crypt: Remove RC4 support.
[paraslash.git] / web / manual.md
1 **Paraslash user manual**
2
3 This document describes how to install, configure and use the paraslash
4 network audio streaming system. Most chapters start with a chapter
5 overview and conclude with an example section. We try to focus on
6 general concepts and on the interaction of the various pieces of the
7 paraslash package. Hence this user manual is not meant as a replacement
8 for the manual pages that describe all command line options of each
9 paraslash executable.
10
11 ============
12 Introduction
13 ============
14
15 In this chapter we give an [overview](#Overview) of the interactions of
16 the two main programs contained in the paraslash package, followed by
17 [brief descriptions](#The.paraslash.executables) of all executables.
18
19 Overview
20 --------
21
22 The core functionality of the para suite is provided by two main
23 executables, para_server and para_audiod. The former maintains a
24 database of audio files and streams these files to para_audiod which
25 receives and plays the stream.
26
27 In a typical setting, both para_server and para_audiod act as
28 background daemons whose functionality is controlled by client
29 programs: the para_audioc client controls para_audiod over a local
30 socket while the para_client program connects to para_server over a
31 local or remote networking connection.
32
33 Typically, these two daemons run on different hosts but a local setup
34 is also possible.
35
36 A simplified picture of a typical setup is as follows
37
38 server_host client_host
39 ~~~~~~~~~~~ ~~~~~~~~~~~
40
41 +-----------+ audio stream +-----------+
42 |para_server| -----------------------------> |para_audiod|
43 +-----------+ +-----------+
44 ^ ^
45 | |
46 | | connect
47 | |
48 | |
49 | +-----------+
50 | |para_audioc|
51 | +-----------+
52 |
53 |
54 | connect +-----------+
55 +-------------------------------------- |para_client|
56 +-----------+
57 The paraslash executables
58 -------------------------
59
60 ### para_server ###
61
62 para_server streams binary audio data (MP3, ...) over local and/or
63 remote networks. It listens on a TCP port and accepts commands such
64 as play, stop, pause, next from authenticated clients. There are
65 many more commands though, see the man page of para_server for a
66 description of all commands.
67
68 It supports three built-in network streaming protocols
69 (senders/receivers): HTTP, DCCP, or UDP. This is explained in more
70 detail in the section on [networking](#Networking).
71
72 The built-in audio file selector of paraslash is used to manage your
73 audio files. It maintains statistics on the usage of all available
74 audio files such as last-played time, and the number of times each
75 file was selected.
76
77 Additional information may be added to the database to allow
78 fine-grained selection based on various properties of the audio file,
79 including information found in (ID3) tags. However, old-fashioned
80 playlists are also supported.
81
82 It is also possible to store images (album covers) and lyrics in the
83 database and associate these to the corresponding audio files.
84
85 The section on the [audio file selector](#The.audio.file.selector)
86 discusses this topic.
87
88
89 ### para_client ###
90
91 The client program to connect to para_server. paraslash commands
92 are sent to para_server and the response is dumped to STDOUT. This
93 can be used by any scripting language to produce user interfaces with
94 little programming effort.
95
96 All connections between para_server and para_client are encrypted
97 with a symmetric session key. For each user of paraslash you must
98 create a public/secret RSA key pair for authentication.
99
100 If para_client is started without non-option arguments, an interactive
101 session (shell) is started. Command history and command completion are
102 supported through libreadline.
103
104 ### para_audiod ###
105
106 The local daemon that collects information from para_server.
107
108 It runs on the client side and connects to para_server. As soon as
109 para_server announces the availability of an audio stream, para_audiod
110 starts an appropriate receiver, any number of filters and a paraslash
111 writer to play the stream.
112
113 Moreover, para_audiod listens on a local socket and sends status
114 information about para_server and para_audiod to local clients on
115 request. Access via this local socket may be restricted by using Unix
116 socket credentials, if available.
117
118
119 ### para_audioc ###
120
121 The client program which talks to para_audiod. Used to control
122 para_audiod, to receive status info, or to grab the stream at any
123 point of the decoding process. Like para_client, para_audioc supports
124 interactive sessions on systems with libreadline.
125
126 ### para_recv ###
127
128 A command line HTTP/DCCP/UDP stream grabber. The http mode is
129 compatible with arbitrary HTTP streaming sources (e.g. icecast).
130 In addition to the three network streaming modes, para_recv can also
131 operate in local (afh) mode. In this mode it writes the content of
132 an audio file on the local file system in complete chunks to stdout,
133 optionally 'just in time'. This allows to cut an audio file without
134 first decoding it, and it enables third-party software which is unaware
135 of the particular audio format to send complete frames in real time.
136
137 ### para_filter ###
138
139 A filter program that reads from STDIN and writes to STDOUT.
140 Like para_recv, this is an atomic building block which can be used to
141 assemble higher-level audio receiving facilities. It combines several
142 different functionalities in one tool: decoders for multiple audio
143 formats and a number of processing filters, among these a normalizer
144 for audio volume.
145
146 ### para_afh ###
147
148 A small stand-alone program that prints tech info about the given
149 audio file to STDOUT. It can be instructed to print a "chunk table",
150 an array of offsets within the audio file.
151
152 ### para_write ###
153
154 A modular audio stream writer. It supports a simple file writer
155 output plug-in and optional WAV/raw players for ALSA (Linux) and for
156 coreaudio (Mac OS). para_write can also be used as a stand-alone WAV
157 or raw audio player.
158
159 ### para_play ###
160
161 A command line audio player.
162
163 ### para_gui ###
164
165 Curses-based gui that presents status information obtained in a curses
166 window. Appearance can be customized via themes. para_gui provides
167 key-bindings for the most common server commands and new key-bindings
168 can be added easily.
169
170 ### para_fade ###
171
172 An alarm clock and volume-fader for OSS and ALSA.
173
174 ===========
175 Quick start
176 ===========
177
178 This chapter lists the [necessary software](#Requirements)
179 that must be installed to compile the paraslash package, describes
180 how to [compile and install](#Installation) the paraslash
181 source code and the steps that have to be performed in order to
182 [set up](#Configuration) a typical server and client.
183
184 Requirements
185 ------------
186 ### For the impatient ###
187
188 git clone git://git.tuebingen.mpg.de/osl
189 cd osl && make && sudo make install && sudo ldconfig
190 sudo apt-get install autoconf libssl-dev help2man gengetopt m4 \
191 libmad0-dev libid3tag0-dev libasound2-dev libvorbis-dev \
192 libfaad-dev libspeex-dev libFLAC-dev libsamplerate-dev realpath \
193 libasound2-dev libao-dev libreadline-dev libncurses-dev \
194 libopus-dev
195
196 ### Detailed description ###
197
198 In any case you will need
199
200 - [libosl](http://people.tuebingen.mpg.de/maan/osl/). The _object
201 storage layer_ library is used by para_server. To clone the source
202 code repository, execute
203
204 git clone git://git.tuebingen.mpg.de/osl
205
206 - [gcc](ftp://ftp.gnu.org/pub/gnu/gcc) or
207 [clang](http://clang.llvm.org). All gcc versions >= 4.2 are currently
208 supported. Clang version 1.1 or newer should work as well.
209
210 - [gnu make](ftp://ftp.gnu.org/pub/gnu/make) is also shipped with the
211 disto. On BSD systems the gnu make executable is often called gmake.
212
213 - [bash](ftp://ftp.gnu.org/pub/gnu/bash). Some scripts which run
214 during compilation require the _Bourne again shell_. It is most
215 likely already installed.
216
217 - [gengetopt](ftp://ftp.gnu.org/pub/gnu/gengetopt/) is needed to
218 generate the C code for the command line parsers of all paraslash
219 executables.
220
221 - [help2man](ftp://ftp.gnu.org/pub/gnu/help2man) is used to create
222 the man pages.
223
224 - [m4](ftp://ftp.gnu.org/pub/gnu/m4/). Some source files are generated
225 from templates by the m4 macro processor.
226
227 Optional:
228
229 - [openssl](http://www.openssl.org/) or
230 [libgcrypt](ftp://ftp.gnupg.org/gcrypt/libgcrypt/). At least one
231 of these two libraries is needed as the backend for cryptographic
232 routines on both the server and the client side. Both openssl and
233 libgcrypt are usually shipped with the distro, but you might have
234 to install the development package (`libssl-dev` or `libgcrypt-dev`
235 on debian systems) as well.
236
237 - [libmad](http://www.underbit.com/products/mad/). To compile in MP3
238 support for paraslash, the development package must be installed. It
239 is called `libmad0-dev` on debian-based systems. Note that libmad is
240 not necessary on the server side, i.e., for sending MP3 files.
241
242 - [libid3tag](http://www.underbit.com/products/mad/). For version-2
243 ID3 tag support, you willl need the libid3tag development package
244 `libid3tag0-dev`. Without libid3tag, only version-1 tags are
245 recognized. The mp3 tagger also needs this library for modifying
246 (id3v1 and id3v2) tags.
247
248 - [ogg vorbis](http://www.xiph.org/downloads/). For ogg vorbis streams
249 you need libogg, libvorbis, libvorbisfile. The corresponding Debian
250 packages are called `libogg-dev` and `libvorbis-dev`.
251
252 - [libfaad](http://www.audiocoding.com/). For aac files (m4a) you
253 need libfaad (`libfaad-dev`).
254
255 - [speex](http://www.speex.org/). In order to stream or decode speex
256 files, libspeex (`libspeex-dev`) is required.
257
258 - [flac](http://flac.sourceforge.net/). To stream or decode files
259 encoded with the _Free Lossless Audio Codec_, libFLAC (`libFLAC-dev`)
260 must be installed.
261
262 - [libsamplerate](http://www.mega-nerd.com/SRC/index.html). The
263 resample filter will only be compiled if this library is
264 installed. Debian package: `libsamplerate-dev`.
265
266 - [alsa-lib](ftp://ftp.alsa-project.org/pub/lib/). On Linux, you will
267 need to have the ALSA development package `libasound2-dev` installed.
268
269 - [libao](http://downloads.xiph.org/releases/ao/). Needed to build
270 the ao writer (ESD, PulseAudio,...). Debian package: `libao-dev`.
271
272 - [curses](ftp://ftp.gnu.org/pub/gnu/ncurses). Needed for
273 para_gui. Debian package: `libncurses-dev`.
274
275 - [GNU
276 Readline](http://cnswww.cns.cwru.edu/php/chet/readline/rltop.html). If
277 this library (`libreadline-dev`) is installed, para_client, para_audioc
278 and para_play support interactive sessions.
