manual: Remove text about permissions of /var/paraslash.
[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 OSS.
156 para_write can also be used as a stand-alone WAV or raw audio player.
157
158 ### para_play ###
159
160 A command line audio player.
161
162 ### para_gui ###
163
164 Curses-based gui that presents status information obtained in a curses
165 window. Appearance can be customized via themes. para_gui provides
166 key-bindings for the most common server commands and new key-bindings
167 can be added easily.
168
169 ### para_mixer ###
170
171 An alarm clock and volume-fader for OSS and ALSA.
172
173 ===========
174 Quick start
175 ===========
176
177 This chapter lists the [necessary software](#Requirements)
178 that must be installed to compile the paraslash package, describes
179 how to [compile and install](#Installation) the paraslash
180 source code and the steps that have to be performed in order to
181 [set up](#Configuration) a typical server and client.
182
183 Requirements
184 ------------
185 ### For the impatient ###
186
187 git clone git://git.tuebingen.mpg.de/lopsub
188 cd lopsub && make && sudo make install
189 git clone git://git.tuebingen.mpg.de/osl
190 cd osl && make && sudo make install && sudo ldconfig
191 sudo apt-get install autoconf libssl-dev m4 \
192 libmad0-dev libid3tag0-dev libasound2-dev libvorbis-dev \
193 libfaad-dev libspeex-dev libFLAC-dev libsamplerate-dev realpath \
194 libasound2-dev libao-dev libreadline-dev libncurses-dev \
195 libopus-dev
196
197 ### Detailed description ###
198
199 In any case you will need
200
201 - [libosl](http://people.tuebingen.mpg.de/maan/osl/). The _object
202 storage layer_ library is used by para_server. To clone the source
203 code repository, execute
204
205 git clone git://git.tuebingen.mpg.de/osl
206
207 - [lopsub](http://people.tuebingen.mpg.de/maan/lopsub/). The long
208 option parser for subcommands generates the command line and config
209 file parsers for all paraslash executables. Clone the source code
210 repository with
211
212 git clone git://git.tuebingen.mpg.de/lopsub
213
214 - [gcc](ftp://ftp.gnu.org/pub/gnu/gcc) or
215 [clang](http://clang.llvm.org). All gcc versions >= 4.2 are currently
216 supported. Clang version 1.1 or newer should work as well.
217
218 - [gnu make](ftp://ftp.gnu.org/pub/gnu/make) is also shipped with the
219 disto. On BSD systems the gnu make executable is often called gmake.
220
221 - [bash](ftp://ftp.gnu.org/pub/gnu/bash). Some scripts which run
222 during compilation require the _Bourne again shell_. It is most
223 likely already installed.
224
225 - [m4](ftp://ftp.gnu.org/pub/gnu/m4/). Some source files are generated
226 from templates by the m4 macro processor.
227
228 Optional:
229
230 - [openssl](http://www.openssl.org/) or
231 [libgcrypt](ftp://ftp.gnupg.org/gcrypt/libgcrypt/). At least one
232 of these two libraries is needed as the backend for cryptographic
233 routines on both the server and the client side. Both openssl and
234 libgcrypt are usually shipped with the distro, but you might have
235 to install the development package (`libssl-dev` or `libgcrypt-dev`
236 on debian systems) as well.
237
238 - [libmad](http://www.underbit.com/products/mad/). To compile in MP3
239 support for paraslash, the development package must be installed. It
240 is called `libmad0-dev` on debian-based systems. Note that libmad is
241 not necessary on the server side, i.e., for sending MP3 files.
242
243 - [libid3tag](http://www.underbit.com/products/mad/). For version-2
244 ID3 tag support, you willl need the libid3tag development package
245 `libid3tag0-dev`. Without libid3tag, only version-1 tags are
246 recognized. The mp3 tagger also needs this library for modifying
247 (id3v1 and id3v2) tags.
248
249 - [ogg vorbis](http://www.xiph.org/downloads/). For ogg vorbis streams
250 you need libogg, libvorbis, libvorbisfile. The corresponding Debian
251 packages are called `libogg-dev` and `libvorbis-dev`.
252
253 - [libfaad and mp4ff](http://www.audiocoding.com/). For aac files
254 (m4a) you need libfaad and libmp4ff (package: `libfaad-dev`). Note
255 that for some distributions, e.g. Ubuntu, mp4ff is not part of the
256 libfaad package. Install the faad library from sources (available
257 through the above link) to get the mp4ff library and header files.
258
259 - [speex](http://www.speex.org/). In order to stream or decode speex
260 files, libspeex (`libspeex-dev`) is required.
261
262 - [flac](http://flac.sourceforge.net/). To stream or decode files
263 encoded with the _Free Lossless Audio Codec_, libFLAC (`libFLAC-dev`)
264 must be installed.
265
266 - [libsamplerate](http://www.mega-nerd.com/SRC/index.html). The
267 resample filter will only be compiled if this library is
268 installed. Debian package: `libsamplerate-dev`.
269
270 - [alsa-lib](ftp://ftp.alsa-project.org/pub/lib/). On Linux, you will
271 need to have the ALSA development package `libasound2-dev` installed.
272
273 - [libao](http://downloads.xiph.org/releases/ao/). Needed to build
274 the ao writer (ESD, PulseAudio,...). Debian package: `libao-dev`.
275
276 - [curses](ftp://ftp.gnu.org/pub/gnu/ncurses). Needed for
277 para_gui. Debian package: `libncurses-dev`.
278
279 - [GNU
280 Readline](http://cnswww.cns.cwru.edu/php/chet/readline/rltop.html). If
281 this library (`libreadline-dev`) is installed, para_client, para_audioc
282 and para_play support interactive sessions.
283
284 Installation
285 ------------
286 To build the sources from a tarball, execute
287
288 ./configure && make
289
290 To build from git or a gitweb snapshot, run this command instead:
291
292 ./autogen.sh
293
294 There should be no errors but probably some warnings about missing
295 packages which usually implies that not all audio formats will be
296 supported. If headers or libs are installed at unusual locations you
297 might need to tell the configure script where to find them. Try
298
299 ./configure --help
300
301 to see a list of options. If the paraslash package was compiled
302 successfully, execute (optionally)
303
304 make test
305
306 to run the paraslash test suite. If all tests pass, execute as root
307
308 make install
309
310 to install executables under /usr/local/bin and the man pages under
311 /usr/local/man.
312
313 Configuration
314 -------------
315
316 ### Create a paraslash user ###
317
318 In order to control para_server at runtime you must create a paraslash
319 user. As authentication is based on the RSA crypto system you'll have
320 to create an RSA key pair. If you already have a user and an RSA key
321 pair, you may skip this step.
322
323 In this section we'll assume a typical setup: You would like to run
324 para_server on some host called server_host as user foo, and you want
325 to connect to para_server from another machine called client_host as
326 user bar.
327
328 As foo@server_host, create ~/.paraslash/server.users by typing the
329 following commands:
330
331 user=bar
332 target=~/.paraslash/server.users
333 key=~/.paraslash/id_rsa.pub.$user
334 perms=AFS_READ,AFS_WRITE,VSS_READ,VSS_WRITE
335 mkdir -p ~/.paraslash
336 echo "user $user $key $perms" >> $target
337
338 Next, change to the "bar" account on client_host and generate the
339 key pair with the commands
340
341 ssh-keygen -q -t rsa -b 2048 -N '' -f $key
342
343 This generates the two files id_rsa and id_rsa.pub in ~/.ssh. Note
344 that para_server won't accept keys shorter than 2048 bits. Moreover,
345 para_client rejects private keys which are world-readable.
346
347 para_server only needs to know the public key of the key pair just
348 created. Copy this public key to server_host:
349
350 src=~/.ssh/id_rsa.pub
351 dest=.paraslash/id_rsa.pub.$LOGNAME
352 scp $src foo@server_host:$dest
353
354 Finally, tell para_client to connect to server_host:
355
356 conf=~/.paraslash/client.conf
357 echo 'hostname server_host' > $conf
358
359
360 ### Start para_server ###
361
362 For this first try, we'll use the info loglevel to make the output
363 of para_server more verbose.
364
365 para_server -l info
366
367 Now you can use para_client to connect to the server and issue
368 commands. Open a new shell as bar@client_host and try
369
370 para_client help
371 para_client si
372
373 to retrieve the list of available commands and some server info.
374 Don't proceed if this doesn't work.
375
376 ### Create and populate the database ###
377
378 An empty database is created with
379
380 para_client init
381
382 This initializes a couple of empty tables under
383 ~/.paraslash/afs_database-0.4. You normally don't need to look at these
384 tables, but it's good to know that you can start from scratch with
385
386 rm -rf ~/.paraslash/afs_database-0.4
387
388 in case something went wrong.
389
390 Next, you need to add some audio files to that database so that
391 para_server knows about them. Choose an absolute path to a directory
392 containing some audio files and add them to the audio file table:
393
394 para_client add /my/mp3/dir
395
396 This might take a while, so it is a good idea to start with a directory
397 containing not too many files. Note that the table only contains data
398 about the audio files found, not the files themselves.
399
400 You may print the list of all known audio files with
401
402 para_client ls
403
404 ### Configure para_audiod ###
405
406 We will have to tell para_audiod that it should receive the audio
407 stream from server_host via http:
408
409 para_audiod -l info -r '.:http -i server_host'
410
411 You should now be able to listen to the audio stream once para_server
412 starts streaming. To activate streaming, execute
413
414 para_client play
415
416 Since no playlist has been specified yet, the "dummy" mode which
417 selects all known audio files is activated automatically. See the
418 section on the [audio file selector](#The.audio.file.selector) for how
419 to use playlists and moods to specify which files should be streamed
420 in which order.
421
422 Troubleshooting
423 ---------------
424
425 To identify streaming problems try to receive, decode and play the
426 stream manually using para_recv, para_filter and para_write as follows.
