VideoLAN Wiki - User contributions [en]

VLMa/Projects/DVBlast integration

2009-11-17T21:46:49Z

Jpountz: creation

== Issue ==

VLMa has been architectured assuming that VLC would be the only streaming facility it would use and the telnet interface is the way it communicates with VLC.

With DVBlast coming as a better solution than VLC for broadcasting DVB channels, this assumption is not right anymore, and DVBlast has no telnet interface (and the README clearly states the DVBlast will keep as simple as possible which means it will probably never implement

The idea behind this project is to provide a facade for communicating from the master (VLMa) to the slaves (the streamers: VLC, DVBlast, etc.).

== Solution ==

There already is a little piece of code in charge of doing process management in the Git repository which can probably be reused.

=== Required features ===

* Monitoring of the server (CPU load, Incoming/Outgoing traffic, etc.)
* Monitoring of the streamer (restarting of the process when needed, CPU & memory used, etc.)
* Extensibility, it should be very easy to add the support of another streaming tool,
* Information retrieval (logs, uptime, version of the software installed)

=== Implementation ===

Python programming language (to reuse some code which has already been written).

REST/XML API:
* Applications only need XML and HTTP bindings (which are prensent in almost every serious programming language) to use the API,
* XML is human-readable (on the contrary to protobuf, etch or thrift).

=== Paths ===

I'm not very good at finding resources' names, feel free to modify them.

* GET '''/server/monitoring''' monitoring metrics of the whole server (CPU load, etc.)
* GET '''/streamers''' list of the streaming programs running (VLC, DVBlast)
* GET '''/streamers/<id>''' description of the streamer whose id is <id>
* GET '''/streamers/<id>/channels''' channels currently handled by streamer <id>
* PUT '''/streamers/<id>/channels/<channel>''' add/modify the channel whose id is <channel>
* DELETE '''/streamers/<id>/channels/<channel>''' remove the channel whose id is <channel>
* GET '''/streamers/<id>/monitoring''' monitoring metrics specific to streamer <id>
* GET '''/streamers/<id>/logs''' logs of streamer <id>
* GET '''/streamers/<id>/restart''' restart the streamer whose id is <id>

VLMa/Projects

2009-05-26T12:38:41Z

Jpountz:

This are ideas of projetcs if you want to start coding on VLMa.

You can contact [[User:Jpountz|Adrien Grand (jpountz)]] if you are interested in one of them.

== Automated deployment ==

VLMa aims at automating as many operations required to setup a farm of streaming servers as possible. A painful operation which is not covered is the start of VLC on every streaming server.

VLMa is provided with a wrapper script around VLC which provides some services that VLC's telnet interface does not. The aim of this project is to set up an automated deployment capability in VLMa by providing mechanisms to deploy this script and running it on hosts which have been identified as streaming servers.

== Remote diagnostic ==

It can be frustrating to notice that some channels are not broadcasted without knowing why.

Currently, to know wether a channel is broadcasted or not, VLMa either joins the multicast group (in case of multicast streaming) or connects the the builtin server in VLC (HTTP or RTSP) so it has no access to debug information that would be useful to trace problems VLC encountered which could explain the fact that the channel is not broacasted.

For other issues (mainly retrieving metrics concerning the running VLC instance), a wrapper around VLC has been written (in Python). This wrapper is able to:
* restart VLC,
* expose metrics such as Memory usage of VLC / CPU load / etc.,
* get the last lines of logs of VLC,
* get VLC uptime,
* get VLC version.

The idea would be to write a log listener for VLC in this wrapper that would be able to analyze errors reported by VLC and expose some methods related to this analysis. Then VLMa daemon could query the wrapper to try to guess why the channel is not broadcasted.

== DVBlast and Darwin streaming server handling ==

Being able to deal with non-VLC streaming servers would be a plus.
See http://www.videolan.org/projects/dvblast.html and http://developer.apple.com/opensource/server/streaming/index.html.

== See also ==

[[VLMa | VLMa overview]]

VLMa/Projects

2009-05-26T12:24:44Z

Jpountz:

This are ideas of projetcs if you want to start coding on VLMa.