279
280 Installation
281 ------------
282 To build the sources from a tarball, execute
283
284 ./configure && make
285
286 To build from git or a gitweb snapshot, run this command instead:
287
288 ./autogen.sh
289
290 There should be no errors but probably some warnings about missing
291 packages which usually implies that not all audio formats will be
292 supported. If headers or libs are installed at unusual locations you
293 might need to tell the configure script where to find them. Try
294
295 ./configure --help
296
297 to see a list of options. If the paraslash package was compiled
298 successfully, execute (optionally)
299
300 make test
301
302 to run the paraslash test suite. If all tests pass, execute as root
303
304 make install
305
306 to install executables under /usr/local/bin and the man pages under
307 /usr/local/man.
308
309 Configuration
310 -------------
311
312 ### Create a paraslash user ###
313
314 In order to control para_server at runtime you must create a paraslash
315 user. As authentication is based on the RSA crypto system you'll have
316 to create an RSA key pair. If you already have a user and an RSA key
317 pair, you may skip this step.
318
319 In this section we'll assume a typical setup: You would like to run
320 para_server on some host called server_host as user foo, and you want
321 to connect to para_server from another machine called client_host as
322 user bar.
323
324 As foo@server_host, create ~/.paraslash/server.users by typing the
325 following commands:
326
327 user=bar
328 target=~/.paraslash/server.users
329 key=~/.paraslash/id_rsa.pub.$user
330 perms=AFS_READ,AFS_WRITE,VSS_READ,VSS_WRITE
331 mkdir -p ~/.paraslash
332 echo "user $user $key $perms" >> $target
333
334 Next, change to the "bar" account on client_host and generate the
335 key pair with the commands
336
337 ssh-keygen -q -t rsa -b 2048 -N '' -f $key
338
339 This generates the two files id_rsa and id_rsa.pub in ~/.ssh. Note
340 that para_server won't accept keys shorter than 2048 bits. Moreover,
341 para_client rejects private keys which are world-readable.
342
343 para_server only needs to know the public key of the key pair just
344 created. Copy this public key to server_host:
345
346 src=~/.ssh/id_rsa.pub
347 dest=.paraslash/id_rsa.pub.$LOGNAME
348 scp $src foo@server_host:$dest
349
350 Finally, tell para_client to connect to server_host:
351
352 conf=~/.paraslash/client.conf
353 echo 'hostname server_host' > $conf
354
355
356 ### Start para_server ###
357
358 For this first try, we'll use the info loglevel to make the output
359 of para_server more verbose.
360
361 para_server -l info
362
363 Now you can use para_client to connect to the server and issue
364 commands. Open a new shell as bar@client_host and try
365
366 para_client help
367 para_client si
368
369 to retrieve the list of available commands and some server info.
370 Don't proceed if this doesn't work.
371
372 ### Create and populate the database ###
373
374 An empty database is created with
375
376 para_client init
377
378 This initializes a couple of empty tables under
379 ~/.paraslash/afs_database-0.4. You normally don't need to look at these
380 tables, but it's good to know that you can start from scratch with
381
382 rm -rf ~/.paraslash/afs_database-0.4
383
384 in case something went wrong.
385
386 Next, you need to add some audio files to that database so that
387 para_server knows about them. Choose an absolute path to a directory
388 containing some audio files and add them to the audio file table:
389
390 para_client add /my/mp3/dir
391
392 This might take a while, so it is a good idea to start with a directory
393 containing not too many files. Note that the table only contains data
394 about the audio files found, not the files themselves.
395
396 You may print the list of all known audio files with
397
398 para_client ls
399
400 ### Configure para_audiod ###
401
402 We will have to tell para_audiod that it should receive the audio
403 stream from server_host via http:
404
405 para_audiod -l info -r '.:http -i server_host'
406
407 You should now be able to listen to the audio stream once para_server
408 starts streaming. To activate streaming, execute
409
410 para_client play
411
412 Since no playlist has been specified yet, the "dummy" mode which
413 selects all known audio files is activated automatically. See the
414 section on the [audio file selector](#The.audio.file.selector) for how
415 to use playlists and moods to specify which files should be streamed
416 in which order.
417
418 Troubleshooting
419 ---------------
420
421 If you receive a socket related error on server or audiod startup,
422 make sure you have write permissions to the /var/paraslash directory:
423
424 sudo chown $LOGNAME /var/paraslash
425
426 Alternatively, use the --afs-socket (para_server) or --socket
427 (para_audiod) option to specify a different socket pathname.
428
429 To identify streaming problems try to receive, decode and play the
430 stream manually using para_recv, para_filter and para_write as follows.
431 For simplicity we assume that you're running Linux/ALSA and that only
432 MP3 files have been added to the database.
433
434 para_recv -r 'http -i server_host' > file.mp3
435 # (interrupt with CTRL+C after a few seconds)
436 ls -l file.mp3 # should not be empty
437 para_filter -f mp3dec -f wav < file.mp3 > file.wav
438 ls -l file.wav # should be much bigger than file.mp3
439 para_write -w alsa < file.wav
440
441 Double check what is logged by para_server and use the --loglevel
442 option of para_recv, para_filter and para_write to increase verbosity.
443
444 ===============
445 User management
446 ===============
447
448 para_server uses a challenge-response mechanism to authenticate
449 requests from incoming connections, similar to ssh's public key
450 authentication method. Authenticated connections are encrypted using
451 the AES stream cipher in integer counter mode.
452
453 In this chapter we briefly describe RSA and AES, and sketch the
454 [authentication handshake](#Client-server.authentication)
455 between para_client and para_server. User management is discussed
456 in the section on [the user_list file](#The.user_list.file).
457 These sections are all about communication between the client and the
458 server. Connecting para_audiod is a different matter and is described
459 in a [separate section](#Connecting.para_audiod).
460
461 RSA and AES
462 -----------
463
464 A block cipher is a transformation which operates on fixed-length
465 blocks. For symmetric block ciphers the transformation is determined
466 by a single key for both encryption and decryption. For asymmetric
467 block ciphers, on the other hand, the key consists of two parts,
468 called the public key and the private key. A message can be encrypted
469 with either key and only the counterpart of that key can decrypt the
470 message. Asymmetric block ciphers can be used for both signing and
471 encrypting a message.
472
473 RSA is an asymmetric block cipher which is used in many applications,
474 including ssh and gpg. The RSA public key encryption and signatures
475 algorithms are defined in detail in RFC 2437. Paraslash relies on
476 RSA for authentication.
477
478 Stream ciphers XOR the input with a pseudo-random key stream to produce
479 the output. Decryption uses the same function calls as encryption.
480 Any block cipher can be turned into a stream cipher by generating the
481 pseudo-random key stream by encrypting successive values of a counter
482 (counter mode).
483
484 AES, the advanced encryption standard, is a well-known symmetric block
485 cipher. Paraslash employs AES in counter mode as described above to
486 encrypt communications. Since a stream cipher key must not be used
487 twice, a random key is generated for every new connection.
488
489 Client-server authentication
490 ----------------------------
491
492 The authentication handshake between para_client and para_server goes
493 as follows:
494
495 - para_client connects to para_server and sends an authentication
496 request for a user. It does so by connecting to TCP port 2990 of the
497 server host. This port is called the para_server _control port_.
498
499 - para_server accepts the connection and forks a child process which
500 handles the incoming request. The parent process keeps listening on the
501 control port while the child process (also called para_server below)
502 continues as follows.
503
504 - para_server loads the RSA public key of that user, fills a
505 fixed-length buffer with random bytes, encrypts that buffer using the
506 public key and sends the encrypted buffer to the client. The first
507 part of the buffer is the challenge which is used for authentication
508 while the second part is the session key.
509
510 - para_client receives the encrypted buffer and decrypts it with the
511 user's private key, thereby obtaining the challenge buffer and the
512 session key. It sends the SHA1 hash value of the challenge back to
513 para_server and stores the session key for further use.
514
515 - para_server also computes the SHA1 hash of the challenge and compares
516 it against what was sent back by the client.
517
518 - If the two hashes do not match, the authentication has failed and
519 para_server closes the connection.
520
521 - Otherwise the user is considered authenticated and the client is
522 allowed to proceed by sending a command to be executed. From this
523 point on the communication is encrypted using the stream cipher with
524 the session key known to both peers.
525
526 paraslash relies on the quality of the pseudo-random bytes provided
527 by the crypto library (openssl or libgcrypt), on the security of
528 the implementation of the RSA and AES crypto routines and on the
529 infeasibility to invert the SHA1 function.
530
531 Neither para_server or para_client create RSA keys on their
532 own. This has to be done once for each user as sketched in
533 [Quick start](#Quick.start) and discussed in more detail
534 [below](#The.user_list.file).
535
536 The user_list file
537 ------------------
538
539 At startup para_server reads the user list file which contains one
540 line per user. The default location of the user list file may be
541 changed with the --user-list option.
542
543 There should be at least one user in this file. Each user must have
544 an RSA key pair. The public part of the key is needed by para_server
545 while the private key is needed by para_client. Each line of the
546 user list file must be of the form
547
548 user <username> <key> <perms>
549
550 where _username_ is an arbitrary string (usually the user's login
551 name), _key_ is the full path to that user's public RSA key, and
552 _perms_ is a comma-separated list of zero or more of the following
553 permission bits:
554
555 +---------------------------------------------------------+
556 | AFS_READ | read the contents of the databases |
557 +-----------+---------------------------------------------+
558 | AFS_WRITE | change database contents |
559 +-----------+---------------------------------------------+
560 | VSS_READ | obtain information about the current stream |
561 +-----------+---------------------------------------------+
562 | VSS_WRITE | change the current stream |
563 +---------------------------------------------------------+
564
565 The permission bits specify which commands the user is allowed to
566 execute. The output of
567
568 para_client help
569
570 contains in the third column the permissions needed to execute the
571 command.
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 Connecting para_audiod
578 ----------------------
579
580 para_audiod listens on a Unix domain socket. Those sockets are
581 for local communication only, so only local users can connect to
582 para_audiod. The default is to let any user connect but this can be
583 restricted on platforms that support UNIX socket credentials which
584 allow para_audiod to obtain the Unix credentials of the connecting
585 process.
586
587 Use para_audiod's --user-allow option to allow connections only for
588 a limited set of users.
