
2033 - Gotos are fine if they improve readability and reduce nesting.
2034 - Don't use C99-style "// ..." comments.
2035 - Names of macros defining constants and labels in enums are capitalized.
2036 - Enums are preferred when defining several related constants.
2037 - Always use the paraslash wrappers for allocating memory.
2038 - If the name of a function is an action or an imperative.
2039 command, the function should return an error-code integer
2040 (<0 means error, >=0 means success). If the name is a
2041 predicate, the function should return a "succeeded" boolean.
2047 Doxygen is a documentation system for various programming
2048 languages. The paraslash project uses Doxygen for generating the API
2049 reference on the web pages, but good source code documentation is
2050 also beneficial to people trying to understand the code structure
2051 and the interactions between the various source files.
2053 It is more illustrative to look at the source code for examples than
2054 to describe the conventions for documenting the source in this manual,
2055 so we only describe which parts of the code need doxygen comments,
2056 but leave out details on documentation conventions.
2058 As a rule, only the public part of the C source is documented with
2059 Doxygen. This includes structures, defines and enumerations in header
2060 files as well as public (non-static) C functions. These should be
2061 documented completely. For example each parameter and the return
2062 value of a public function should get a descriptive comment.
2064 No doxygen comments are necessary for static functions and for
2065 structures and enumerations in C files (which are used only within
2066 this file). This does not mean, however, that those entities need
2067 no documentation at all. Instead, common sense should be applied to
2068 document what is not obvious from reading the code.
2077 *IP*. The _Internet Protocol_ is the primary networking protocol
2078 used for the Internet. All protocols described below use IP as the
2079 underlying layer. Both the prevalent IPv4 and the next-generation
2080 IPv6 variant are being deployed actively worldwide.
2082 *Connection-oriented and connectionless protocols*. Connectionless
2083 protocols differ from connection-oriented ones in that state
2084 associated with the sending/receiving endpoints is treated
2085 implicitly. Connectionless protocols maintain no internal knowledge
2086 about the state of the connection. Hence they are not capable of
2087 reacting to state changes, such as sudden loss or congestion on the
2088 connection medium. Connection-oriented protocols, in contrast, make
2089 this knowledge explicit. The connection is established only after
2090 a bidirectional handshake which requires both endpoints to agree
2091 on the state of the connection, and may also involve negotiating
2092 specific parameters for the particular connection. Maintaining an
2093 up-to-date internal state of the connection also in general means
2094 that the sending endpoints perform congestion control, adapting to
2095 qualitative changes of the connection medium.
2097 *Reliability*. In IP networking, packets can be lost, duplicated,
2098 or delivered out of order, and different network protocols handle
2099 these problems in different ways. We call a transport-layer protocol
2100 _reliable_, if it turns the unreliable IP delivery into an ordered,
2101 duplicate- and loss-free delivery of packets. Sequence numbers
2102 are used to discard duplicates and re-arrange packets delivered
2103 out-of-order. Retransmission is used to guarantee loss-free
2104 delivery. Unreliable protocols, in contrast, do not guarantee ordering
2107 *Classification*. With these definitions the protocols which are used
2108 by paraslash for steaming audio data may be classified as follows.
2110 - HTTP/TCP: connection-oriented, reliable,
2111 - UDP: connectionless, unreliable,
2112 - DCCP: connection-oriented, unreliable.
2114 Below we give a short descriptions of these protocols.
2116 *TCP*. The _Transmission Control Protocol_ provides reliable,
2117 ordered delivery of a stream and a classic window-based congestion
2118 control. In contrast to UDP and DCCP (see below), TCP does not have
2119 record-oriented or datagram-based syntax, i.e. it provides a stream
2120 which is unaware and independent of any record (packet) boundaries.
2121 TCP is used extensively by many application layers. Besides HTTP (the
2122 Hypertext Transfer Protocol), also FTP (the File Transfer protocol),
2123 SMTP (Simple Mail Transfer Protocol), SSH (Secure Shell) all sit on
2126 *UDP*. The _User Datagram Protocol_ is the simplest transport-layer
2127 protocol, built as a thin layer directly on top of IP. For this reason,
2128 it offers the same best-effort service as IP itself, i.e. there is no
2129 detection of duplicate or reordered packets. Being a connectionless
2130 protocol, only minimal internal state about the connection is
2131 maintained, which means that there is no protection against packet
2132 loss or network congestion. Error checking and correction (if at all)
2133 are performed in the application.
2135 *DCCP*. The _Datagram Congestion Control Protocol_ combines the
2136 connection-oriented state maintenance known from TCP with the
2137 unreliable, datagram-based transport of UDP. This means that it
2138 is capable of reacting to changes in the connection by performing
2139 congestion control, offering multiple alternative approaches. But it
2140 is bound to datagram boundaries (the maximum packet size supported
2141 by a medium), and like UDP it lacks retransmission to protect
2142 against loss. Due to the use of sequence numbers, it is however
2143 able to react to loss (interpreted as a congestion indication) and
2144 to ignore out-of-order and duplicate packets. Unlike TCP it allows
2145 to negotiate specific, binding features for a connection, such as
2146 the choice of congestion control: classic, window-based congestion
2147 control known from TCP is available as CCID-2, rate-based, "smooth"
2148 congestion control is offered as CCID-3.