You can contact [[User:Jpountz|Adrien Grand]] if you are interested in one of them.

== Automated deployment ==

VLMa aims at automating as many operations required to setup a farm of streaming servers as possible. A painful operation which is not covered is the start of VLC on every streaming server.

VLMa is provided with a wrapper script around VLC which provides some services that VLC's telnet interface does not. The aim of this project is to set up an automated deployment capability in VLMa by providing mechanisms to deploy this script and running it on hosts which have been identified as streaming servers.

== Remote diagnostic ==

It can be frustrating to notice that some channels are not broadcasted without knowing why.

Currently, to know wether a channel is broadcasted or not, VLMa either joins the multicast group (in case of multicast streaming) or connects the the builtin server in VLC (HTTP or RTSP) so it has no access to debug information that would be useful to trace problems VLC encountered which could explain the fact that the channel is not broacasted.

For other issues (mainly retrieving metrics concerning the running VLC instance), a wrapper around VLC has been written (in Python). This wrapper is able to:
* restart VLC,
* expose metrics such as Memory usage of VLC / CPU load / etc.,
* get the last lines of logs of VLC,
* get VLC uptime,
* get VLC version.

The idea would be to write a log listener for VLC in this wrapper that would be able to analyze errors reported by VLC and expose some methods related to this analysis. Then VLMa daemon could query the wrapper to try to guess why the channel is not broadcasted.

== DVBlast and Darwin streaming server handling ==

Being able to deal with non-VLC streaming servers would be a plus.
See http://www.videolan.org/projects/dvblast.html and http://developer.apple.com/opensource/server/streaming/index.html.

== See also ==

[[VLMa | VLMa overview]]

User:Jpountz

2009-03-24T14:01:15Z

Jpountz:

Adrien Grand, VLMa developer.

You can contact me by IRC:
* jpountz on #videolan@irc.freenode.net
or by mail
* jpountz (at) videolan (dot) org

VLMa/Projects

2009-03-24T13:56:33Z

Jpountz:

SoC 2009

2009-02-26T03:53:13Z

Jpountz: Typos.

Google Summer of Code 2009.

== Introduction & Information ==

This wiki page covers the attempt by the [[VideoLAN]] project to act as a mentoring organization for [http://code.google.com/soc/ Google Summer of Code], in order to improve {{VLC}} and [[x264]], the H.264 video encoder.

* This page list the ideas for VLC.
* There is also a separate page for project ideas for the [[x264]]: [[SoC x264]].

If you are a '''student''', please read the [[SoC_2008_Rules|rules]] and [[SoC_2008_Rules#Advices|advices]].

=== VLC media player ===
{{VLC}} is a cross-platform Multimedia player, and encoding and streaming application.
It is one of the most successful open-source projects without any professional structure underneath.

{{VLC}} is downloaded at an approximate monthly rate of 5 millions from the main website and that's not including third-party distributions (Linux)!

You can find more information on [http://en.wikipedia.org/wiki/VLC_media_player VLC on Wikipedia] or on this [[Main_Page|wiki]].

=== Summer of Code ===

If selected and developed, SoC ideas will be included in the VLC 1.1 or any later version release, and covered by the GPL licence.

=== Ideas ===

'''''Read this'''''

This page is split in three lists of ideas:
* The ''main ideas'' are the ones we would like to focus on, and seem to be key projects for VLC and a lot more ''thrilling'' than the other ones; we have assigned a potential mentor to each of these.
* The other ideas are less detailed but could be good ideas too.
* Original good ideas will be valued. Last year, half of the ideas were given by students. [[Mini_Projects]] has some ideas too.

We don't want to impose anything. This is free and Open Source software. Last year, many projects were external ideas.

If you don't want to apply for SoC but want other ideas to develop on, check the [[Mini Projects]] page!

And on IRC channel, you can have even more ideas.

===x264===
For x264 ideas, please see [[SoC_x264]].

== Main ideas ==

=== Test Suite ===
This project aims at making automatic tests to improve VLC quality.

* Write a series of tests for vlc-internal
* Integrate a framework for automated plugins testing with the automake build system
* Automate the different codec playback/mux/etc tests
* Automate the subtitles tests
* Write tests for the different bindings: Mac OS X Framework/Python Bindings... etc.