589
590 =======================
591 The audio file selector
592 =======================
593
594 paraslash comes with a sophisticated audio file selector (AFS),
595 whose main task is to determine which file to stream next, based on
596 information on the audio files stored in a database. It communicates
597 also with para_client whenever an AFS command is executed, for example
598 to answer a database query.
599
600 Besides the traditional playlists, AFS supports audio file selection
601 based on _moods_ which act as a filter that limits the set of all
602 known audio files to those which satisfy certain criteria. It also
603 maintains tables containing images (e.g. album cover art) and lyrics
604 that can be associated with one or more audio files.
605
606 AFS employs [libosl](http://people.tuebingen.mpg.de/maan/osl/), the
607 object storage layer library, as the backend library for storing
608 information on audio files, playlists, etc. This library offers
609 functionality similar to a relational database, but is much more
610 lightweight than a full database backend.
611
612 In this chapter we sketch the setup of the [AFS
613 process](#The.AFS.process) during server startup and proceed with the
614 description of the [layout](#Database.layout) of the various database
615 tables. The section on [playlists and moods](#Playlists.and.moods)
616 explains these two audio file selection mechanisms in detail
617 and contains pratical examples. The way [file renames and content
618 changes](#File.renames.and.content.changes) are detected is discussed
619 briefly before the [Troubleshooting](#Troubleshooting) section
620 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 once
651 when the file is added to the database. Whenever AFS selects this
652 audio file for streaming the hash value is recomputed and checked
653 against the value stored in the database to detect content changes.
654
655 - The time when this audio file was last played.
656
657 - The number of times the file has been played so far.
658
659 - The attribute bitmask.
660
661 - The image id which describes the image associated with this audio
662 file.
663
664 - The lyrics id which describes the lyrics associated with this
665 audio file.
666
667 - The audio format id (MP3, OGG, ...).
668
669 - An amplification value that can be used by the amplification filter
670 to pre-amplify the decoded audio stream.
671
672 - The chunk table. It describes the location and the timing of the
673 building blocks of the audio file. This is used by para_server to
674 send chunks of the file at appropriate times.
675
676 - The duration of the audio file.
677
678 - Tag information contained in the audio file (ID3 tags, Vorbis
679 comments, ...).
680
681 - The number of channels
682
683 - The encoding bitrate.
684
685 - The sampling frequency.
686
687 To add or refresh the data contained in the audio file table, the _add_
688 command is used. It takes the full path of either an audio file or a
689 directory. In the latter case, the directory is traversed recursively
690 and all files which are recognized as valid audio files are added to
691 the database.
692
693 ### The attribute table ###
694
695 The attribute table contains two columns, _name_ and _bitnum_. An
696 attribute is simply a name for a certain bit number in the attribute
697 bitmask of the audio file table.
698
699 Each of the 64 bits of the attribute bitmask can be set for each
700 audio file individually. Hence up to 64 different attributes may be
701 defined. For example, "pop", "rock", "blues", "jazz", "instrumental",
702 "german_lyrics", "speech", whatever. You are free to choose as
703 many attributes as you like and there are no naming restrictions
704 for attributes.
705
706 A new attribute "test" is created by
707
708 para_client addatt test
709 and
710 para_client lsatt
711
712 lists all available attributes. You can set the "test" attribute for
713 an audio file by executing
714
715 para_client setatt test+ /path/to/the/audio/file
716
717 Similarly, the "test" bit can be removed from an audio file with
718
719 para_client setatt test- /path/to/the/audio/file
720
721 Instead of a path you may use a shell wildcard pattern. The attribute
722 is applied to all audio files matching this pattern:
723
724 para_client setatt test+ '/test/directory/*'
725
726 The command
727
728 para_client -- ls -l=v
729
730 gives you a verbose listing of your audio files also showing which
731 attributes are set.
732
733 In case you wonder why the double-dash in the above command is needed:
734 It tells para_client to not interpret the options after the dashes. If
735 you find this annoying, just say
736
737 alias para='para_client --'
738
739 and be happy. In what follows we shall use this alias.
740
741 The "test" attribute can be dropped from the database with
742
743 para rmatt test
744
745 Read the output of
746
747 para help ls
748 para help setatt
749
750 for more information and a complete list of command line options to
751 these commands.
752
753 ### Blob tables ###
754
755 The image, lyrics, moods and playlists tables are all blob tables.
756 Blob tables consist of three columns each: The identifier which is
757 a positive number that is auto-incremented, the name (an arbitrary
758 string) and the content (the blob).
759
760 All blob tables support the same set of actions: cat, ls, mv, rm
761 and add. Of course, _add_ is used for adding new blobs to the table
762 while the other actions have the same meaning as the corresponding
763 Unix commands. The paraslash commands to perform these actions are
764 constructed as the concatenation of the table name and the action. For
765 example addimg, catimg, lsimg, mvimg, rmimg are the commands that
766 manipulate or query the image table.
767
768 The add variant of these commands is special as these commands read
769 the blob contents from stdin. To add an image to the image table the
770 command
771
772 para addimg image_name < file.jpg
773
774 can be used.
775
776 Note that the images and lyrics are not interpreted at all, and also
777 the playlist and the mood blobs are only investigated when the mood
778 or playlist is activated with the select command.
779
780 ### The score table ###
781
782 The score table describes those audio files which are admissible for
783 the current mood or playlist (see below). The table has two columns:
784 a pointer to a row of the audio file table and a score value.
785
786 Unlike all other tables of the database, the score table remains in
787 memory and is never stored on disk. It is initialized at startup and
788 recomputed when the select command loads a new mood or playlist.
789
790 When the audio file selector is asked to open the next audio file,
791 it picks the row with the highest score, opens the corresponding
792 file and passes the file descriptor to the virtual streaming system.
793 At this point the last_played and the num_played fields of the selected
794 file are updated and the score is recomputed.
795
796 Playlists and moods
797 -------------------
798
799 Playlists and moods offer two different ways of specifying the set of
800 admissible files. A playlist in itself describes a set of admissible
801 files. A mood, in contrast, describes the set of admissible files in
802 terms of attributes and other type of information available in the
803 audio file table. As an example, a mood can define a filename pattern,
804 which is then matched against the names of audio files in the table.
805
806 ### Playlists ###
807
808 Playlists are accommodated in the playlist table of the afs database,
809 using the aforementioned blob format for tables. A new playlist is
810 created with the addpl command by specifying the full (absolute)
811 paths of all desired audio files, separated by newlines. Example:
812
813 find /my/mp3/dir -name "*.mp3" | para addpl my_playlist
814
815 If _my_playlist_ already exists it is overwritten. To activate the
816 new playlist, execute
817
818 para select p/my_playlist
819
820 The audio file selector will assign scores to each entry of the list,
821 in descending order so that files will be selected in order. If a
822 file could not be opened for streaming, its entry is removed from
823 the score table (but not from the playlist).
824
825 ### Moods ###
826
827 A mood consists of a unique name and its *mood definition*, which is
828 a set of *mood lines* containing expressions in terms of attributes
829 and other data contained in the database.
830
831 At any time at most one mood can be *active* which means that
832 para_server is going to select only files from that subset of
833 admissible files.
834
835 So in order to create a mood definition one has to write a set of
836 mood lines. Mood lines come in three flavours: Accept lines, deny
837 lines and score lines.
838
839 The general syntax of the three types of mood lines is
840
841
842 accept [with score <score>] [if] [not] <mood_method> [options]
843 deny [with score <score>] [if] [not] <mood_method> [options]
844 score <score> [if] [not] <mood_method> [options]
845
846
847 Here <score> is either an integer or the string "random" which assigns
848 a random score to all matching files. The score value changes the
849 order in which admissible files are going to be selected, but is of
850 minor importance for this introduction.
851
852 So we concentrate on the first two forms, i.e. accept and deny
853 lines. As usual, everything in square brackets is optional, i.e.
854 accept/deny lines take the following form when ignoring scores:
855
856 accept [if] [not] <mood_method> [options]
857
858 and analogously for the deny case. The "if" keyword is only syntactic
859 sugar and has no function. The "not" keyword just inverts the result,
860 so the essence of a mood line is the mood method part and the options
861 following thereafter.
862
863 A *mood method* is realized as a function which takes an audio file
864 and computes a number from the data contained in the database.
865 If this number is non-negative, we say the file *matches* the mood
866 method. The file matches the full mood line if it either
867
868 - matches the mood method and the "not" keyword is not given,
869 or
870 - does not match the mood method, but the "not" keyword is given.
871
872 The set of admissible files for the whole mood is now defined as those
873 files which match at least one accept mood line, but no deny mood line.
874 More formally, an audio file F is admissible if and only if
875
876 (F ~ AL1 or F ~ AL2...) and not (F ~ DL1 or F ~ DN2 ...)
877
878 where AL1, AL2... are the accept lines, DL1, DL2... are the deny
879 lines and "~" means "matches".
880
881 The cases where no mood lines of accept/deny type are defined need
882 special treatment:
883
884 - Neither accept nor deny lines: This treats all files as
885 admissible (in fact, that is the definition of the dummy mood
886 which is activated automatically if no moods are available).
887
888 - Only accept lines: A file is admissible iff it matches at
889 least one accept line:
890
891 F ~ AL1 or F ~ AL2 or ...
892
893 - Only deny lines: A file is admissible iff it matches no
894 deny line:
895
896 not (F ~ DL1 or F ~ DN2 ...)
897
898
899
900 ### List of mood_methods ###
901
902 no_attributes_set
903
904 Takes no arguments and matches an audio file if and only if no
905 attributes are set.
906
907 is_set <attribute_name>
908
909 Takes the name of an attribute and matches iff that attribute is set.
910
911 path_matches <pattern>
912
913 Takes a filename pattern and matches iff the path of the audio file
914 matches the pattern.
915
916 artist_matches <pattern>
917 album_matches <pattern>
918 title_matches <pattern>
919 comment_matches <pattern>
920
921 Takes an extended regular expression and matches iff the text of the
922 corresponding tag of the audio file matches the pattern. If the tag
923 is not set, the empty string is matched against the pattern.
924
925 year ~ <num>
926 bitrate ~ <num>
927 frequency ~ <num>
928 channels ~ <num>
929 num_played ~ <num>
930 image_id ~ <num>
931 lyrics_id ~ <num>
932
933 Takes a comparator ~ of the set {<, =, <=, >, >=, !=} and a number
934 <num>. Matches an audio file iff the condition <val> ~ <num> is
935 satisfied where val is the corresponding value of the audio file
936 (value of the year tag, bitrate in kbit/s, etc.).