427 For simplicity we assume that you're running Linux/ALSA and that only
428 MP3 files have been added to the database.
429
430 para_recv -r 'http -i server_host' > file.mp3
431 # (interrupt with CTRL+C after a few seconds)
432 ls -l file.mp3 # should not be empty
433 para_filter -f mp3dec -f wav < file.mp3 > file.wav
434 ls -l file.wav # should be much bigger than file.mp3
435 para_write -w alsa < file.wav
436
437 Double check what is logged by para_server and use the --loglevel
438 option of para_recv, para_filter and para_write to increase verbosity.
439
440 ===============
441 User management
442 ===============
443
444 para_server uses a challenge-response mechanism to authenticate
445 requests from incoming connections, similar to ssh's public key
446 authentication method. Authenticated connections are encrypted using
447 the AES stream cipher in integer counter mode.
448
449 In this chapter we briefly describe RSA and AES, and sketch the
450 [authentication handshake](#Client-server.authentication)
451 between para_client and para_server. User management is discussed
452 in the section on [the user_list file](#The.user_list.file).
453 These sections are all about communication between the client and the
454 server. Connecting para_audiod is a different matter and is described
455 in a [separate section](#Connecting.para_audiod).
456
457 RSA and AES
458 -----------
459
460 A block cipher is a transformation which operates on fixed-length
461 blocks. For symmetric block ciphers the transformation is determined
462 by a single key for both encryption and decryption. For asymmetric
463 block ciphers, on the other hand, the key consists of two parts,
464 called the public key and the private key. A message can be encrypted
465 with either key and only the counterpart of that key can decrypt the
466 message. Asymmetric block ciphers can be used for both signing and
467 encrypting a message.
468
469 RSA is an asymmetric block cipher which is used in many applications,
470 including ssh and gpg. The RSA public key encryption and signatures
471 algorithms are defined in detail in RFC 2437. Paraslash relies on
472 RSA for authentication.
473
474 Stream ciphers XOR the input with a pseudo-random key stream to produce
475 the output. Decryption uses the same function calls as encryption.
476 Any block cipher can be turned into a stream cipher by generating the
477 pseudo-random key stream by encrypting successive values of a counter
478 (counter mode).
479
480 AES, the advanced encryption standard, is a well-known symmetric block
481 cipher. Paraslash employs AES in counter mode as described above to
482 encrypt communications. Since a stream cipher key must not be used
483 twice, a random key is generated for every new connection.
484
485 Client-server authentication
486 ----------------------------
487
488 The authentication handshake between para_client and para_server goes
489 as follows:
490
491 - para_client connects to para_server and sends an authentication
492 request for a user. It does so by connecting to TCP port 2990 of the
493 server host. This port is called the para_server _control port_.
494
495 - para_server accepts the connection and forks a child process which
496 handles the incoming request. The parent process keeps listening on the
497 control port while the child process (also called para_server below)
498 continues as follows.
499
500 - para_server loads the RSA public key of that user, fills a
501 fixed-length buffer with random bytes, encrypts that buffer using the
502 public key and sends the encrypted buffer to the client. The first
503 part of the buffer is the challenge which is used for authentication
504 while the second part is the session key.
505
506 - para_client receives the encrypted buffer and decrypts it with the
507 user's private key, thereby obtaining the challenge buffer and the
508 session key. It sends the SHA1 hash value of the challenge back to
509 para_server and stores the session key for further use.
510
511 - para_server also computes the SHA1 hash of the challenge and compares
512 it against what was sent back by the client.
513
514 - If the two hashes do not match, the authentication has failed and
515 para_server closes the connection.
516
517 - Otherwise the user is considered authenticated and the client is
518 allowed to proceed by sending a command to be executed. From this
519 point on the communication is encrypted using the stream cipher with
520 the session key known to both peers.
521
522 paraslash relies on the quality of the pseudo-random bytes provided
523 by the crypto library (openssl or libgcrypt), on the security of
524 the implementation of the RSA and AES crypto routines and on the
525 infeasibility to invert the SHA1 function.
526
527 Neither para_server or para_client create RSA keys on their
528 own. This has to be done once for each user as sketched in
529 [Quick start](#Quick.start) and discussed in more detail
530 [below](#The.user_list.file).
531
532 The user_list file
533 ------------------
534
535 At startup para_server reads the user list file which contains one
536 line per user. The default location of the user list file may be
537 changed with the --user-list option.
538
539 There should be at least one user in this file. Each user must have
540 an RSA key pair. The public part of the key is needed by para_server
541 while the private key is needed by para_client. Each line of the
542 user list file must be of the form
543
544 user <username> <key> <perms>
545
546 where _username_ is an arbitrary string (usually the user's login
547 name), _key_ is the full path to that user's public RSA key, and
548 _perms_ is a comma-separated list of zero or more of the following
549 permission bits:
550
551 +---------------------------------------------------------+
552 | AFS_READ | read the contents of the databases |
553 +-----------+---------------------------------------------+
554 | AFS_WRITE | change database contents |
555 +-----------+---------------------------------------------+
556 | VSS_READ | obtain information about the current stream |
557 +-----------+---------------------------------------------+
558 | VSS_WRITE | change the current stream |
559 +---------------------------------------------------------+
560
561 The permission bits specify which commands the user is allowed to
562 execute. The output of
563
564 para_client help
565
566 contains the permissions needed to execute the command.
567
568 It is possible to make para_server reread the user_list file by
569 executing the paraslash "hup" command or by sending SIGHUP to the
570 PID of para_server.
571
572 Connecting para_audiod
573 ----------------------
574
575 para_audiod listens on a Unix domain socket. Those sockets are
576 for local communication only, so only local users can connect to
577 para_audiod. The default is to let any user connect but this can be
578 restricted on platforms that support UNIX socket credentials which
579 allow para_audiod to obtain the Unix credentials of the connecting
580 process.
581
582 Use para_audiod's --user-allow option to allow connections only for
583 a limited set of users.
584
585 =======================
586 The audio file selector
587 =======================
588
589 paraslash comes with a sophisticated audio file selector (AFS),
590 whose main task is to determine which file to stream next, based on
591 information on the audio files stored in a database. It communicates
592 also with para_client whenever an AFS command is executed, for example
593 to answer a database query.
594
595 Besides the traditional playlists, AFS supports audio file selection
596 based on _moods_ which act as a filter that limits the set of all
597 known audio files to those which satisfy certain criteria. It also
598 maintains tables containing images (e.g. album cover art) and lyrics
599 that can be associated with one or more audio files.
600
601 AFS employs [libosl](http://people.tuebingen.mpg.de/maan/osl/), the
602 object storage layer library, as the backend library for storing
603 information on audio files, playlists, etc. This library offers
604 functionality similar to a relational database, but is much more
605 lightweight than a full database backend.
606
607 In this chapter we sketch the setup of the [AFS
608 process](#The.AFS.process) during server startup and proceed with the
609 description of the [layout](#Database.layout) of the various database
610 tables. The section on [playlists and moods](#Playlists.and.moods)
611 explains these two audio file selection mechanisms in detail
612 and contains pratical examples. The way [file renames and content
613 changes](#File.renames.and.content.changes) are detected is discussed
614 briefly before the [Troubleshooting](#Troubleshooting) section
615 concludes the chapter.
616
617 The AFS process
618 ---------------
619
620 On startup, para_server forks to create the AFS process which opens
621 the OSL database tables. The server process communicates with the
622 AFS process via pipes and shared memory. Usually, the AFS process
623 awakes only briefly whenever the current audio file changes. The AFS
624 process determines the next audio file, opens it, verifies it has
625 not been changed since it was added to the database and passes the
626 open file descriptor to the server process, along with audio file
627 meta-data such as file name, duration, audio format and so on. The
628 server process then starts to stream the audio file.
629
630 The AFS process also accepts connections from local clients via
631 a well-known socket. However, only child processes of para_server
632 may connect through this socket. All server commands that have the
633 AFS_READ or AFS_WRITE permission bits use this mechanism to query or
634 change the database.
635
636 Database layout
637 ---------------
638
639 ### The audio file table ###
640
641 This is the most important and usually also the largest table of the
642 AFS database. It contains the information needed to stream each audio
643 file. In particular the following data is stored for each audio file.
644
645 - SHA1 hash value of the audio file contents. This is computed once
646 when the file is added to the database. Whenever AFS selects this
647 audio file for streaming the hash value is recomputed and checked
648 against the value stored in the database to detect content changes.
649
650 - The time when this audio file was last played.
651
652 - The number of times the file has been played so far.
653
654 - The attribute bitmask.
655
656 - The image id which describes the image associated with this audio
657 file.
658
659 - The lyrics id which describes the lyrics associated with this
660 audio file.
661
662 - The audio format id (MP3, OGG, ...).
663
664 - An amplification value that can be used by the amplification filter
665 to pre-amplify the decoded audio stream.
666
667 - The chunk table. It describes the location and the timing of the
668 building blocks of the audio file. This is used by para_server to
669 send chunks of the file at appropriate times.
670
671 - The duration of the audio file.
672
673 - Tag information contained in the audio file (ID3 tags, Vorbis
674 comments, ...).
675
676 - The number of channels
677
678 - The encoding bitrate.
679
680 - The sampling frequency.
681
682 To add or refresh the data contained in the audio file table, the _add_
683 command is used. It takes the full path of either an audio file or a
684 directory. In the latter case, the directory is traversed recursively
685 and all files which are recognized as valid audio files are added to
686 the database.
687
688 ### The attribute table ###
689
690 The attribute table contains two columns, _name_ and _bitnum_. An
691 attribute is simply a name for a certain bit number in the attribute
692 bitmask of the audio file table.
693
694 Each of the 64 bits of the attribute bitmask can be set for each
695 audio file individually. Hence up to 64 different attributes may be
696 defined. For example, "pop", "rock", "blues", "jazz", "instrumental",
697 "german_lyrics", "speech", whatever. You are free to choose as
698 many attributes as you like and there are no naming restrictions
699 for attributes.