2150 *HTTP*. _The Hypertext Transfer Protocol_ is an application layer
2151 protocol on top of TCP. It is spoken by web servers and is most often
2152 used for web services. However, as can be seen by the many Internet
2153 radio stations and YouTube/Flash videos, http is by far not limited to
2154 the delivery of web pages only. Being a simple request/response based
2155 protocol, the semantics of the protocol also allow the delivery of
2156 multimedia content, such as audio over http.
2158 *Multicast*. IP multicast is not really a protocol but a technique
2159 for one-to-many communication over an IP network. The challenge is to
2160 deliver information to a group of destinations simultaneously using
2161 the most efficient strategy to send the messages over each link of
2162 the network only once. This has benefits for streaming multimedia:
2163 the standard one-to-one unicast offered by TCP/DCCP means that
2164 n clients listening to the same stream also consume n-times the
2165 resources, whereas multicast requires to send the stream just once,
2166 irrespective of the number of receivers. Since it would be costly to
2167 maintain state for each listening receiver, multicast often implies
2168 connectionless transport, which is the reason that it is currently
2169 only available via UDP.
2174 Paraslash is licensed under the GPL, version 2. Most of the code
2175 base has been written from scratch, and those parts are GPL V2
2176 throughout. Notable exceptions are FEC and the WMA decoder. See the
2177 corresponding source files for licencing details for these parts. Some
2178 code sniplets of several other third party software packages have
2179 been incorporated into the paraslash sources, for example log message
2180 coloring was taken from the git sources. These third party software
2181 packages are all published under the GPL or some other license
2182 compatible to the GPL.
2187 Many thanks to Gerrit Renker who read an early draft of this manual
2188 and contributed significant improvements.
2196 - Reed, Irving S.; Solomon, Gustave (1960),
2197 XREFERENCE(http://kom.aau.dk/~heb/kurser/NOTER/KOFA01.PDF,
2198 Polynomial Codes over Certain Finite Fields), Journal of the
2199 Society for Industrial and Applied Mathematics (SIAM) 8 (2):
2200 300-304, doi:10.1137/0108018)
2205 - XREFERENCE(http://www.ietf.org/rfc/rfc768.txt, RFC 768) (1980):
2206 User Datagram Protocol
2207 - XREFERENCE(http://www.ietf.org/rfc/rfc791.txt, RFC 791) (1981):
2209 - XREFERENCE(http://www.ietf.org/rfc/rfc2437.txt, RFC 2437) (1998):
2210 RSA Cryptography Specifications
2211 - XREFERENCE(http://www.ietf.org/rfc/rfc4340.txt, RFC 4340)
2212 (2006): Datagram Congestion Control Protocol (DCCP)
2213 - XREFERENCE(http://www.ietf.org/rfc/rfc4341.txt, RFC 4341) (2006):
2214 Congestion Control ID 2: TCP-like Congestion Control
2215 - XREFERENCE(http://www.ietf.org/rfc/rfc4342.txt, RFC 4342) (2006):
2216 Congestion Control ID 3: TCP-Friendly Rate Control (TFRC)
2217 - XREFERENCE(http://www.ietf.org/rfc/rfc6716.txt, RFC 6716) (2012):
2218 Definition of the Opus Audio Codec
2220 Application web pages
2221 ~~~~~~~~~~~~~~~~~~~~~
2223 - XREFERENCE(http://people.tuebingen.mpg.de/maan/paraslash/, paraslash)
2224 - XREFERENCE(http://paraslash.systemlinux.org/, paraslash (alternative page))
2225 - XREFERENCE(http://xmms2.org/wiki/Main_Page, xmms)
2226 - XREFERENCE(http://www.mpg123.de/, mpg123)
2227 - XREFERENCE(http://gstreamer.freedesktop.org/, gstreamer)
2228 - XREFERENCE(http://www.icecast.org/, icecast)
2229 - XREFERENCE(http://beesbuzz.biz/code/audiocompress.php, Audio Compress)
2231 External documentation
2232 ~~~~~~~~~~~~~~~~~~~~~~
2234 - XREFERENCE(http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf,
2235 H. Peter Anvin: The mathematics of Raid6)
2236 - XREFERENCE(http://info.iet.unipi.it/~luigi/fec_ccr.ps.gz,
2237 Luigi Rizzo: Effective Erasure Codes for reliable Computer
2238 Communication Protocols)
2242 - XREFERENCE(http://info.iet.unipi.it/~luigi/vdm.tar.gz,
2243 Original FEC implementation by Luigi Rizzo)