937
938 The year tag is special as its value is undefined if the audio file
939 has no year tag or the content of the year tag is not a number. Such
940 audio files never match. Another difference is the special treatment
941 if the year tag is a two-digit number. In this case either 1900 or
942 2000 is added to the tag value, depending on whether the number is
943 greater than 2000 plus the current year.
944
945
946 ### Mood usage ###
947
948 To create a new mood called "my_mood", write its definition into
949 some temporary file, say "tmpfile", and add it to the mood table
950 by executing
951
952 para addmood my_mood < tmpfile
953
954 If the mood definition is really short, you may just pipe it to the
955 client instead of using temporary files. Like this:
956
957 echo "$MOOD_DEFINITION" | para addmood my_mood
958
959 There is no need to keep the temporary file since you can always use
960 the catmood command to get it back:
961
962 para catmood my_mood
963
964 A mood can be activated by executing
965
966 para select m/my_mood
967
968 Once active, the list of admissible files is shown by the ls command
969 if the "-a" switch is given:
970
971 para ls -a
972
973
974 ### Example mood definition ###
975
976 Suppose you have defined attributes "punk" and "rock" and want to define
977 a mood containing only Punk-Rock songs. That is, an audio file should be
978 admissible if and only if both attributes are set. Since
979
980 punk and rock
981
982 is obviously the same as
983
984 not (not punk or not rock)
985
986 (de Morgan's rule), a mood definition that selects only Punk-Rock
987 songs is
988
989 deny if not is_set punk
990 deny if not is_set rock
991
992
993
994 File renames and content changes
995 --------------------------------
996
997 Since the audio file selector knows the SHA1 of each audio file that
998 has been added to the afs database, it recognizes if the content of
999 a file has changed, e.g. because an ID3 tag was added or modified.
1000 Also, if a file has been renamed or moved to a different location,
1001 afs will detect that an entry with the same hash value already exists
1002 in the audio file table.
1003
1004 In both cases it is enough to just re-add the new file. In the
1005 first case (file content changed), the audio table is updated, while
1006 metadata such as the num_played and last_played fields, as well as
1007 the attributes, remain unchanged. In the other case, when the file
1008 is moved or renamed, only the path information is updated, all other
1009 data remains as before.
1010
1011 It is possible to change the behaviour of the add command by using the
1012 "-l" (lazy add) or the "-f" (force add) option.
1013
1014 Troubleshooting
1015 ---------------
1016
1017 Use the debug loglevel (-l debug) to show debugging info. All paraslash
1018 executables have a brief online help which is displayed when -h is
1019 given. The --detailed-help option prints the full help text.
1020
1021 If para_server crashed or was killed by SIGKILL (signal 9), it
1022 may refuse to start again because of "dirty osl tables". In this
1023 case you'll have to run the oslfsck program of libosl to fix your
1024 database:
1025
1026 oslfsck -fd ~/.paraslash/afs_database-0.4
1027
1028 However, make sure para_server isn't running before executing oslfsck.
1029
1030 If you don't mind to recreate your database you can start
1031 from scratch by removing the entire database directory, i.e.
1032
1033 rm -rf ~/.paraslash/afs_database-0.4
1034
1035 Be aware that this removes all attribute definitions, all playlists
1036 and all mood definitions and requires to re-initialize the tables.
1037
1038 Although oslfsck fixes inconsistencies in database tables it doesn't
1039 care about the table contents. To check for invalid table contents, use
1040
1041 para_client check
1042
1043 This prints out references to missing audio files as well as invalid
1044 playlists and mood definitions.
1045
1046 Similarly, para_audiod refuses to start if its socket file exists, since
1047 this indicates that another instance of para_audiod is running. After
1048 a crash a stale socket file might remain and you must run
1049
1050 para_audiod --force
1051
1052 once to fix it up.
1053
1054 =======================================
1055 Audio formats and audio format handlers
1056 =======================================
1057
1058 Audio formats
1059 -------------
1060
1061 The following audio formats are supported by paraslash:
1062
1063 ### MP3 ###
1064
1065 Mp3, MPEG-1 Audio Layer 3, is a common audio format for audio storage,
1066 designed as part of its MPEG-1 standard. An MP3 file is made up of
1067 multiple MP3 frames, which consist of a header and a data block. The
1068 size of an MP3 frame depends on the bit rate and on the number
1069 of channels. For a typical CD-audio file (sample rate of 44.1 kHz
1070 stereo), encoded with a bit rate of 128 kbit, an MP3 frame is about
1071 400 bytes large.
1072
1073 ### OGG/Vorbis ###
1074
1075 OGG is a standardized audio container format, while Vorbis is an
1076 open source codec for lossy audio compression. Since Vorbis is most
1077 commonly made available via the OGG container format, it is often
1078 referred to as OGG/Vorbis. The OGG container format divides data into
1079 chunks called OGG pages. A typical OGG page is about 4KB large. The
1080 Vorbis codec creates variable-bitrate (VBR) data, where the bitrate
1081 may vary considerably.
1082
1083 ### OGG/Speex ###
1084
1085 Speex is an open-source speech codec that is based on CELP (Code
1086 Excited Linear Prediction) coding. It is designed for voice
1087 over IP applications, has modest complexity and a small memory
1088 footprint. Wideband and narrowband (telephone quality) speech are
1089 supported. As for Vorbis audio, Speex bit-streams are often stored
1090 in OGG files. As of 2012 this codec is considered obsolete since the
1091 Oppus codec, described below, surpasses its performance in all areas.
1092
1093 ### OGG/Opus ###
1094
1095 Opus is a lossy audio compression format standardized through RFC
1096 6716 in 2012. It combines the speech-oriented SILK codec and the
1097 low-latency CELT (Constrained Energy Lapped Transform) codec. Like
1098 OGG/Vorbis and OGG/Speex, Opus data is usually encapsulated in OGG
1099 containers. All known software patents which cover Opus are licensed
1100 under royalty-free terms.
1101
1102 ### AAC ###
1103
1104 Advanced Audio Coding (AAC) is a standardized, lossy compression
1105 and encoding scheme for digital audio which is the default audio
1106 format for Apple's iPhone, iPod, iTunes. Usually MPEG-4 is used as
1107 the container format and audio files encoded with AAC have the .m4a
1108 extension. A typical AAC frame is about 700 bytes large.
1109
1110 ### WMA ###
1111
1112 Windows Media Audio (WMA) is an audio data compression technology
1113 developed by Microsoft. A WMA file is usually encapsulated in the
1114 Advanced Systems Format (ASF) container format, which also specifies
1115 how meta data about the file is to be encoded. The bit stream of WMA
1116 is composed of superframes, each containing one or more frames of
1117 2048 samples. For 16 bit stereo a WMA superframe is about 8K large.
1118
1119 ### FLAC ###
1120
1121 The Free Lossless Audio Codec (FLAC) compresses audio without quality
1122 loss. It gives better compression ratios than a general purpose
1123 compressor like zip or bzip2 because FLAC is designed specifically
1124 for audio. A FLAC-encoded file consists of frames of varying size, up
1125 to 16K. Each frame starts with a header that contains all information
1126 necessary to decode the frame.
1127
1128 Meta data
1129 ---------
1130
1131 Unfortunately, each audio format has its own conventions how meta
1132 data is added as tags to the audio file.
1133
1134 For MP3 files, ID3, version 1 and 2 are widely used. ID3 version 1
1135 is rather simple but also very limited as it supports only artist,
1136 title, album, year and comment tags. Each of these can only be at most
1137 32 characters long. ID3, version 2 is much more flexible but requires
1138 a separate library being installed for paraslash to support it.
1139
1140 Ogg vorbis, ogg speex and flac files contain meta data as Vorbis
1141 comments, which are typically implemented as strings of the form
1142 "[TAG]=[VALUE]". Unlike ID3 version 1 tags, one may use whichever
1143 tags are appropriate for the content.
1144
1145 AAC files usually use the MPEG-4 container format for storing meta
1146 data while WMA files wrap meta data as special objects within the
1147 ASF container format.
1148
1149 paraslash only tracks the most common tags that are supported by
1150 all tag variants: artist, title, year, album, comment. When a file
1151 is added to the AFS database, the meta data of the file is extracted
1152 and stored in the audio file table.
1153
1154 Chunks and chunk tables
1155 -----------------------
1156
1157 paraslash uses the word "chunk" as common term for the building blocks
1158 of an audio file. For MP3 files, a chunk is the same as an MP3 frame,
1159 while for OGG files a chunk is an OGG page, etc. Therefore the chunk
1160 size varies considerably between audio formats, from a few hundred
1161 bytes (MP3) up to 16K (FLAC).
1162
1163 The chunk table contains the offsets within the audio file that
1164 correspond to the chunk boundaries of the file. Like the meta data,
1165 the chunk table is computed and stored in the database whenever an
1166 audio file is added.
1167
1168 The paraslash senders (see below) always send complete chunks. The
1169 granularity for seeking is therefore determined by the chunk size.
1170
1171 Audio format handlers
1172 ---------------------
1173
1174 For each audio format paraslash contains an audio format handler whose
1175 first task is to tell whether a given file is a valid audio file of
1176 this type. If so, the audio file handler extracts some technical data
1177 (duration, sampling rate, number of channels etc.), computes the
1178 chunk table and reads the meta data.
1179
1180 The audio format handler code is linked into para_server and executed
1181 via the _add_ command. The same code is also available as a stand-alone
1182 tool, para_afh, which prints the technical data, the chunk table
1183 and the meta data of a file. Moreover, all audio format handlers are
1184 combined in the afh receiver which is part of para_recv and para_play.
1185
1186 ==========
1187 Networking
1188 ==========
1189
1190 Paraslash uses different network connections for control and data.
1191 para_client communicates with para_server over a dedicated TCP control
1192 connection. To transport audio data, separate data connections are
1193 used. For these data connections, a variety of transports (UDP, DCCP,
1194 HTTP) can be chosen.
1195
1196 The chapter starts with the [control
1197 service](#The.paraslash.control.service), followed by a section
1198 on the various [streaming protocols](#Streaming.protocols)
1199 in which the data connections are described. The way
1200 audio file headers are embedded into the stream is discussed
1201 [briefly](#Streams.with.headers.and.headerless.streams) before the
1202 [example section](#Networking.examples) which illustrates typical
1203 commands for real-life scenarios.
1204
1205 Both IPv4 and IPv6 are supported.
1206
1207 The paraslash control service
1208 -----------------------------
1209
1210 para_server is controlled at runtime via the paraslash control
1211 connection. This connection is used for server commands (play, stop,
1212 ...) as well as for afs commands (ls, select, ...).