700
701 A new attribute "test" is created by
702
703 para_client addatt test
704 and
705 para_client lsatt
706
707 lists all available attributes. You can set the "test" attribute for
708 an audio file by executing
709
710 para_client setatt test+ /path/to/the/audio/file
711
712 Similarly, the "test" bit can be removed from an audio file with
713
714 para_client setatt test- /path/to/the/audio/file
715
716 Instead of a path you may use a shell wildcard pattern. The attribute
717 is applied to all audio files matching this pattern:
718
719 para_client setatt test+ '/test/directory/*'
720
721 The command
722
723 para_client -- ls -l=v
724
725 gives you a verbose listing of your audio files also showing which
726 attributes are set.
727
728 In case you wonder why the double-dash in the above command is needed:
729 It tells para_client to not interpret the options after the dashes. If
730 you find this annoying, just say
731
732 alias para='para_client --'
733
734 and be happy. In what follows we shall use this alias.
735
736 The "test" attribute can be dropped from the database with
737
738 para rmatt test
739
740 Read the output of
741
742 para help ls
743 para help setatt
744
745 for more information and a complete list of command line options to
746 these commands.
747
748 ### Blob tables ###
749
750 The image, lyrics, moods and playlists tables are all blob tables.
751 Blob tables consist of three columns each: The identifier which is
752 a positive number that is auto-incremented, the name (an arbitrary
753 string) and the content (the blob).
754
755 All blob tables support the same set of actions: cat, ls, mv, rm
756 and add. Of course, _add_ is used for adding new blobs to the table
757 while the other actions have the same meaning as the corresponding
758 Unix commands. The paraslash commands to perform these actions are
759 constructed as the concatenation of the table name and the action. For
760 example addimg, catimg, lsimg, mvimg, rmimg are the commands that
761 manipulate or query the image table.
762
763 The add variant of these commands is special as these commands read
764 the blob contents from stdin. To add an image to the image table the
765 command
766
767 para addimg image_name < file.jpg
768
769 can be used.
770
771 Note that the images and lyrics are not interpreted at all, and also
772 the playlist and the mood blobs are only investigated when the mood
773 or playlist is activated with the select command.
774
775 ### The score table ###
776
777 The score table describes those audio files which are admissible for
778 the current mood or playlist (see below). The table has two columns:
779 a pointer to a row of the audio file table and a score value.
780
781 Unlike all other tables of the database, the score table remains in
782 memory and is never stored on disk. It is initialized at startup and
783 recomputed when the select command loads a new mood or playlist.
784
785 When the audio file selector is asked to open the next audio file,
786 it picks the row with the highest score, opens the corresponding
787 file and passes the file descriptor to the virtual streaming system.
788 At this point the last_played and the num_played fields of the selected
789 file are updated and the score is recomputed.
790
791 Playlists and moods
792 -------------------
793
794 Playlists and moods offer two different ways of specifying the set of
795 admissible files. A playlist in itself describes a set of admissible
796 files. A mood, in contrast, describes the set of admissible files in
797 terms of attributes and other type of information available in the
798 audio file table. As an example, a mood can define a filename pattern,
799 which is then matched against the names of audio files in the table.
800
801 ### Playlists ###
802
803 Playlists are accommodated in the playlist table of the afs database,
804 using the aforementioned blob format for tables. A new playlist is
805 created with the addpl command by specifying the full (absolute)
806 paths of all desired audio files, separated by newlines. Example:
807
808 find /my/mp3/dir -name "*.mp3" | para addpl my_playlist
809
810 If _my_playlist_ already exists it is overwritten. To activate the
811 new playlist, execute
812
813 para select p/my_playlist
814
815 The audio file selector will assign scores to each entry of the list,
816 in descending order so that files will be selected in order. If a
817 file could not be opened for streaming, its entry is removed from
818 the score table (but not from the playlist).
819
820 ### Moods ###
821
822 A mood consists of a unique name and its *mood definition*, which is
823 a set of *mood lines* containing expressions in terms of attributes
824 and other data contained in the database.
825
826 At any time at most one mood can be *active* which means that
827 para_server is going to select only files from that subset of
828 admissible files.
829
830 So in order to create a mood definition one has to write a set of
831 mood lines. Mood lines come in three flavours: Accept lines, deny
832 lines and score lines.
833
834 The general syntax of the three types of mood lines is
835
836
837 accept [with score <score>] [if] [not] <mood_method> [options]
838 deny [with score <score>] [if] [not] <mood_method> [options]
839 score <score> [if] [not] <mood_method> [options]
840
841
842 Here <score> is either an integer or the string "random" which assigns
843 a random score to all matching files. The score value changes the
844 order in which admissible files are going to be selected, but is of
845 minor importance for this introduction.
846
847 So we concentrate on the first two forms, i.e. accept and deny
848 lines. As usual, everything in square brackets is optional, i.e.
849 accept/deny lines take the following form when ignoring scores:
850
851 accept [if] [not] <mood_method> [options]
852
853 and analogously for the deny case. The "if" keyword is only syntactic
854 sugar and has no function. The "not" keyword just inverts the result,
855 so the essence of a mood line is the mood method part and the options
856 following thereafter.
857
858 A *mood method* is realized as a function which takes an audio file
859 and computes a number from the data contained in the database.
860 If this number is non-negative, we say the file *matches* the mood
861 method. The file matches the full mood line if it either
862
863 - matches the mood method and the "not" keyword is not given,
864 or
865 - does not match the mood method, but the "not" keyword is given.
866
867 The set of admissible files for the whole mood is now defined as those
868 files which match at least one accept mood line, but no deny mood line.
869 More formally, an audio file F is admissible if and only if
870
871 (F ~ AL1 or F ~ AL2...) and not (F ~ DL1 or F ~ DN2 ...)
872
873 where AL1, AL2... are the accept lines, DL1, DL2... are the deny
874 lines and "~" means "matches".
875
876 The cases where no mood lines of accept/deny type are defined need
877 special treatment:
878
879 - Neither accept nor deny lines: This treats all files as
880 admissible (in fact, that is the definition of the dummy mood
881 which is activated automatically if no moods are available).
882
883 - Only accept lines: A file is admissible iff it matches at
884 least one accept line:
885
886 F ~ AL1 or F ~ AL2 or ...
887
888 - Only deny lines: A file is admissible iff it matches no
889 deny line:
890
891 not (F ~ DL1 or F ~ DN2 ...)
892
893
894
895 ### List of mood_methods ###
896
897 no_attributes_set
898
899 Takes no arguments and matches an audio file if and only if no
900 attributes are set.
901
902 is_set <attribute_name>
903
904 Takes the name of an attribute and matches iff that attribute is set.
905
906 path_matches <pattern>
907
908 Takes a filename pattern and matches iff the path of the audio file
909 matches the pattern.
910
911 artist_matches <pattern>
912 album_matches <pattern>
913 title_matches <pattern>
914 comment_matches <pattern>
915
916 Takes an extended regular expression and matches iff the text of the
917 corresponding tag of the audio file matches the pattern. If the tag
918 is not set, the empty string is matched against the pattern.
919
920 year ~ <num>
921 bitrate ~ <num>
922 frequency ~ <num>
923 channels ~ <num>
924 num_played ~ <num>
925 image_id ~ <num>
926 lyrics_id ~ <num>
927
928 Takes a comparator ~ of the set {<, =, <=, >, >=, !=} and a number
929 <num>. Matches an audio file iff the condition <val> ~ <num> is
930 satisfied where val is the corresponding value of the audio file
931 (value of the year tag, bitrate in kbit/s, etc.).
932
933 The year tag is special as its value is undefined if the audio file
934 has no year tag or the content of the year tag is not a number. Such
935 audio files never match. Another difference is the special treatment
936 if the year tag is a two-digit number. In this case either 1900 or
937 2000 is added to the tag value, depending on whether the number is
938 greater than 2000 plus the current year.
939
940
941 ### Mood usage ###
942
943 To create a new mood called "my_mood", write its definition into
944 some temporary file, say "tmpfile", and add it to the mood table
945 by executing
946
947 para addmood my_mood < tmpfile
948
949 If the mood definition is really short, you may just pipe it to the
950 client instead of using temporary files. Like this:
951
952 echo "$MOOD_DEFINITION" | para addmood my_mood
953
954 There is no need to keep the temporary file since you can always use
955 the catmood command to get it back:
956
957 para catmood my_mood
958
959 A mood can be activated by executing
960
961 para select m/my_mood
962
963 Once active, the list of admissible files is shown by the ls command
964 if the "-a" switch is given:
965
966 para ls -a
967
968
969 ### Example mood definition ###
970
971 Suppose you have defined attributes "punk" and "rock" and want to define
972 a mood containing only Punk-Rock songs. That is, an audio file should be
973 admissible if and only if both attributes are set. Since
974
975 punk and rock
976
977 is obviously the same as
978
979 not (not punk or not rock)
980
981 (de Morgan's rule), a mood definition that selects only Punk-Rock
982 songs is
983
984 deny if not is_set punk
985 deny if not is_set rock
986
987
988
989 File renames and content changes
990 --------------------------------
991
992 Since the audio file selector knows the SHA1 of each audio file that
993 has been added to the afs database, it recognizes if the content of
994 a file has changed, e.g. because an ID3 tag was added or modified.
995 Also, if a file has been renamed or moved to a different location,
996 afs will detect that an entry with the same hash value already exists
997 in the audio file table.
998
999 In both cases it is enough to just re-add the new file. In the
1000 first case (file content changed), the audio table is updated, while
1001 metadata such as the num_played and last_played fields, as well as
1002 the attributes, remain unchanged. In the other case, when the file
1003 is moved or renamed, only the path information is updated, all other
1004 data remains as before.
1005
1006 It is possible to change the behaviour of the add command by using the
1007 "-l" (lazy add) or the "-f" (force add) option.
1008
1009 Troubleshooting
1010 ---------------
1011
1012 Use the debug loglevel (-l debug) to show debugging info. All paraslash
1013 executables have a brief online help which is displayed when -h is
1014 given. The --detailed-help option prints the full help text.
1015
1016 If para_server crashed or was killed by SIGKILL (signal 9), it
1017 may refuse to start again because of "dirty osl tables". In this
1018 case you'll have to run the oslfsck program of libosl to fix your
1019 database:
1020
1021 oslfsck -fd ~/.paraslash/afs_database-0.4
1022
1023 However, make sure para_server isn't running before executing oslfsck.
1024
1025 If you don't mind to recreate your database you can start
1026 from scratch by removing the entire database directory, i.e.
1027
1028 rm -rf ~/.paraslash/afs_database-0.4
1029
1030 Be aware that this removes all attribute definitions, all playlists
1031 and all mood definitions and requires to re-initialize the tables.
1032
1033 Although oslfsck fixes inconsistencies in database tables it doesn't
1034 care about the table contents. To check for invalid table contents, use
1035
1036 para_client check
1037
1038 This prints out references to missing audio files as well as invalid
1039 playlists and mood definitions.
1040
1041 Similarly, para_audiod refuses to start if its socket file exists, since
1042 this indicates that another instance of para_audiod is running. After
1043 a crash a stale socket file might remain and you must run
1044
1045 para_audiod --force
1046
1047 once to fix it up.
1048
1049 =======================================
1050 Audio formats and audio format handlers
1051 =======================================
1052
1053 Audio formats
1054 -------------
1055
1056 The following audio formats are supported by paraslash:
1057
1058 ### MP3 ###
1059
1060 Mp3, MPEG-1 Audio Layer 3, is a common audio format for audio storage,
1061 designed as part of its MPEG-1 standard. An MP3 file is made up of
1062 multiple MP3 frames, which consist of a header and a data block. The
1063 size of an MP3 frame depends on the bit rate and on the number
1064 of channels. For a typical CD-audio file (sample rate of 44.1 kHz
1065 stereo), encoded with a bit rate of 128 kbit, an MP3 frame is about
1066 400 bytes large.
1067
1068 ### OGG/Vorbis ###
1069
1070 OGG is a standardized audio container format, while Vorbis is an
1071 open source codec for lossy audio compression. Since Vorbis is most
1072 commonly made available via the OGG container format, it is often
1073 referred to as OGG/Vorbis. The OGG container format divides data into
1074 chunks called OGG pages. A typical OGG page is about 4KB large. The
1075 Vorbis codec creates variable-bitrate (VBR) data, where the bitrate
1076 may vary considerably.
1077
1078 ### OGG/Speex ###
1079
1080 Speex is an open-source speech codec that is based on CELP (Code
1081 Excited Linear Prediction) coding. It is designed for voice
1082 over IP applications, has modest complexity and a small memory
1083 footprint. Wideband and narrowband (telephone quality) speech are
1084 supported. As for Vorbis audio, Speex bit-streams are often stored
1085 in OGG files. As of 2012 this codec is considered obsolete since the
1086 Oppus codec, described below, surpasses its performance in all areas.
1087
1088 ### OGG/Opus ###
1089
1090 Opus is a lossy audio compression format standardized through RFC
1091 6716 in 2012. It combines the speech-oriented SILK codec and the
1092 low-latency CELT (Constrained Energy Lapped Transform) codec. Like
1093 OGG/Vorbis and OGG/Speex, Opus data is usually encapsulated in OGG
1094 containers. All known software patents which cover Opus are licensed
1095 under royalty-free terms.
1096
1097 ### AAC ###
1098
1099 Advanced Audio Coding (AAC) is a standardized, lossy compression
1100 and encoding scheme for digital audio which is the default audio
1101 format for Apple's iPhone, iPod, iTunes. Usually MPEG-4 is used as
1102 the container format and audio files encoded with AAC have the .m4a
1103 extension. A typical AAC frame is about 700 bytes large.
1104
1105 ### WMA ###
1106
1107 Windows Media Audio (WMA) is an audio data compression technology
1108 developed by Microsoft. A WMA file is usually encapsulated in the
1109 Advanced Systems Format (ASF) container format, which also specifies
1110 how meta data about the file is to be encoded. The bit stream of WMA
1111 is composed of superframes, each containing one or more frames of
1112 2048 samples. For 16 bit stereo a WMA superframe is about 8K large.
1113
1114 ### FLAC ###
1115
1116 The Free Lossless Audio Codec (FLAC) compresses audio without quality
1117 loss. It gives better compression ratios than a general purpose
1118 compressor like zip or bzip2 because FLAC is designed specifically
1119 for audio. A FLAC-encoded file consists of frames of varying size, up
1120 to 16K. Each frame starts with a header that contains all information
1121 necessary to decode the frame.
1122
1123 Meta data
1124 ---------
1125
1126 Unfortunately, each audio format has its own conventions how meta
1127 data is added as tags to the audio file.
1128
1129 For MP3 files, ID3, version 1 and 2 are widely used. ID3 version 1
1130 is rather simple but also very limited as it supports only artist,
1131 title, album, year and comment tags. Each of these can only be at most
1132 32 characters long. ID3, version 2 is much more flexible but requires
1133 a separate library being installed for paraslash to support it.
1134
1135 Ogg vorbis, ogg speex and flac files contain meta data as Vorbis
1136 comments, which are typically implemented as strings of the form
1137 "[TAG]=[VALUE]". Unlike ID3 version 1 tags, one may use whichever
1138 tags are appropriate for the content.
1139
1140 AAC files usually use the MPEG-4 container format for storing meta
1141 data while WMA files wrap meta data as special objects within the
1142 ASF container format.
1143
1144 paraslash only tracks the most common tags that are supported by
1145 all tag variants: artist, title, year, album, comment. When a file
1146 is added to the AFS database, the meta data of the file is extracted
1147 and stored in the audio file table.
1148
1149 Chunks and chunk tables
1150 -----------------------
1151
1152 paraslash uses the word "chunk" as common term for the building blocks
1153 of an audio file. For MP3 files, a chunk is the same as an MP3 frame,
1154 while for OGG files a chunk is an OGG page, etc. Therefore the chunk
1155 size varies considerably between audio formats, from a few hundred
1156 bytes (MP3) up to 16K (FLAC).
1157
1158 The chunk table contains the offsets within the audio file that
1159 correspond to the chunk boundaries of the file. Like the meta data,
1160 the chunk table is computed and stored in the database whenever an
1161 audio file is added.
1162
1163 The paraslash senders (see below) always send complete chunks. The
1164 granularity for seeking is therefore determined by the chunk size.
1165
1166 Audio format handlers
1167 ---------------------
1168
1169 For each audio format paraslash contains an audio format handler whose
1170 first task is to tell whether a given file is a valid audio file of
1171 this type. If so, the audio file handler extracts some technical data
1172 (duration, sampling rate, number of channels etc.), computes the
1173 chunk table and reads the meta data.
1174
1175 The audio format handler code is linked into para_server and executed
1176 via the _add_ command. The same code is also available as a stand-alone
1177 tool, para_afh, which prints the technical data, the chunk table
1178 and the meta data of a file. Moreover, all audio format handlers are
1179 combined in the afh receiver which is part of para_recv and para_play.
1180
1181 ==========
1182 Networking
1183 ==========
1184
1185 Paraslash uses different network connections for control and data.
1186 para_client communicates with para_server over a dedicated TCP control
1187 connection. To transport audio data, separate data connections are
1188 used. For these data connections, a variety of transports (UDP, DCCP,
1189 HTTP) can be chosen.
1190
1191 The chapter starts with the [control
1192 service](#The.paraslash.control.service), followed by a section
1193 on the various [streaming protocols](#Streaming.protocols)
1194 in which the data connections are described. The way
1195 audio file headers are embedded into the stream is discussed
1196 [briefly](#Streams.with.headers.and.headerless.streams) before the
1197 [example section](#Networking.examples) which illustrates typical
1198 commands for real-life scenarios.
1199
1200 Both IPv4 and IPv6 are supported.
1201
1202 The paraslash control service
1203 -----------------------------
1204
1205 para_server is controlled at runtime via the paraslash control
1206 connection. This connection is used for server commands (play, stop,
1207 ...) as well as for afs commands (ls, select, ...).
1208
1209 The server listens on a TCP port and accepts connections from clients
1210 that connect the open port. Each connection causes the server to fork
1211 off a client process which inherits the connection and deals with that
1212 client only. In this classical accept/fork approach the server process
1213 is unaffected if the child dies or goes crazy for whatever reason. In
1214 fact, the child process can not change address space of server process.
1215
1216 The section on [client-server
1217 authentication](#Client-server.authentication) above described the
1218 early connection establishment from the crypto point of view. Here
1219 it is described what happens after the connection (including crypto
1220 setup) has been established. There are four processes involved during
1221 command dispatch as sketched in the following diagram.
1222
1223 server_host client_host
1224 ~~~~~~~~~~~ ~~~~~~~~~~~
1225
1226 +-----------+ connect +-----------+
1227 |para_server|<------------------------------ |para_client|
1228 +-----------+ +-----------+
1229 | ^
1230 | fork +---+ |
1231 +----------> |AFS| |
1232 | +---+ |
1233 | ^ |
1234 | | |
1235 | | connect (cookie) |
1236 | | |
1237 | | |
1238 | fork +-----+ inherited connection |
1239 +---------->|child|<--------------------------+
1240 +-----+
1241
1242 Note that the child process is not a child of the afs process,
1243 so communication of these two processes has to happen via local
1244 sockets. In order to avoid abuse of the local socket by unrelated
1245 processes, a magic cookie is created once at server startup time just
1246 before the server process forks off the AFS process. This cookie is
1247 known to the server, AFS and the child, but not to unrelated processes.
1248
1249 There are two different kinds of commands: First there are commands
1250 that cause the server to respond with some answer such as the list
1251 of all audio files. All but the addblob commands (addimg, addlyr,
1252 addpl, addmood) are of this kind. The addblob commands add contents
1253 to the database, so they need to transfer data the other way round,
1254 from the client to the server.