1213
1214 The server listens on a TCP port and accepts connections from clients
1215 that connect the open port. Each connection causes the server to fork
1216 off a client process which inherits the connection and deals with that
1217 client only. In this classical accept/fork approach the server process
1218 is unaffected if the child dies or goes crazy for whatever reason. In
1219 fact, the child process can not change address space of server process.
1220
1221 The section on [client-server
1222 authentication](#Client-server.authentication) above described the
1223 early connection establishment from the crypto point of view. Here
1224 it is described what happens after the connection (including crypto
1225 setup) has been established. There are four processes involved during
1226 command dispatch as sketched in the following diagram.
1227
1228 server_host client_host
1229 ~~~~~~~~~~~ ~~~~~~~~~~~
1230
1231 +-----------+ connect +-----------+
1232 |para_server|<------------------------------ |para_client|
1233 +-----------+ +-----------+
1234 | ^
1235 | fork +---+ |
1236 +----------> |AFS| |
1237 | +---+ |
1238 | ^ |
1239 | | |
1240 | | connect (cookie) |
1241 | | |
1242 | | |
1243 | fork +-----+ inherited connection |
1244 +---------->|child|<--------------------------+
1245 +-----+
1246
1247 Note that the child process is not a child of the afs process,
1248 so communication of these two processes has to happen via local
1249 sockets. In order to avoid abuse of the local socket by unrelated
1250 processes, a magic cookie is created once at server startup time just
1251 before the server process forks off the AFS process. This cookie is
1252 known to the server, AFS and the child, but not to unrelated processes.
1253
1254 There are two different kinds of commands: First there are commands
1255 that cause the server to respond with some answer such as the list
1256 of all audio files. All but the addblob commands (addimg, addlyr,
1257 addpl, addmood) are of this kind. The addblob commands add contents
1258 to the database, so they need to transfer data the other way round,
1259 from the client to the server.
1260
1261 There is no knowledge about the server commands built into para_client,
1262 so it does not know about addblob commands. Instead, the server sends
1263 a special "awaiting data" packet for these commands. If the client
1264 receives this packet, it sends STDIN to the server, otherwise it
1265 dumps data from the server to STDOUT.
1266
1267 Streaming protocols
1268 -------------------
1269
1270 A network (audio) stream usually consists of one streaming source,
1271 the _sender_, and one or more _receivers_ which read data over the
1272 network from the streaming source.
1273
1274 Senders are thus part of para_server while receivers are part of
1275 para_audiod. Moreover, there is the stand-alone tool para_recv which
1276 can be used to manually download a stream, either from para_server
1277 or from a web-based audio streaming service.
1278
1279 The following three streaming protocols are supported by paraslash:
1280
1281 - HTTP. Recommended for public streams that can be played by any
1282 player like mpg123, xmms, itunes, winamp, etc. The HTTP sender is
1283 supported on all operating systems and all platforms.
1284
1285 - DCCP. Recommended for LAN streaming. DCCP is currently available
1286 only for Linux.
1287
1288 - UDP. Recommended for multicast LAN streaming.
1289
1290 See the Appendix on [network protocols](/#Network.protocols)
1291 for brief descriptions of the various protocols relevant for network
1292 audio streaming with paraslash.
1293
1294 It is possible to activate more than one sender simultaneously.
1295 Senders can be controlled at run time and via config file and command
1296 line options.
1297
1298 Note that audio connections are _not_ encrypted. Transport or Internet
1299 layer encryption should be used if encrypted data connections are
1300 needed.
1301
1302 Since DCCP and TCP are both connection-oriented protocols, connection
1303 establishment/teardown and access control are very similar between
1304 these two streaming protocols. UDP is the most lightweight option,
1305 since in contrast to TCP/DCCP it is connectionless. It is also the
1306 only protocol supporting IP multicast.
1307
1308 The HTTP and the DCCP sender listen on a (TCP/DCCP) port waiting for
1309 clients to connect and establish a connection via some protocol-defined
1310 handshake mechanism. Both senders maintain two linked lists each:
1311 The list of all clients which are currently connected, and the list
1312 of access control entries which determines who is allowed to connect.
1313 IP-based access control may be configured through config file and
1314 command line options and via the "allow" and "deny" sender subcommands.
1315
1316 Upon receiving a GET request from the client, the HTTP sender sends
1317 back a status line and a message. The body of this message is the
1318 audio stream. This is common practice and is supported by many popular
1319 clients which can thus be used to play a stream offered by para_server.
1320 For DCCP things are a bit simpler: No messages are exchanged between
1321 the receiver and sender. The client simply connects and the sender
1322 starts to stream.
1323
1324 DCCP is an experimental protocol which offers a number of new features
1325 not available for TCP. Both ends can negotiate these features using
1326 a built-in negotiation mechanism. In contrast to TCP/HTTP, DCCP is
1327 datagram-based (no retransmissions) and thus should not be used over
1328 lossy media (e.g. WiFi networks). One useful feature offered by DCCP
1329 is access to a variety of different congestion-control mechanisms
1330 called CCIDs. Two different CCIDs are available per default on Linux:
1331
1332
1333 - _CCID 2_. A Congestion Control mechanism similar to that of TCP. The
1334 sender maintains a congestion window and halves this window in response
1335 to congestion.
1336
1337
1338 - _CCID-3_. Designed to be fair when competing for bandwidth.
1339 It has lower variation of throughput over time compared with TCP,
1340 which makes it suitable for streaming media.
1341
1342 Unlike the HTTP and DCCP senders, the UDP sender maintains only a
1343 single list, the _target list_. This list describes the set of clients
1344 to which the stream is sent. There is no list for access control and
1345 no "allow" and "deny" commands for the UDP sender. Instead, the "add"
1346 and "delete" commands can be used to modify the target list.
1347
1348 Since both UDP and DCCP offer an unreliable datagram-based transport,
1349 additional measures are necessary to guard against disruptions over
1350 networks that are lossy or which may be subject to interference (as
1351 is for instance the case with WiFi). Paraslash uses FEC (Forward
1352 Error Correction) to guard against packet losses and reordering. The
1353 stream is FEC-encoded before it is sent through the UDP socket and
1354 must be decoded accordingly on the receiver side.
1355
1356 The packet size and the amount of redundancy introduced by FEC can
1357 be configured via the FEC parameters which are dictated by server
1358 and may also be configured through the "sender" command. The FEC
1359 parameters are encoded in the header of each network packet, so no
1360 configuration is necessary on the receiver side. See the section on
1361 [FEC](#Forward.error.correction) below.
1362
1363 Streams with headers and headerless streams
1364 -------------------------------------------
1365
1366 For OGG/Vorbis, OGG/Speex and wma streams, some of the information
1367 needed to decode the stream is only contained in the audio file
1368 header of the container format but not in each data chunk. Clients
1369 must be able to obtain this information in case streaming starts in
1370 the middle of the file or if para_audiod is started while para_server
1371 is already sending a stream.
1372
1373 This is accomplished in different ways, depending on the streaming
1374 protocol. For connection-oriented streams (HTTP, DCCP) the audio file
1375 header is sent prior to audio file data. This technique however does
1376 not work for the connectionless UDP transport. Hence the audio file
1377 header is periodically being embedded into the UDP audio data stream.
1378 By default, the header is resent after five seconds. The receiver has
1379 to wait until the next header arrives before it can start decoding
1380 the stream.
1381
1382 Networking examples
1383 -------------------
1384
1385 The "si" (server info) command lists some information about the
1386 currently running server process.
1387
1388 -> Show PIDs, number of connected clients, uptime, and more:
1389
1390 para_client si
1391
1392 The sender command of para_server prints information about senders,
1393 like the various access control lists, and it allows to (de-)activate
1394 senders and to change the access permissions at runtime.
1395
1396 -> List all senders
1397
1398 para_client sender
1399
1400 -> Obtain general help for the sender command:
1401
1402 para_client help sender
1403
1404 -> Get help for a specific sender (contains further examples):
1405
1406 s=http # or dccp or udp
1407 para_client sender $s help
1408
1409 -> Show status of the http sender
1410
1411 para_client sender http status
1412
1413 By default para_server activates both the HTTP and th DCCP sender on
1414 startup. This can be changed via command line options or para_server's
1415 config file.
1416
1417 -> List config file options for senders:
1418
1419 para_server -h
1420
1421 All senders share the "on" and "off" commands, so senders may be
1422 activated and deactivated independently of each other.
1423
1424 -> Switch off the http sender:
1425
1426 para_client sender http off
1427
1428 -> Receive a DCCP stream using CCID2 and write the output into a file:
1429
1430 host=foo.org; ccid=2; filename=bar
1431 para_recv --receiver "dccp --host $host --ccid $ccid" > $filename
1432
1433 Note the quotes around the arguments for the dccp receiver. Each
1434 receiver has its own set of command line options and its own command
1435 line parser, so arguments for the dccp receiver must be protected
1436 from being interpreted by para_recv.
1437
1438 -> Start UDP multicast, using the default multicast address:
1439
1440 para_client sender udp add 224.0.1.38
1441
1442 -> Receive FEC-encoded multicast stream and write the output into a file:
1443
1444 filename=foo
1445 para_recv -r udp > $filename
1446
1447 -> Add an UDP unicast for a client to the target list of the UDP sender:
1448
1449 t=client.foo.org
1450 para_client sender udp add $t
1451
1452 -> Receive this (FEC-encoded) unicast stream:
1453
1454 filename=foo
1455 para_recv -r 'udp -i 0.0.0.0' > $filename
1456
1457 -> Create a minimal config for para_audiod for HTTP streams:
1458
1459 c=$HOME/.paraslash/audiod.conf.min; s=server.foo.com
1460 echo receiver \".:http -i $s\" > $c
1461 para_audiod --config $c
1462
1463 =======
1464 Filters
1465 =======
1466
1467 A paraslash filter is a module which transforms an input stream into
1468 an output stream. Filters are included in the para_audiod executable
1469 and in the stand-alone tool para_filter which usually contains the
1470 same modules.
1471
1472 While para_filter reads its input stream from STDIN and writes
1473 the output to STDOUT, the filter modules of para_audiod are always
1474 connected to a receiver which produces the input stream and a writer
1475 which absorbs the output stream.