1255
1256 There is no knowledge about the server commands built into para_client,
1257 so it does not know about addblob commands. Instead, the server sends
1258 a special "awaiting data" packet for these commands. If the client
1259 receives this packet, it sends STDIN to the server, otherwise it
1260 dumps data from the server to STDOUT.
1261
1262 Streaming protocols
1263 -------------------
1264
1265 A network (audio) stream usually consists of one streaming source,
1266 the _sender_, and one or more _receivers_ which read data over the
1267 network from the streaming source.
1268
1269 Senders are thus part of para_server while receivers are part of
1270 para_audiod. Moreover, there is the stand-alone tool para_recv which
1271 can be used to manually download a stream, either from para_server
1272 or from a web-based audio streaming service.
1273
1274 The following three streaming protocols are supported by paraslash:
1275
1276 - HTTP. Recommended for public streams that can be played by any
1277 player like mpg123, xmms, itunes, winamp, etc. The HTTP sender is
1278 supported on all operating systems and all platforms.
1279
1280 - DCCP. Recommended for LAN streaming. DCCP is currently available
1281 only for Linux.
1282
1283 - UDP. Recommended for multicast LAN streaming.
1284
1285 See the Appendix on [network protocols](/#Network.protocols)
1286 for brief descriptions of the various protocols relevant for network
1287 audio streaming with paraslash.
1288
1289 It is possible to activate more than one sender simultaneously.
1290 Senders can be controlled at run time and via config file and command
1291 line options.
1292
1293 Note that audio connections are _not_ encrypted. Transport or Internet
1294 layer encryption should be used if encrypted data connections are
1295 needed.
1296
1297 Since DCCP and TCP are both connection-oriented protocols, connection
1298 establishment/teardown and access control are very similar between
1299 these two streaming protocols. UDP is the most lightweight option,
1300 since in contrast to TCP/DCCP it is connectionless. It is also the
1301 only protocol supporting IP multicast.
1302
1303 The HTTP and the DCCP sender listen on a (TCP/DCCP) port waiting for
1304 clients to connect and establish a connection via some protocol-defined
1305 handshake mechanism. Both senders maintain two linked lists each:
1306 The list of all clients which are currently connected, and the list
1307 of access control entries which determines who is allowed to connect.
1308 IP-based access control may be configured through config file and
1309 command line options and via the "allow" and "deny" sender subcommands.
1310
1311 Upon receiving a GET request from the client, the HTTP sender sends
1312 back a status line and a message. The body of this message is the
1313 audio stream. This is common practice and is supported by many popular
1314 clients which can thus be used to play a stream offered by para_server.
1315 For DCCP things are a bit simpler: No messages are exchanged between
1316 the receiver and sender. The client simply connects and the sender
1317 starts to stream.
1318
1319 DCCP is an experimental protocol which offers a number of new features
1320 not available for TCP. Both ends can negotiate these features using
1321 a built-in negotiation mechanism. In contrast to TCP/HTTP, DCCP is
1322 datagram-based (no retransmissions) and thus should not be used over
1323 lossy media (e.g. WiFi networks). One useful feature offered by DCCP
1324 is access to a variety of different congestion-control mechanisms
1325 called CCIDs. Two different CCIDs are available per default on Linux:
1326
1327
1328 - _CCID 2_. A Congestion Control mechanism similar to that of TCP. The
1329 sender maintains a congestion window and halves this window in response
1330 to congestion.
1331
1332
1333 - _CCID-3_. Designed to be fair when competing for bandwidth.
1334 It has lower variation of throughput over time compared with TCP,
1335 which makes it suitable for streaming media.
1336
1337 Unlike the HTTP and DCCP senders, the UDP sender maintains only a
1338 single list, the _target list_. This list describes the set of clients
1339 to which the stream is sent. There is no list for access control and
1340 no "allow" and "deny" commands for the UDP sender. Instead, the "add"
1341 and "delete" commands can be used to modify the target list.
1342
1343 Since both UDP and DCCP offer an unreliable datagram-based transport,
1344 additional measures are necessary to guard against disruptions over
1345 networks that are lossy or which may be subject to interference (as
1346 is for instance the case with WiFi). Paraslash uses FEC (Forward
1347 Error Correction) to guard against packet losses and reordering. The
1348 stream is FEC-encoded before it is sent through the UDP socket and
1349 must be decoded accordingly on the receiver side.
1350
1351 The packet size and the amount of redundancy introduced by FEC can
1352 be configured via the FEC parameters which are dictated by server
1353 and may also be configured through the "sender" command. The FEC
1354 parameters are encoded in the header of each network packet, so no
1355 configuration is necessary on the receiver side. See the section on
1356 [FEC](#Forward.error.correction) below.
1357
1358 Streams with headers and headerless streams
1359 -------------------------------------------
1360
1361 For OGG/Vorbis, OGG/Speex and wma streams, some of the information
1362 needed to decode the stream is only contained in the audio file
1363 header of the container format but not in each data chunk. Clients
1364 must be able to obtain this information in case streaming starts in
1365 the middle of the file or if para_audiod is started while para_server
1366 is already sending a stream.
1367
1368 This is accomplished in different ways, depending on the streaming
1369 protocol. For connection-oriented streams (HTTP, DCCP) the audio file
1370 header is sent prior to audio file data. This technique however does
1371 not work for the connectionless UDP transport. Hence the audio file
1372 header is periodically being embedded into the UDP audio data stream.
1373 By default, the header is resent after five seconds. The receiver has
1374 to wait until the next header arrives before it can start decoding
1375 the stream.
1376
1377 Networking examples
1378 -------------------
1379
1380 The "si" (server info) command lists some information about the
1381 currently running server process.
1382
1383 -> Show PIDs, number of connected clients, uptime, and more:
1384
1385 para_client si
1386
1387 The sender command of para_server prints information about senders,
1388 like the various access control lists, and it allows to (de-)activate
1389 senders and to change the access permissions at runtime.
1390
1391 -> List all senders
1392
1393 para_client sender
1394
1395 -> Obtain general help for the sender command:
1396
1397 para_client help sender
1398
1399 -> Get help for a specific sender (contains further examples):
1400
1401 s=http # or dccp or udp
1402 para_client sender $s help
1403
1404 -> Show status of the http sender
1405
1406 para_client sender http status
1407
1408 By default para_server activates both the HTTP and th DCCP sender on
1409 startup. This can be changed via command line options or para_server's
1410 config file.
1411
1412 -> List config file options for senders:
1413
1414 para_server -h
1415
1416 All senders share the "on" and "off" commands, so senders may be
1417 activated and deactivated independently of each other.
1418
1419 -> Switch off the http sender:
1420
1421 para_client sender http off
1422
1423 -> Receive a DCCP stream using CCID2 and write the output into a file:
1424
1425 host=foo.org; ccid=2; filename=bar
1426 para_recv --receiver "dccp --host $host --ccid $ccid" > $filename
1427
1428 Note the quotes around the arguments for the dccp receiver. Each
1429 receiver has its own set of command line options and its own command
1430 line parser, so arguments for the dccp receiver must be protected
1431 from being interpreted by para_recv.
1432
1433 -> Start UDP multicast, using the default multicast address:
1434
1435 para_client sender udp add 224.0.1.38
1436
1437 -> Receive FEC-encoded multicast stream and write the output into a file:
1438
1439 filename=foo
1440 para_recv -r udp > $filename
1441
1442 -> Add an UDP unicast for a client to the target list of the UDP sender:
1443
1444 t=client.foo.org
1445 para_client sender udp add $t
1446
1447 -> Receive this (FEC-encoded) unicast stream:
1448
1449 filename=foo
1450 para_recv -r 'udp -i 0.0.0.0' > $filename
1451
1452 -> Create a minimal config for para_audiod for HTTP streams:
1453
1454 c=$HOME/.paraslash/audiod.conf.min; s=server.foo.com
1455 echo receiver \".:http -i $s\" > $c
1456 para_audiod --config $c
1457
1458 =======
1459 Filters
1460 =======
1461
1462 A paraslash filter is a module which transforms an input stream into
1463 an output stream. Filters are included in the para_audiod executable
1464 and in the stand-alone tool para_filter which usually contains the
1465 same modules.
1466
1467 While para_filter reads its input stream from STDIN and writes
1468 the output to STDOUT, the filter modules of para_audiod are always
1469 connected to a receiver which produces the input stream and a writer
1470 which absorbs the output stream.
1471
1472 Some filters depend on a specific library and are not compiled in
1473 if this library was not found at compile time. To see the list of
1474 supported filters, run para_filter and para_audiod with the --help
1475 option. The output looks similar to the following:
1476
1477 Available filters:
1478 compress wav amp fecdec wmadec prebuffer oggdec aacdec mp3dec
1479
1480 Out of these filter modules, a chain of filters can be constructed,
1481 much in the way Unix pipes can be chained, and analogous to the use
1482 of modules in gstreamer: The output of the first filter becomes the
1483 input of the second filter. There is no limitation on the number of
1484 filters and the same filter may occur more than once.
1485
1486 Like receivers, each filter has its own command line options which
1487 must be quoted to protect them from the command line options of
1488 the driving application (para_audiod or para_filter). Example:
1489
1490 para_filter -f 'mp3dec --ignore-crc' -f 'compress --damp 1'
1491
1492 For para_audiod, each audio format has its own set of filters. The
1493 name of the audio format for which the filter should be applied can
1494 be used as the prefix for the filter option. Example:
1495
1496 para_audiod -f 'mp3:prebuffer --duration 300'
1497
1498 The "mp3" prefix above is actually interpreted as a POSIX extended
1499 regular expression. Therefore
1500
1501 para_audiod -f '.:prebuffer --duration 300'
1502
1503 activates the prebuffer filter for all supported audio formats (because
1504 "." matches all audio formats) while
1505
1506 para_audiod -f 'wma|ogg:prebuffer --duration 300'
1507
1508 activates it only for wma and ogg streams.
1509
1510 Decoders
1511 --------
1512
1513 For each supported audio format there is a corresponding filter
1514 which decodes audio data in this format to 16 bit PCM data which
1515 can be directly sent to the sound device or any other software that
1516 operates on undecoded PCM data (visualizers, equalizers etc.). Such
1517 filters are called _decoders_ in general, and xxxdec is the name of
1518 the paraslash decoder for the audio format xxx. For example, the mp3
1519 decoder is called mp3dec.
1520
1521 Note that the output of the decoder is about 10 times larger than
1522 its input. This means that filters that operate on the decoded audio
1523 stream have to deal with much more data than filters that transform
1524 the audio stream before it is fed to the decoder.
1525
1526 Paraslash relies on external libraries for most decoders, so these
1527 libraries must be installed for the decoder to be included in the
1528 executables. For example, the mp3dec filter depends on the mad library.
1529
1530 Forward error correction
1531 ------------------------
1532
1533 As already mentioned [earlier](#Streaming.protocols), paraslash
1534 uses forward error correction (FEC) for the unreliable UDP and
1535 DCCP transports. FEC is a technique which was invented already in
1536 1960 by Reed and Solomon and which is widely used for the parity
1537 calculations of storage devices (RAID arrays). It is based on the
1538 algebraic concept of finite fields, today called Galois fields, in
1539 honour of the mathematician Galois (1811-1832). The FEC implementation
1540 of paraslash is based on code by Luigi Rizzo.
1541
1542 Although the details require a sound knowledge of the underlying
1543 mathematics, the basic idea is not hard to understand: For positive
1544 integers k and n with k < n it is possible to compute for any k given
1545 data bytes d_1, ..., d_k the corresponding r := n -k parity bytes p_1,
1546 ..., p_r such that all data bytes can be reconstructed from *any*
1547 k bytes of the set
1548
1549 {d_1, ..., d_k, p_1, ..., p_r}.
1550
1551 FEC-encoding for unreliable network transports boils down to slicing
1552 the audio stream into groups of k suitably sized pieces called _slices_
1553 and computing the r corresponding parity slices. This step is performed
1554 in para_server which then sends both the data and the parity slices
1555 over the unreliable network connection. If the client was able
1556 to receive at least k of the n = k + r slices, it can reconstruct
1557 (FEC-decode) the original audio stream.
1558
1559 From these observations it is clear that there are three different
1560 FEC parameters: The slice size, the number of data slices k, and the
1561 total number of slices n. It is crucial to choose the slice size
1562 such that no fragmentation of network packets takes place because
1563 FEC only guards against losses and reordering but fails if slices are
1564 received partially.
1565
1566 FEC decoding in paralash is performed through the fecdec filter which
1567 usually is the first filter (there can be other filters before fecdec
1568 if these do not alter the audio stream).
1569
1570 Volume adjustment (amp and compress)
1571 ------------------------------------
1572
1573 The amp and the compress filter both adjust the volume of the audio
1574 stream. These filters operate on uncompressed audio samples. Hence
1575 they are usually placed directly after the decoding filter. Each
1576 sample is multiplied with a scaling factor (>= 1) which makes amp
1577 and compress quite expensive in terms of computing power.
1578
1579 ### amp ###
1580
1581 The amp filter amplifies the audio stream by a fixed scaling factor
1582 that must be known in advance. For para_audiod this factor is derived
1583 from the amplification field of the audio file's entry in the audio
1584 file table while para_filter uses the value given at the command line.
1585
1586 The optimal scaling factor F for an audio file is the largest real
1587 number F >= 1 such that after multiplication with F all samples still
1588 fit into the sample interval [-32768, 32767]. One can use para_filter
1589 in combination with the sox utility to compute F:
1590
1591 para_filter -f mp3dec -f wav < file.mp3 | sox -t wav - -e stat -v
1592
1593 The amplification value V which is stored in the audio file table,
1594 however, is an integer between 0 and 255 which is connected to F
1595 through the formula
1596
1597 V = (F - 1) * 64.
1598
1599 To store V in the audio file table, the command
1600
1601 para_client -- touch -a=V file.mp3
1602
1603 is used. The reader is encouraged to write a script that performs
1604 these computations :)
1605
1606 ### compress ###
1607
1608 Unlike the amplification filter, the compress filter adjusts the volume
1609 of the audio stream dynamically without prior knowledge about the peak
1610 value. It maintains the maximal volume of the last n samples of the
1611 audio stream and computes a suitable amplification factor based on that
1612 value and the various configuration options. It tries to chose this
1613 factor such that the adjusted volume meets the desired target level.
1614
1615 Note that it makes sense to combine amp and compress.
1616
1617 Misc filters (wav and prebuffer)
1618 --------------------------------
1619
1620 These filters are rather simple and do not modify the audio stream at
1621 all. The wav filter is only useful with para_filter and in connection
1622 with a decoder. It asks the decoder for the number of channels and the
1623 sample rate of the stream and adds a Microsoft wave header containing
1624 this information at the beginning. This allows to write wav files
1625 rather than raw PCM files (which do not contain any information about
1626 the number of channels and the sample rate).
1627
1628 The prebuffer filter simply delays the output until the given time has
1629 passed (starting from the time the first byte was available in its
1630 input queue) or until the given amount of data has accumulated. It
1631 is mainly useful for para_audiod if the standard parameters result
1632 in buffer underruns.
1633
1634 Both filters require almost no additional computing time, even when
1635 operating on uncompressed audio streams, since data buffers are simply
1636 "pushed down" rather than copied.
1637
1638 Examples
1639 --------
1640
1641 -> Decode an mp3 file to wav format:
1642
1643 para_filter -f mp3dec -f wav < file.mp3 > file.wav
1644
1645 -> Amplify a raw audio file by a factor of 1.5:
1646
1647 para_filter -f amp --amp 32 < foo.raw > bar.raw
1648
1649 ======
1650 Output
1651 ======
1652
1653 Once an audio stream has been received and decoded to PCM format,
1654 it can be sent to a sound device for playback. This part is performed
1655 by paraslash _writers_ which are described in this chapter.
1656
1657 Writers
1658 -------
1659
1660 A paraslash writer acts as a data sink that consumes but does not
1661 produce audio data. Paraslash writers operate on the client side and
1662 are contained in para_audiod and in the stand-alone tool para_write.
1663
1664 The para_write program reads uncompressed audio data from STDIN. If
1665 this data starts with a wav header, sample rate, sample format and
1666 channel count are read from the header. Otherwise CD audio (44.1KHz
1667 16 bit little endian, stereo) is assumed but this can be overridden
1668 by command line options. para_audiod, on the other hand, obtains
1669 the sample rate and the number of channels from the decoder.
1670
1671 Like receivers and filters, each writer has an individual set of
1672 command line options, and for para_audiod writers can be configured
1673 per audio format separately. It is possible to activate more than
1674 one writer for the same stream simultaneously.
1675
1676 OS-dependent APIs
1677 -----------------
1678
1679 Unfortunately, the various flavours of Unix on which paraslash
1680 runs on have different APIs for opening a sound device and starting
1681 playback. Hence for each such API there is a paraslash writer that
1682 can play the audio stream via this API.
1683
1684 - *ALSA*. The _Advanced Linux Sound Architecture_ is only available on
1685 Linux systems. Although there are several mid-layer APIs in use by
1686 the various Linux distributions (ESD, Jack, PulseAudio), paraslash
1687 currently supports only the low-level ALSA API which is not supposed
1688 to be change. ALSA is very feature-rich, in particular it supports
1689 software mixing via its DMIX plugin. ALSA is the default writer on
1690 Linux systems.
1691
1692 - *OSS*. The _Open Sound System_ is the only API on \*BSD Unixes and
1693 is also available on Linux systems, usually provided by ALSA as an
1694 emulation for backwards compatibility. This API is rather simple but
1695 also limited. For example only one application can open the device
1696 at any time. The OSS writer is activated by default on BSD Systems.
1697
1698 - *FILE*. The file writer allows to capture the audio stream and
1699 write the PCM data to a file on the file system rather than playing
1700 it through a sound device. It is supported on all platforms and is
1701 always compiled in.
1702
1703 - *AO*. _Libao_ is a cross-platform audio library which supports a wide
1704 variety of platforms including PulseAudio (gnome), ESD (Enlightened
1705 Sound Daemon), AIX, Solaris and IRIX. The ao writer plays audio
1706 through an output plugin of libao.
1707
1708 Examples
1709 --------
1710
1711 -> Use the OSS writer to play a wav file:
1712
1713 para_write --writer oss < file.wav
1714
1715 -> Enable ALSA software mixing for mp3 streams:
1716
1717 para_audiod --writer 'mp3:alsa -d plug:swmix'
1718
1719
1720 ===
1721 Gui
1722 ===
1723
1724 para_gui executes an arbitrary command which is supposed to print
1725 status information to STDOUT. It then displays this information in
1726 a curses window. By default the command
1727
1728 para_audioc -- stat -p
1729
1730 is executed, but this can be customized via the --stat-cmd option. In
1731 particular it possible to use
1732
1733 para_client -- stat -p
1734
1735 to make para_gui work on systems on which para_audiod is not running.
1736
1737 Key bindings
1738 ------------
1739
1740 It is possible to bind keys to arbitrary commands via custom
1741 key-bindings. Besides the internal keys which can not be changed (help,
1742 quit, loglevel, version...), the following flavours of key-bindings
1743 are supported:
1744
1745 - external: Shutdown curses before launching the given command.
1746 Useful for starting other ncurses programs from within para_gui,
1747 e.g. aumix or dialog scripts. Or, use the mbox output format to write
1748 a mailbox containing one mail for each (admissible) file the audio
1749 file selector knows about. Then start mutt from within para_gui to
1750 browse your collection!
1751
1752 - display: Launch the command and display its stdout in para_gui's
1753 bottom window.
1754
1755 - para: Like display, but start "para_client <specified command>"
1756 instead of "<specified command>".
1757
1758 The general form of a key binding is
1759
1760 key_map k:m:c
1761
1762 which maps key k to command c using mode m. Mode may be x, d or p
1763 for external, display and paraslash commands, respectively.