1476
1477 Some filters depend on a specific library and are not compiled in
1478 if this library was not found at compile time. To see the list of
1479 supported filters, run para_filter and para_audiod with the --help
1480 option. The output looks similar to the following:
1481
1482 Available filters:
1483 compress wav amp fecdec wmadec prebuffer oggdec aacdec mp3dec
1484
1485 Out of these filter modules, a chain of filters can be constructed,
1486 much in the way Unix pipes can be chained, and analogous to the use
1487 of modules in gstreamer: The output of the first filter becomes the
1488 input of the second filter. There is no limitation on the number of
1489 filters and the same filter may occur more than once.
1490
1491 Like receivers, each filter has its own command line options which
1492 must be quoted to protect them from the command line options of
1493 the driving application (para_audiod or para_filter). Example:
1494
1495 para_filter -f 'mp3dec --ignore-crc' -f 'compress --damp 1'
1496
1497 For para_audiod, each audio format has its own set of filters. The
1498 name of the audio format for which the filter should be applied can
1499 be used as the prefix for the filter option. Example:
1500
1501 para_audiod -f 'mp3:prebuffer --duration 300'
1502
1503 The "mp3" prefix above is actually interpreted as a POSIX extended
1504 regular expression. Therefore
1505
1506 para_audiod -f '.:prebuffer --duration 300'
1507
1508 activates the prebuffer filter for all supported audio formats (because
1509 "." matches all audio formats) while
1510
1511 para_audiod -f 'wma|ogg:prebuffer --duration 300'
1512
1513 activates it only for wma and ogg streams.
1514
1515 Decoders
1516 --------
1517
1518 For each supported audio format there is a corresponding filter
1519 which decodes audio data in this format to 16 bit PCM data which
1520 can be directly sent to the sound device or any other software that
1521 operates on undecoded PCM data (visualizers, equalizers etc.). Such
1522 filters are called _decoders_ in general, and xxxdec is the name of
1523 the paraslash decoder for the audio format xxx. For example, the mp3
1524 decoder is called mp3dec.
1525
1526 Note that the output of the decoder is about 10 times larger than
1527 its input. This means that filters that operate on the decoded audio
1528 stream have to deal with much more data than filters that transform
1529 the audio stream before it is fed to the decoder.
1530
1531 Paraslash relies on external libraries for most decoders, so these
1532 libraries must be installed for the decoder to be included in the
1533 executables. For example, the mp3dec filter depends on the mad library.
1534
1535 Forward error correction
1536 ------------------------
1537
1538 As already mentioned [earlier](#Streaming.protocols), paraslash
1539 uses forward error correction (FEC) for the unreliable UDP and
1540 DCCP transports. FEC is a technique which was invented already in
1541 1960 by Reed and Solomon and which is widely used for the parity
1542 calculations of storage devices (RAID arrays). It is based on the
1543 algebraic concept of finite fields, today called Galois fields, in
1544 honour of the mathematician Galois (1811-1832). The FEC implementation
1545 of paraslash is based on code by Luigi Rizzo.
1546
1547 Although the details require a sound knowledge of the underlying
1548 mathematics, the basic idea is not hard to understand: For positive
1549 integers k and n with k < n it is possible to compute for any k given
1550 data bytes d_1, ..., d_k the corresponding r := n -k parity bytes p_1,
1551 ..., p_r such that all data bytes can be reconstructed from *any*
1552 k bytes of the set
1553
1554 {d_1, ..., d_k, p_1, ..., p_r}.
1555
1556 FEC-encoding for unreliable network transports boils down to slicing
1557 the audio stream into groups of k suitably sized pieces called _slices_
1558 and computing the r corresponding parity slices. This step is performed
1559 in para_server which then sends both the data and the parity slices
1560 over the unreliable network connection. If the client was able
1561 to receive at least k of the n = k + r slices, it can reconstruct
1562 (FEC-decode) the original audio stream.
1563
1564 From these observations it is clear that there are three different
1565 FEC parameters: The slice size, the number of data slices k, and the
1566 total number of slices n. It is crucial to choose the slice size
1567 such that no fragmentation of network packets takes place because
1568 FEC only guards against losses and reordering but fails if slices are
1569 received partially.
1570
1571 FEC decoding in paralash is performed through the fecdec filter which
1572 usually is the first filter (there can be other filters before fecdec
1573 if these do not alter the audio stream).
1574
1575 Volume adjustment (amp and compress)
1576 ------------------------------------
1577
1578 The amp and the compress filter both adjust the volume of the audio
1579 stream. These filters operate on uncompressed audio samples. Hence
1580 they are usually placed directly after the decoding filter. Each
1581 sample is multiplied with a scaling factor (>= 1) which makes amp
1582 and compress quite expensive in terms of computing power.
1583
1584 ### amp ###
1585
1586 The amp filter amplifies the audio stream by a fixed scaling factor
1587 that must be known in advance. For para_audiod this factor is derived
1588 from the amplification field of the audio file's entry in the audio
1589 file table while para_filter uses the value given at the command line.
1590
1591 The optimal scaling factor F for an audio file is the largest real
1592 number F >= 1 such that after multiplication with F all samples still
1593 fit into the sample interval [-32768, 32767]. One can use para_filter
1594 in combination with the sox utility to compute F:
1595
1596 para_filter -f mp3dec -f wav < file.mp3 | sox -t wav - -e stat -v
1597
1598 The amplification value V which is stored in the audio file table,
1599 however, is an integer between 0 and 255 which is connected to F
1600 through the formula
1601
1602 V = (F - 1) * 64.
1603
1604 To store V in the audio file table, the command
1605
1606 para_client -- touch -a=V file.mp3
1607
1608 is used. The reader is encouraged to write a script that performs
1609 these computations :)
1610
1611 ### compress ###
1612
1613 Unlike the amplification filter, the compress filter adjusts the volume
1614 of the audio stream dynamically without prior knowledge about the peak
1615 value. It maintains the maximal volume of the last n samples of the
1616 audio stream and computes a suitable amplification factor based on that
1617 value and the various configuration options. It tries to chose this
1618 factor such that the adjusted volume meets the desired target level.
1619
1620 Note that it makes sense to combine amp and compress.
1621
1622 Misc filters (wav and prebuffer)
1623 --------------------------------
1624
1625 These filters are rather simple and do not modify the audio stream at
1626 all. The wav filter is only useful with para_filter and in connection
1627 with a decoder. It asks the decoder for the number of channels and the
1628 sample rate of the stream and adds a Microsoft wave header containing
1629 this information at the beginning. This allows to write wav files
1630 rather than raw PCM files (which do not contain any information about
1631 the number of channels and the sample rate).
1632
1633 The prebuffer filter simply delays the output until the given time has
1634 passed (starting from the time the first byte was available in its
1635 input queue) or until the given amount of data has accumulated. It
1636 is mainly useful for para_audiod if the standard parameters result
1637 in buffer underruns.
1638
1639 Both filters require almost no additional computing time, even when
1640 operating on uncompressed audio streams, since data buffers are simply
1641 "pushed down" rather than copied.
1642
1643 Examples
1644 --------
1645
1646 -> Decode an mp3 file to wav format:
1647
1648 para_filter -f mp3dec -f wav < file.mp3 > file.wav
1649
1650 -> Amplify a raw audio file by a factor of 1.5:
1651
1652 para_filter -f amp --amp 32 < foo.raw > bar.raw
1653
1654 ======
1655 Output
1656 ======
1657
1658 Once an audio stream has been received and decoded to PCM format,
1659 it can be sent to a sound device for playback. This part is performed
1660 by paraslash _writers_ which are described in this chapter.
1661
1662 Writers
1663 -------
1664
1665 A paraslash writer acts as a data sink that consumes but does not
1666 produce audio data. Paraslash writers operate on the client side and
1667 are contained in para_audiod and in the stand-alone tool para_write.
1668
1669 The para_write program reads uncompressed audio data from STDIN. If
1670 this data starts with a wav header, sample rate, sample format and
1671 channel count are read from the header. Otherwise CD audio (44.1KHz
1672 16 bit little endian, stereo) is assumed but this can be overridden
1673 by command line options. para_audiod, on the other hand, obtains
1674 the sample rate and the number of channels from the decoder.
1675
1676 Like receivers and filters, each writer has an individual set of
1677 command line options, and for para_audiod writers can be configured
1678 per audio format separately. It is possible to activate more than
1679 one writer for the same stream simultaneously.
1680
1681 OS-dependent APIs
1682 -----------------
1683
1684 Unfortunately, the various flavours of Unix on which paraslash
1685 runs on have different APIs for opening a sound device and starting
1686 playback. Hence for each such API there is a paraslash writer that
1687 can play the audio stream via this API.
1688
1689 - *ALSA*. The _Advanced Linux Sound Architecture_ is only available on
1690 Linux systems. Although there are several mid-layer APIs in use by
1691 the various Linux distributions (ESD, Jack, PulseAudio), paraslash
1692 currently supports only the low-level ALSA API which is not supposed
1693 to be change. ALSA is very feature-rich, in particular it supports
1694 software mixing via its DMIX plugin. ALSA is the default writer on
1695 Linux systems.
1696
1697 - *OSS*. The _Open Sound System_ is the only API on \*BSD Unixes and
1698 is also available on Linux systems, usually provided by ALSA as an
1699 emulation for backwards compatibility. This API is rather simple but
1700 also limited. For example only one application can open the device
1701 at any time. The OSS writer is activated by default on BSD Systems.
1702
1703 - *OSX*. Mac OS X has yet another API called CoreAudio. The OSX writer
1704 for this API is only compiled in on such systems and is of course
1705 the default there.
1706
1707 - *FILE*. The file writer allows to capture the audio stream and
1708 write the PCM data to a file on the file system rather than playing
1709 it through a sound device. It is supported on all platforms and is
1710 always compiled in.
1711
1712 - *AO*. _Libao_ is a cross-platform audio library which supports a wide
1713 variety of platforms including PulseAudio (gnome), ESD (Enlightened
1714 Sound Daemon), AIX, Solaris and IRIX. The ao writer plays audio
1715 through an output plugin of libao.
1716
1717 Examples
1718 --------
1719
1720 -> Use the OSS writer to play a wav file:
1721
1722 para_write --writer oss < file.wav
1723
1724 -> Enable ALSA software mixing for mp3 streams:
1725
1726 para_audiod --writer 'mp3:alsa -d plug:swmix'
1727
1728
1729 ===
1730 Gui
1731 ===
1732
1733 para_gui executes an arbitrary command which is supposed to print
1734 status information to STDOUT. It then displays this information in
1735 a curses window. By default the command
1736
1737 para_audioc -- stat -p
1738
1739 is executed, but this can be customized via the --stat-cmd option. In
1740 particular it possible to use
1741
1742 para_client -- stat -p
1743
1744 to make para_gui work on systems on which para_audiod is not running.