1764
1765 Themes
1766 ------
1767
1768 Currently there are only two themes for para_gui. It is easy, however,
1769 to add more themes. To create a new theme one has to define the
1770 position, color and geometry for for each status item that should be
1771 shown by this theme. See gui_theme.c for examples.
1772
1773 The "." and "," keys are used to switch between themes.
1774
1775 Examples
1776 --------
1777
1778 -> Show server info:
1779
1780 key_map "i:p:si"
1781
1782 -> Jump to the middle of the current audio file by pressing F5:
1783
1784 key_map "<F5>:p:jmp 50"
1785
1786 -> vi-like bindings for jumping around:
1787
1788 key_map "l:p:ff 10"
1789 key_map "h:p:ff 10-"
1790 key_map "w:p:ff 60"
1791 key_map "b:p:ff 60-"
1792
1793 -> Print the current date and time:
1794
1795 key_map "D:d:date"
1796
1797 -> Call other curses programs:
1798
1799 key_map "U:x:aumix"
1800 key_map "!:x:/bin/bash"
1801 key_map "^E:x:/bin/sh -c 'vi ~/.paraslash/gui.conf'"
1802
1803 ===========
1804 Development
1805 ===========
1806
1807 Contributing
1808 ------------
1809
1810 Paraslash is an open source project and contributions are
1811 welcome. Here's a list of things you can do to help the project:
1812
1813 - Report problems with building, installing or running the software.
1814 In particular, test the experimental git branches ("next" and "pu").
1815 This helps to identify and fix problems before the code gets merged
1816 and thus keeps the master branch as stable as possible.
1817 - Proofread the documentation (manual, web pages, man pages, source
1818 code documentation) and point out unclear or poorly written parts. If
1819 you are a native English speaker you will easily find a lot of text
1820 that could be improved.
1821 - Run analysis tools (coverity, afl, sparse, etc.) and report issues
1822 found by those tools.
1823 - Suggest new features you would like to see implemented.
1824 - Compile and test on your favorite architecture or operating
1825 system. The code is tested only on a limited set of systems, so you
1826 will probably encounter problems when building on different systems.
1827 - Post about about paraslash on your blog or on social networks.
1828 - Build and maintain Debian/RPM packages for your favorite distribution.
1829
1830 Note that there is no mailing list, no bug tracker and no discussion
1831 forum for paraslash. If you'd like to contribute, or have questions
1832 about contributing, send email to Andre Noll <maan@tuebingen.mpg.de>.
1833
1834 Tools
1835 -----
1836
1837 In order to compile the sources from the git repository (rather than
1838 from tar balls) and for contributing non-trivial changes to the
1839 paraslash project, some additional tools should be installed on a
1840 developer machine.
1841
1842 - [git](http://git.or.cz/). As described in more detail
1843 [below](#Git.branches), the git source code management tool is used for
1844 paraslash development. It is necessary for cloning the git repository
1845 and for getting updates.
1846
1847 - [autoconf](ftp://ftp.gnu.org/pub/gnu/autoconf/) GNU autoconf creates
1848 the configure file which is shipped in the tarballs but has to be
1849 generated when compiling from git.
1850
1851 - [discount](http://www.pell.portland.or.us/~orc/Code/discount). The
1852 HTML version of this manual and some of the paraslash web pages are
1853 written in the Markdown markup language and are translated into html
1854 with the converter of the *Discount* package.
1855
1856 - [doxygen](http://www.stack.nl/~dimitri/doxygen/). The documentation
1857 of paraslash's C sources uses the doxygen documentation system. The
1858 conventions for documenting the source code is described in the
1859 [Doxygen section](#Doxygen).
1860
1861 - [global](ftp://ftp.gnu.org/pub/gnu/global). This is used to generate
1862 browsable HTML from the C sources. It is needed by doxygen.
1863
1864 Git branches
1865 ------------
1866
1867 Paraslash has been developed using the git source code management
1868 tool since 2006. Development is organized roughly in the same spirit
1869 as the git development itself, as described below.
1870
1871 The following text passage is based on "A note from the maintainer",
1872 written by Junio C Hamano, the maintainer of git.
1873
1874 There are four branches in the paraslash repository that track the
1875 source tree: "master", "maint", "next", and "pu".
1876
1877 The "master" branch is meant to contain what is well tested and
1878 ready to be used in a production setting. There could occasionally be
1879 minor breakages or brown paper bag bugs but they are not expected to
1880 be anything major, and more importantly quickly and easily fixable.
1881 Every now and then, a "feature release" is cut from the tip of this
1882 branch, named with three dotted decimal digits, like 0.4.2.
1883
1884 Whenever changes are about to be included that will eventually lead to
1885 a new major release (e.g. 0.5.0), a "maint" branch is forked off from
1886 "master" at that point. Obvious, safe and urgent fixes after the major
1887 release are applied to this branch and maintenance releases are cut
1888 from it. New features never go to this branch. This branch is also
1889 merged into "master" to propagate the fixes forward.
1890
1891 A trivial and safe enhancement goes directly on top of "master".
1892 New development does not usually happen on "master", however.
1893 Instead, a separate topic branch is forked from the tip of "master",
1894 and it first is tested in isolation; Usually there are a handful such
1895 topic branches that are running ahead of "master". The tip of these
1896 branches is not published in the public repository to keep the number
1897 of branches that downstream developers need to worry about low.
1898
1899 The quality of topic branches varies widely. Some of them start out as
1900 "good idea but obviously is broken in some areas" and then with some
1901 more work become "more or less done and can now be tested by wider
1902 audience". Luckily, most of them start out in the latter, better shape.
1903
1904 The "next" branch is to merge and test topic branches in the latter
1905 category. In general, this branch always contains the tip of "master".
1906 It might not be quite rock-solid production ready, but is expected to
1907 work more or less without major breakage. The maintainer usually uses
1908 the "next" version of paraslash for his own pleasure, so it cannot
1909 be _that_ broken. The "next" branch is where new and exciting things
1910 take place.
1911
1912 The two branches "master" and "maint" are never rewound, and "next"
1913 usually will not be either (this automatically means the topics that
1914 have been merged into "next" are usually not rebased, and you can find
1915 the tip of topic branches you are interested in from the output of
1916 "git log next"). You should be able to safely build on top of them.
1917
1918 However, at times "next" will be rebuilt from the tip of "master" to
1919 get rid of merge commits that will never be in "master". The commit
1920 that replaces "next" will usually have the identical tree, but it
1921 will have different ancestry from the tip of "master".
1922
1923 The "pu" (proposed updates) branch bundles the remainder of the
1924 topic branches. The "pu" branch, and topic branches that are only in
1925 "pu", are subject to rebasing in general. By the above definition
1926 of how "next" works, you can tell that this branch will contain quite
1927 experimental and obviously broken stuff.
1928
1929 When a topic that was in "pu" proves to be in testable shape, it
1930 graduates to "next". This is done with
1931
1932 git checkout next
1933 git merge that-topic-branch
1934
1935 Sometimes, an idea that looked promising turns out to be not so good
1936 and the topic can be dropped from "pu" in such a case.
1937
1938 A topic that is in "next" is expected to be polished to perfection
1939 before it is merged to "master". Similar to the above, this is
1940 done with
1941
1942 git checkout master
1943 git merge that-topic-branch
1944 git branch -d that-topic-branch
1945
1946 Note that being in "next" is not a guarantee to appear in the next
1947 release (being in "master" is such a guarantee, unless it is later
1948 found seriously broken and reverted), nor even in any future release.
1949
1950 Coding Style
1951 ------------
1952
1953 The preferred coding style for paraslash coincides more or less
1954 with the style of the Linux kernel. So rather than repeating what is
1955 written [there](http://www.kernel.org/doc/Documentation/process/coding-style.rst),
1956 here are the most important points.
1957
1958 - Burn the GNU coding standards.
1959 - Never use spaces for indentation.
1960 - Tabs are 8 characters, and thus indentations are also 8 characters.
1961 - Don't put multiple assignments on a single line.
1962 - Avoid tricky expressions.
1963 - Don't leave whitespace at the end of lines.
1964 - The limit on the length of lines is 80 columns.
1965 - Use K&R style for placing braces and spaces:
1966
1967 if (x is true) {
1968 we do y
1969 }
1970
1971 - Use a space after (most) keywords.
1972 - Do not add spaces around (inside) parenthesized expressions.
1973 - Use one space around (on each side of) most binary and ternary operators.
1974 - Do not use cute names like ThisVariableIsATemporaryCounter, call it tmp.
1975 - Mixed-case names are frowned upon.
1976 - Descriptive names for global variables are a must.
1977 - Avoid typedefs.
1978 - Functions should be short and sweet, and do just one thing.
1979 - The number of local variables shouldn't exceed 10.
1980 - Gotos are fine if they improve readability and reduce nesting.
1981 - Don't use C99-style "// ..." comments.
1982 - Names of macros defining constants and labels in enums are capitalized.
1983 - Enums are preferred when defining several related constants.
1984 - Always use the paraslash wrappers for allocating memory.
1985 - If the name of a function is an action or an imperative.
1986 command, the function should return an error-code integer
1987 (<0 means error, >=0 means success). If the name is a
1988 predicate, the function should return a "succeeded" boolean.
1989
1990 Doxygen
1991 -------
1992
1993 Doxygen is a documentation system for various programming
1994 languages. The API reference on the paraslash web page is generated
1995 by doxygen.
1996
1997 It is more illustrative to look at the source code for examples than
1998 to describe the conventions in this manual, so we only describe which
1999 parts of the code need doxygen comments, but leave out details on
2000 documentation conventions.
2001
2002 As a rule, only the public part of the C source is documented with
2003 Doxygen. This includes structures, defines and enumerations in header
2004 files as well as public (non-static) C functions. These should be
2005 documented completely. For example, each parameter and the return
2006 value of a public function should get a descriptive doxygen comment.
2007
2008 No doxygen comments are necessary for static functions and for
2009 structures and enumerations in C files (which are used only within
2010 this file). This does not mean, however, that those entities need
2011 no documentation at all. Instead, common sense should be applied to
2012 document what is not obvious from reading the code.
2013
2014 ========
2015 Appendix
2016 ========
2017
2018 Network protocols
2019 -----------------
2020
2021 ### IP ###
2022
2023 The _Internet Protocol_ is the primary networking protocol used for
2024 the Internet. All protocols described below use IP as the underlying
2025 layer. Both the prevalent IPv4 and the next-generation IPv6 variant
2026 are being deployed actively worldwide.
2027
2028 ### Connection-oriented and connectionless protocols ###
2029
2030 Connectionless protocols differ from connection-oriented ones in
2031 that state associated with the sending/receiving endpoints is treated
2032 implicitly. Connectionless protocols maintain no internal knowledge
2033 about the state of the connection. Hence they are not capable of
2034 reacting to state changes, such as sudden loss or congestion on the
2035 connection medium. Connection-oriented protocols, in contrast, make
2036 this knowledge explicit. The connection is established only after
2037 a bidirectional handshake which requires both endpoints to agree
2038 on the state of the connection, and may also involve negotiating
2039 specific parameters for the particular connection. Maintaining an
2040 up-to-date internal state of the connection also in general means
2041 that the sending endpoints perform congestion control, adapting to
2042 qualitative changes of the connection medium.
2043
2044 ### Reliability ###
2045
2046 In IP networking, packets can be lost, duplicated, or delivered
2047 out of order, and different network protocols handle these
2048 problems in different ways. We call a transport-layer protocol
2049 _reliable_, if it turns the unreliable IP delivery into an ordered,
2050 duplicate- and loss-free delivery of packets. Sequence numbers
2051 are used to discard duplicates and re-arrange packets delivered
2052 out-of-order. Retransmission is used to guarantee loss-free
2053 delivery. Unreliable protocols, in contrast, do not guarantee ordering
2054 or data integrity.
2055
2056 ### Classification ###
2057
2058 With these definitions the protocols which are used by paraslash for
2059 steaming audio data may be classified as follows.
2060
2061 - HTTP/TCP: connection-oriented, reliable,
2062 - UDP: connectionless, unreliable,
2063 - DCCP: connection-oriented, unreliable.
2064
2065 Below we give a short descriptions of these protocols.
2066
2067 ### TCP ###
2068
2069 The _Transmission Control Protocol_ provides reliable, ordered delivery
2070 of a stream and a classic window-based congestion control. In contrast
2071 to UDP and DCCP (see below), TCP does not have record-oriented or
2072 datagram-based syntax, i.e. it provides a stream which is unaware
2073 and independent of any record (packet) boundaries. TCP is used
2074 extensively by many application layers. Besides HTTP (the Hypertext
2075 Transfer Protocol), also FTP (the File Transfer protocol), SMTP (Simple
2076 Mail Transfer Protocol), SSH (Secure Shell) all sit on top of TCP.
2077
2078 ### UDP ###
2079
2080 The _User Datagram Protocol_ is the simplest transport-layer protocol,
2081 built as a thin layer directly on top of IP. For this reason, it offers
2082 the same best-effort service as IP itself, i.e. there is no detection
2083 of duplicate or reordered packets. Being a connectionless protocol,
2084 only minimal internal state about the connection is maintained, which
2085 means that there is no protection against packet loss or network
2086 congestion. Error checking and correction (if at all) are performed
2087 in the application.
2088
2089 ### DCCP ###
2090
2091 The _Datagram Congestion Control Protocol_ combines the
2092 connection-oriented state maintenance known from TCP with the
2093 unreliable, datagram-based transport of UDP. This means that it
2094 is capable of reacting to changes in the connection by performing
2095 congestion control, offering multiple alternative approaches. But it
2096 is bound to datagram boundaries (the maximum packet size supported
2097 by a medium), and like UDP it lacks retransmission to protect
2098 against loss. Due to the use of sequence numbers, it is however
2099 able to react to loss (interpreted as a congestion indication) and
2100 to ignore out-of-order and duplicate packets. Unlike TCP it allows
2101 to negotiate specific, binding features for a connection, such as
2102 the choice of congestion control: classic, window-based congestion
2103 control known from TCP is available as CCID-2, rate-based, "smooth"
2104 congestion control is offered as CCID-3.
2105
2106 ### HTTP ###
2107
2108 The _Hypertext Transfer Protocol_ is an application layer protocol
2109 on top of TCP. It is spoken by web servers and is most often used
2110 for web services. However, as can be seen by the many Internet radio
2111 stations and YouTube/Flash videos, http is by far not limited to the
2112 delivery of web pages only. Being a simple request/response based
2113 protocol, the semantics of the protocol also allow the delivery of
2114 multimedia content, such as audio over http.
2115
2116 ### Multicast ###
2117
2118 IP multicast is not really a protocol but a technique for one-to-many
2119 communication over an IP network. The challenge is to deliver
2120 information to a group of destinations simultaneously using the
2121 most efficient strategy to send the messages over each link of the
2122 network only once. This has benefits for streaming multimedia: the
2123 standard one-to-one unicast offered by TCP/DCCP means that n clients
2124 listening to the same stream also consume n-times the resources,
2125 whereas multicast requires to send the stream just once, irrespective
2126 of the number of receivers. Since it would be costly to maintain state
2127 for each listening receiver, multicast often implies connectionless
2128 transport, which is the reason that it is currently only available
2129 via UDP.
2130
2131 Abstract socket namespace
2132 -------------------------
2133 UNIX domain sockets are a traditional way to communicate between
2134 processes on the same machine. They are always reliable (see above)
2135 and don't reorder datagrams. Unlike TCP and UDP, UNIX domain sockets
2136 support passing open file descriptors or process credentials to
2137 other processes.
2138
2139 The usual way to set up a UNIX domain socket (as obtained from
2140 socket(2)) for listening is to first bind the socket to a file system
2141 pathname and then call listen(2), then accept(2). Such sockets are
2142 called _pathname sockets_ because bind(2) creates a special socket
2143 file at the specified path. Pathname sockets allow unrelated processes
2144 to communicate with the listening process by binding to the same path
2145 and calling connect(2).
2146
2147 There are two problems with pathname sockets:
2148
2149 * The listing process must be able to (safely) create the
2150 socket special in a directory which is also accessible to
2151 the connecting process.
2152
2153 * After an unclean shutdown of the listening process, a stale
2154 socket special may reside on the file system.
2155
2156 The abstract socket namespace is a non-portable Linux feature which
2157 avoids these problems. Abstract sockets are still bound to a name,
2158 but the name has no connection with file system pathnames.
2159
2160 License
2161 -------
2162
2163 Paraslash is licensed under the GPL, version 2. Most of the code
2164 base has been written from scratch, and those parts are GPL V2
2165 throughout. Notable exceptions are FEC and the WMA decoder. See the
2166 corresponding source files for licencing details for these parts. Some
2167 code sniplets of several other third party software packages have
2168 been incorporated into the paraslash sources, for example log message
2169 coloring was taken from the git sources. These third party software
2170 packages are all published under the GPL or some other license
2171 compatible to the GPL.
2172
2173 Acknowledgements
2174 ----------------
2175
2176 Many thanks to Gerrit Renker who read an early draft of this manual
2177 and contributed significant improvements.
2178
2179 ==========
2180 References
2181 ==========
2182
2183 Articles
2184 --------
2185 - [Polynomial Codes over Certain Finite
2186 Fields](http://kom.aau.dk/~heb/kurser/NOTER/KOFA01.PDF) by Reed, Irving
2187 S.; Solomon, Gustave (1960), Journal of the Society for Industrial
2188 and Applied Mathematics (SIAM) 8 (2): 300-304, doi:10.1137/0108018)
2189
2190 RFCs
2191 ----
2192
2193 - [RFC 768](http://www.ietf.org/rfc/rfc768.txt) (1980): User Datagram
2194 Protocol
2195
2196 - [RFC 791](http://www.ietf.org/rfc/rfc791.txt) (1981): Internet
2197 Protocol
2198
2199 - [RFC 2437](http://www.ietf.org/rfc/rfc2437.txt) (1998): RSA
2200 Cryptography Specifications
2201
2202 - [RFC 4340](http://www.ietf.org/rfc/rfc4340.txt) (2006): Datagram
2203 Congestion Control Protocol (DCCP)
2204
2205 - [RFC 4341](http://www.ietf.org/rfc/rfc4341.txt) (2006): Congestion
2206 Control ID 2: TCP-like Congestion Control
2207
2208 - [RFC 4342](http://www.ietf.org/rfc/rfc4342.txt) (2006): Congestion
2209 Control ID 3: TCP-Friendly Rate Control (TFRC)
2210
2211 - [RFC 6716](http://www.ietf.org/rfc/rfc6716.txt) (2012): Definition
2212 of the Opus Audio Codec
2213
2214 Application web pages
2215 ---------------------
2216
2217 - [paraslash](http://people.tuebingen.mpg.de/maan/paraslash/)
2218 - [xmms](http://xmms2.org/wiki/Main_Page)
2219 - [mpg123](http://www.mpg123.de/)
2220 - [gstreamer](http://gstreamer.freedesktop.org/)
2221 - [icecast](http://www.icecast.org/)
2222 - [Audio Compress](http://beesbuzz.biz/code/audiocompress.php)
2223
2224 External documentation
2225 ----------------------
2226
2227 - [The mathematics of
2228 Raid6](http://kernel.org/pub/linux/kernel/people/hpa/raid6.pdf)
2229 by H. Peter Anvin
2230
2231 - [Effective Erasure Codes for reliable Computer Communication
2232 Protocols](http://info.iet.unipi.it/~luigi/fec_ccr.ps.gz) by Luigi
2233 Rizzo
2234
2235 Code
2236 ----
2237 - [Original FEC
2238 implementation](http://info.iet.unipi.it/~luigi/vdm.tar.gz) by
2239 Luigi Rizzo)