1745
1746 Key bindings
1747 ------------
1748
1749 It is possible to bind keys to arbitrary commands via custom
1750 key-bindings. Besides the internal keys which can not be changed (help,
1751 quit, loglevel, version...), the following flavours of key-bindings
1752 are supported:
1753
1754 - external: Shutdown curses before launching the given command.
1755 Useful for starting other ncurses programs from within para_gui,
1756 e.g. aumix or dialog scripts. Or, use the mbox output format to write
1757 a mailbox containing one mail for each (admissible) file the audio
1758 file selector knows about. Then start mutt from within para_gui to
1759 browse your collection!
1760
1761 - display: Launch the command and display its stdout in para_gui's
1762 bottom window.
1763
1764 - para: Like display, but start "para_client <specified command>"
1765 instead of "<specified command>".
1766
1767 The general form of a key binding is
1768
1769 key_map k:m:c
1770
1771 which maps key k to command c using mode m. Mode may be x, d or p
1772 for external, display and paraslash commands, respectively.
1773
1774 Themes
1775 ------
1776
1777 Currently there are only two themes for para_gui. It is easy, however,
1778 to add more themes. To create a new theme one has to define the
1779 position, color and geometry for for each status item that should be
1780 shown by this theme. See gui_theme.c for examples.
1781
1782 The "." and "," keys are used to switch between themes.
1783
1784 Examples
1785 --------
1786
1787 -> Show server info:
1788
1789 key_map "i:p:si"
1790
1791 -> Jump to the middle of the current audio file by pressing F5:
1792
1793 key_map "<F5>:p:jmp 50"
1794
1795 -> vi-like bindings for jumping around:
1796
1797 key_map "l:p:ff 10"
1798 key_map "h:p:ff 10-"
1799 key_map "w:p:ff 60"
1800 key_map "b:p:ff 60-"
1801
1802 -> Print the current date and time:
1803
1804 key_map "D:d:date"
1805
1806 -> Call other curses programs:
1807
1808 key_map "U:x:aumix"
1809 key_map "!:x:/bin/bash"
1810 key_map "^E:x:/bin/sh -c 'vi ~/.paraslash/gui.conf'"
1811
1812 ===========
1813 Development
1814 ===========
1815
1816 Contributing
1817 ------------
1818
1819 Paraslash is an open source project and contributions are
1820 welcome. Here's a list of things you can do to help the project:
1821
1822 - Report problems with building, installing or running the software.
1823 In particular, test the experimental git branches ("next" and "pu").
1824 This helps to identify and fix problems before the code gets merged
1825 and thus keeps the master branch as stable as possible.
1826 - Proofread the documentation (manual, web pages, man pages, source
1827 code documentation) and point out unclear or poorly written parts. If
1828 you are a native English speaker you will easily find a lot of text
1829 that could be improved.
1830 - Run analysis tools (coverity, afl, sparse, etc.) and report issues
1831 found by those tools.
1832 - Suggest new features you would like to see implemented.
1833 - Compile and test on your favorite architecture or operating
1834 system. The code is tested only on a limited set of systems, so you
1835 will probably encounter problems when building on different systems.
1836 - Post about about paraslash on your blog or on social networks.
1837 - Build and maintain Debian/RPM packages for your favorite distribution.
1838
1839 Note that there is no mailing list, no bug tracker and no discussion
1840 forum for paraslash. If you'd like to contribute, or have questions
1841 about contributing, send email to Andre Noll <maan@tuebingen.mpg.de>.
1842
1843 Tools
1844 -----
1845
1846 In order to compile the sources from the git repository (rather than
1847 from tar balls) and for contributing non-trivial changes to the
1848 paraslash project, some additional tools should be installed on a
1849 developer machine.
1850
1851 - [git](http://git.or.cz/). As described in more detail
1852 [below](#Git.branches), the git source code management tool is used for
1853 paraslash development. It is necessary for cloning the git repository
1854 and for getting updates.
1855
1856 - [autoconf](ftp://ftp.gnu.org/pub/gnu/autoconf/) GNU autoconf creates
1857 the configure file which is shipped in the tarballs but has to be
1858 generated when compiling from git.
1859
1860 - [discount](http://www.pell.portland.or.us/~orc/Code/discount). The
1861 HTML version of this manual and some of the paraslash web pages are
1862 written in the Markdown markup language and are translated into html
1863 with the converter of the *Discount* package.
1864
1865 - [doxygen](http://www.stack.nl/~dimitri/doxygen/). The documentation
1866 of paraslash's C sources uses the doxygen documentation system. The
1867 conventions for documenting the source code is described in the
1868 [Doxygen section](#Doxygen).
1869
1870 - [global](ftp://ftp.gnu.org/pub/gnu/global). This is used to generate
1871 browsable HTML from the C sources. It is needed by doxygen.
1872
1873 Git branches
1874 ------------
1875
1876 Paraslash has been developed using the git source code management
1877 tool since 2006. Development is organized roughly in the same spirit
1878 as the git development itself, as described below.
1879
1880 The following text passage is based on "A note from the maintainer",
1881 written by Junio C Hamano, the maintainer of git.
1882
1883 There are four branches in the paraslash repository that track the
1884 source tree: "master", "maint", "next", and "pu".
1885
1886 The "master" branch is meant to contain what is well tested and
1887 ready to be used in a production setting. There could occasionally be
1888 minor breakages or brown paper bag bugs but they are not expected to
1889 be anything major, and more importantly quickly and easily fixable.
1890 Every now and then, a "feature release" is cut from the tip of this
1891 branch, named with three dotted decimal digits, like 0.4.2.
1892
1893 Whenever changes are about to be included that will eventually lead to
1894 a new major release (e.g. 0.5.0), a "maint" branch is forked off from
1895 "master" at that point. Obvious, safe and urgent fixes after the major
1896 release are applied to this branch and maintenance releases are cut
1897 from it. New features never go to this branch. This branch is also
1898 merged into "master" to propagate the fixes forward.
1899
1900 A trivial and safe enhancement goes directly on top of "master".
1901 New development does not usually happen on "master", however.
1902 Instead, a separate topic branch is forked from the tip of "master",
1903 and it first is tested in isolation; Usually there are a handful such
1904 topic branches that are running ahead of "master". The tip of these
1905 branches is not published in the public repository to keep the number
1906 of branches that downstream developers need to worry about low.
1907
1908 The quality of topic branches varies widely. Some of them start out as
1909 "good idea but obviously is broken in some areas" and then with some
1910 more work become "more or less done and can now be tested by wider
1911 audience". Luckily, most of them start out in the latter, better shape.
1912
1913 The "next" branch is to merge and test topic branches in the latter
1914 category. In general, this branch always contains the tip of "master".
1915 It might not be quite rock-solid production ready, but is expected to
1916 work more or less without major breakage. The maintainer usually uses
1917 the "next" version of paraslash for his own pleasure, so it cannot
1918 be _that_ broken. The "next" branch is where new and exciting things
1919 take place.
1920
1921 The two branches "master" and "maint" are never rewound, and "next"
1922 usually will not be either (this automatically means the topics that
1923 have been merged into "next" are usually not rebased, and you can find
1924 the tip of topic branches you are interested in from the output of
1925 "git log next"). You should be able to safely build on top of them.
1926
1927 However, at times "next" will be rebuilt from the tip of "master" to
1928 get rid of merge commits that will never be in "master". The commit
1929 that replaces "next" will usually have the identical tree, but it
1930 will have different ancestry from the tip of "master".
1931
1932 The "pu" (proposed updates) branch bundles the remainder of the
1933 topic branches. The "pu" branch, and topic branches that are only in
1934 "pu", are subject to rebasing in general. By the above definition
1935 of how "next" works, you can tell that this branch will contain quite
1936 experimental and obviously broken stuff.
1937
1938 When a topic that was in "pu" proves to be in testable shape, it
1939 graduates to "next". This is done with
1940
1941 git checkout next
1942 git merge that-topic-branch
1943
1944 Sometimes, an idea that looked promising turns out to be not so good
1945 and the topic can be dropped from "pu" in such a case.
1946
1947 A topic that is in "next" is expected to be polished to perfection
1948 before it is merged to "master". Similar to the above, this is
1949 done with
1950
1951 git checkout master
1952 git merge that-topic-branch
1953 git branch -d that-topic-branch
1954
1955 Note that being in "next" is not a guarantee to appear in the next
1956 release (being in "master" is such a guarantee, unless it is later
1957 found seriously broken and reverted), nor even in any future release.
1958
1959 Coding Style
1960 ------------
1961
1962 The preferred coding style for paraslash coincides more or less
1963 with the style of the Linux kernel. So rather than repeating what is
1964 written [there](http://www.kernel.org/doc/Documentation/process/coding-style.rst),
1965 here are the most important points.
1966
1967 - Burn the GNU coding standards.
1968 - Never use spaces for indentation.
1969 - Tabs are 8 characters, and thus indentations are also 8 characters.
1970 - Don't put multiple assignments on a single line.
1971 - Avoid tricky expressions.
1972 - Don't leave whitespace at the end of lines.
1973 - The limit on the length of lines is 80 columns.
1974 - Use K&R style for placing braces and spaces:
1975
1976 if (x is true) {
1977 we do y
1978 }
1979
1980 - Use a space after (most) keywords.
1981 - Do not add spaces around (inside) parenthesized expressions.
1982 - Use one space around (on each side of) most binary and ternary operators.
1983 - Do not use cute names like ThisVariableIsATemporaryCounter, call it tmp.
1984 - Mixed-case names are frowned upon.
1985 - Descriptive names for global variables are a must.
1986 - Avoid typedefs.
1987 - Functions should be short and sweet, and do just one thing.
1988 - The number of local variables shouldn't exceed 10.
1989 - Gotos are fine if they improve readability and reduce nesting.
1990 - Don't use C99-style "// ..." comments.
1991 - Names of macros defining constants and labels in enums are capitalized.
1992 - Enums are preferred when defining several related constants.
1993 - Always use the paraslash wrappers for allocating memory.
1994 - If the name of a function is an action or an imperative.
1995 command, the function should return an error-code integer
1996 (<0 means error, >=0 means success). If the name is a
1997 predicate, the function should return a "succeeded" boolean.
1998
1999 Doxygen
2000 -------
2001
2002 Doxygen is a documentation system for various programming
2003 languages. The API reference on the paraslash web page is generated
2004 by doxygen.
2005
2006 It is more illustrative to look at the source code for examples than
2007 to describe the conventions in this manual, so we only describe which
2008 parts of the code need doxygen comments, but leave out details on
2009 documentation conventions.
2010
2011 As a rule, only the public part of the C source is documented with
2012 Doxygen. This includes structures, defines and enumerations in header
2013 files as well as public (non-static) C functions. These should be
2014 documented completely. For example, each parameter and the return
2015 value of a public function should get a descriptive doxygen comment.
2016
2017 No doxygen comments are necessary for static functions and for
2018 structures and enumerations in C files (which are used only within
2019 this file). This does not mean, however, that those entities need
2020 no documentation at all. Instead, common sense should be applied to
2021 document what is not obvious from reading the code.
2022
2023 ========
2024 Appendix
2025 ========
2026
2027 Network protocols
2028 -----------------
2029
2030 ### IP ###
2031
2032 The _Internet Protocol_ is the primary networking protocol used for
2033 the Internet. All protocols described below use IP as the underlying
2034 layer. Both the prevalent IPv4 and the next-generation IPv6 variant
2035 are being deployed actively worldwide.
2036
2037 ### Connection-oriented and connectionless protocols ###
2038
2039 Connectionless protocols differ from connection-oriented ones in
2040 that state associated with the sending/receiving endpoints is treated
2041 implicitly. Connectionless protocols maintain no internal knowledge
2042 about the state of the connection. Hence they are not capable of
2043 reacting to state changes, such as sudden loss or congestion on the
2044 connection medium. Connection-oriented protocols, in contrast, make
2045 this knowledge explicit. The connection is established only after
2046 a bidirectional handshake which requires both endpoints to agree
2047 on the state of the connection, and may also involve negotiating
2048 specific parameters for the particular connection. Maintaining an
2049 up-to-date internal state of the connection also in general means
2050 that the sending endpoints perform congestion control, adapting to
2051 qualitative changes of the connection medium.
2052
2053 ### Reliability ###
2054
2055 In IP networking, packets can be lost, duplicated, or delivered
2056 out of order, and different network protocols handle these
2057 problems in different ways. We call a transport-layer protocol
2058 _reliable_, if it turns the unreliable IP delivery into an ordered,
2059 duplicate- and loss-free delivery of packets. Sequence numbers
2060 are used to discard duplicates and re-arrange packets delivered
2061 out-of-order. Retransmission is used to guarantee loss-free
2062 delivery. Unreliable protocols, in contrast, do not guarantee ordering
2063 or data integrity.
2064
2065 ### Classification ###
2066
2067 With these definitions the protocols which are used by paraslash for
2068 steaming audio data may be classified as follows.
2069
2070 - HTTP/TCP: connection-oriented, reliable,
2071 - UDP: connectionless, unreliable,
2072 - DCCP: connection-oriented, unreliable.
2073
2074 Below we give a short descriptions of these protocols.
2075
2076 ### TCP ###
2077
2078 The _Transmission Control Protocol_ provides reliable, ordered delivery
2079 of a stream and a classic window-based congestion control. In contrast
2080 to UDP and DCCP (see below), TCP does not have record-oriented or
2081 datagram-based syntax, i.e. it provides a stream which is unaware
2082 and independent of any record (packet) boundaries. TCP is used
2083 extensively by many application layers. Besides HTTP (the Hypertext
2084 Transfer Protocol), also FTP (the File Transfer protocol), SMTP (Simple
2085 Mail Transfer Protocol), SSH (Secure Shell) all sit on top of TCP.
2086
2087 ### UDP ###
2088
2089 The _User Datagram Protocol_ is the simplest transport-layer protocol,
2090 built as a thin layer directly on top of IP. For this reason, it offers
2091 the same best-effort service as IP itself, i.e. there is no detection
2092 of duplicate or reordered packets. Being a connectionless protocol,
2093 only minimal internal state about the connection is maintained, which
2094 means that there is no protection against packet loss or network
2095 congestion. Error checking and correction (if at all) are performed
2096 in the application.
2097
2098 ### DCCP ###
2099
2100 The _Datagram Congestion Control Protocol_ combines the
2101 connection-oriented state maintenance known from TCP with the
2102 unreliable, datagram-based transport of UDP. This means that it
2103 is capable of reacting to changes in the connection by performing
2104 congestion control, offering multiple alternative approaches. But it
2105 is bound to datagram boundaries (the maximum packet size supported
2106 by a medium), and like UDP it lacks retransmission to protect
2107 against loss. Due to the use of sequence numbers, it is however
2108 able to react to loss (interpreted as a congestion indication) and
2109 to ignore out-of-order and duplicate packets. Unlike TCP it allows
2110 to negotiate specific, binding features for a connection, such as
2111 the choice of congestion control: classic, window-based congestion
2112 control known from TCP is available as CCID-2, rate-based, "smooth"
2113 congestion control is offered as CCID-3.
2114
2115 ### HTTP ###
2116
2117 The _Hypertext Transfer Protocol_ is an application layer protocol
2118 on top of TCP. It is spoken by web servers and is most often used
2119 for web services. However, as can be seen by the many Internet radio
2120 stations and YouTube/Flash videos, http is by far not limited to the
2121 delivery of web pages only. Being a simple request/response based
2122 protocol, the semantics of the protocol also allow the delivery of
2123 multimedia content, such as audio over http.
2124
2125 ### Multicast ###
2126
2127 IP multicast is not really a protocol but a technique for one-to-many
2128 communication over an IP network. The challenge is to deliver
2129 information to a group of destinations simultaneously using the
2130 most efficient strategy to send the messages over each link of the
2131 network only once. This has benefits for streaming multimedia: the
2132 standard one-to-one unicast offered by TCP/DCCP means that n clients
2133 listening to the same stream also consume n-times the resources,
2134 whereas multicast requires to send the stream just once, irrespective
2135 of the number of receivers. Since it would be costly to maintain state
2136 for each listening receiver, multicast often implies connectionless
2137 transport, which is the reason that it is currently only available
2138 via UDP.
2139
2140 Abstract socket namespace
2141 -------------------------
2142 UNIX domain sockets are a traditional way to communicate between
2143 processes on the same machine. They are always reliable (see above)
2144 and don't reorder datagrams. Unlike TCP and UDP, UNIX domain sockets
2145 support passing open file descriptors or process credentials to
2146 other processes.
2147
2148 The usual way to set up a UNIX domain socket (as obtained from
2149 socket(2)) for listening is to first bind the socket to a file system
2150 pathname and then call listen(2), then accept(2). Such sockets are
2151 called _pathname sockets_ because bind(2) creates a special socket
2152 file at the specified path. Pathname sockets allow unrelated processes
2153 to communicate with the listening process by binding to the same path
2154 and calling connect(2).
2155
2156 There are two problems with pathname sockets:
2157
2158 * The listing process must be able to (safely) create the
2159 socket special in a directory which is also accessible to
2160 the connecting process.
2161
2162 * After an unclean shutdown of the listening process, a stale
2163 socket special may reside on the file system.
2164
2165 The abstract socket namespace is a non-portable Linux feature which
2166 avoids these problems. Abstract sockets are still bound to a name,
2167 but the name has no connection with file system pathnames.
2168
2169 License
2170 -------
2171
2172 Paraslash is licensed under the GPL, version 2. Most of the code
2173 base has been written from scratch, and those parts are GPL V2
2174 throughout. Notable exceptions are FEC and the WMA decoder. See the
2175 corresponding source files for licencing details for these parts. Some
2176 code sniplets of several other third party software packages have
2177 been incorporated into the paraslash sources, for example log message
2178 coloring was taken from the git sources. These third party software
2179 packages are all published under the GPL or some other license
2180 compatible to the GPL.
2181
2182 Acknowledgements
2183 ----------------
2184
2185 Many thanks to Gerrit Renker who read an early draft of this manual
2186 and contributed significant improvements.
2187
2188 ==========
2189 References
2190 ==========
2191
2192 Articles
2193 --------
2194 - [Polynomial Codes over Certain Finite
2195 Fields](http://kom.aau.dk/~heb/kurser/NOTER/KOFA01.PDF) by Reed, Irving
2196 S.; Solomon, Gustave (1960), Journal of the Society for Industrial
2197 and Applied Mathematics (SIAM) 8 (2): 300-304, doi:10.1137/0108018)
2198
2199 RFCs
2200 ----
2201
2202 - [RFC 768](http://www.ietf.org/rfc/rfc768.txt) (1980): User Datagram
2203 Protocol
2204
2205 - [RFC 791](http://www.ietf.org/rfc/rfc791.txt) (1981): Internet
2206 Protocol
2207
2208 - [RFC 2437](http://www.ietf.org/rfc/rfc2437.txt) (1998): RSA
2209 Cryptography Specifications
2210
2211 - [RFC 4340](http://www.ietf.org/rfc/rfc4340.txt) (2006): Datagram
2212 Congestion Control Protocol (DCCP)
2213
2214 - [RFC 4341](http://www.ietf.org/rfc/rfc4341.txt) (2006): Congestion
2215 Control ID 2: TCP-like Congestion Control
2216
2217 - [RFC 4342](http://www.ietf.org/rfc/rfc4342.txt) (2006): Congestion
2218 Control ID 3: TCP-Friendly Rate Control (TFRC)
2219
2220 - [RFC 6716](http://www.ietf.org/rfc/rfc6716.txt) (2012): Definition
2221 of the Opus Audio Codec
2222
2223 Application web pages
2224 ---------------------
2225
2226 - [paraslash](http://people.tuebingen.mpg.de/maan/paraslash/)
2227 - [xmms](http://xmms2.org/wiki/Main_Page)
2228 - [mpg123](http://www.mpg123.de/)
2229 - [gstreamer](http://gstreamer.freedesktop.org/)
2230 - [icecast](http://www.icecast.org/)
2231 - [Audio Compress](http://beesbuzz.biz/code/audiocompress.php)
2232
2233 External documentation
2234 ----------------------
2235
2236 - [The mathematics of
2237 Raid6](http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf)
2238 by H. Peter Anvin
2239
2240 - [Effective Erasure Codes for reliable Computer Communication
2241 Protocols](http://info.iet.unipi.it/~luigi/fec_ccr.ps.gz) by Luigi
2242 Rizzo
2243
2244 Code
2245 ----
2246 - [Original FEC
2247 implementation](http://info.iet.unipi.it/~luigi/vdm.tar.gz) by
2248 Luigi Rizzo)