This project is a code project, require some knowledge of C

''Proposed mentor: ?''

=== QuickTimeVR Playback ===
This project deals with integration QuickTimeVR inside VLC using work from [http://freepv.sf.net FreePV] (SoC 2007).
Other panoramic data inputs include video surveillance cameras with a 360° lens, regular images, etc.

This project is likely to require some modification of the VLC image handling code.

''Proposed mentor: ?''

=== GUI interfaces for the browsers plug-ins ===
This project focuses on adding interfaces to our browser plug-ins, in order to have, in the web pages embedded mode, buttons to control {{VLC}} as in WMP or Youtube player. These buttons need to be cross-platform and cross-browsers.

The second part of this project aims to provide a replacement for all preconfigured media players embedded in web pages, if possible.

For example, when a webpage asks for WMP embedded plugin, youtube player, quicktime player or any other, VLC would detect the stream and display it. This would mean less CPU needed than Flash based players...

This project may need to work in concert with some improvements on the cache handling module access.

The look of the interface should not be a problem (it can be added later on).

''Proposed mentor: ?''

== More Ideas ==

=== QTKit access ===
This project means to deal with building a QTKit-based access module for OS X (allowing input from USB / FireWire cameras, audio devices, camcorders, tape recorders [, ...your favourite here] ).

This access module will also need a nice GUI Integration.

This project requires C, preferably Obj-C, knowledge and interest in video/audio capture devices. Obviously, the student is required to own at a least a Mac with a (embedded) webcam and an audio input facility for development. Knowledge of the QuickTime APIs and previous experience with Cocoa are strong pluses.

''Proposed mentor: ?''

=== Leopard integration ===
* a QuickLook-plugin for OS X Leopard
* Video Output animations
* VLC_app (VLCKit based) improvements.
* VLCKit oneclick-Streaming Server implementation.

This project requires C, preferably Obj-C, knowledge and interest in GUI programming. Previous experience with Cocoa preferred. Basic Core Animation knowledge is a plus, but far from being absolutely necessary.

''Proposed mentors: ?''

=== Automated threading error detection ===
VLC is heavily multithreaded.

This project would involve extending the VLC threading APIs to provide better run-time consistency checks, and possibly proof of correctness.

This could be similar to the lock consistency checking in the Linux kernel, albeit more simple - VLC does not have to deal with interruptions neither atomic operation and has fewer thread and synchronisation constructs.

This project requires working knowledge of the C programming language and POSIX threads.

''proposed mentor: ?''

=== Assembly optimizations in VLC ===

If you are fond of writing ASM and you are fluent in MMX/SSE2, there are many interesting things in VLC to speed up.
* memcpy implementations in SSE2
* video filters ASM speedups
* audio filters ASM speedups
* Porting other ASM video filters to VLC.
* Profiling VLC.

This project needs some good ASM knowledge and good C experience.

''Proposed mentor: ?''

=== Matroska (mkv) muxing ===

This project is about writing a mkv muxer for VLC with advanced support (chapters, multiple audio and multiple subtitles, etc...) like when you need to rip or archive a DVD.

FFMPEG project has a similar project, and we could use the libavformat muxer and adapt ourselves, but we are not sure if we will be able to use therefore all the advanced features.

''Proposed mentor: ?''

=== Skins improvements ===

Better video output integration.

Better skin support.

''Proposed mentor: ?''

=== SVCD handling ===

Make SVCD's work. Menu selection, navigation and hot spots.

''Proposed mentor: ?''

=== Lua audio filters ===

VLC currently uses Lua as an interface, to handle specific URLs, and to fetch media information. This project aims at creating an audio filter in Lua, and enabling users to create whatever audio filtering function they want in a Lua script.

''Proposed mentor: Geal''

== Mini Projects - More ideas ==

'''Original''' ideas will be '''rewarded.'''

We have a more general page for ideas for Project in {{VLC}}, that could give you ideas: [[Mini_Projects]].

* Visualisation improvements (libprojectM implementation ? )
* Video rework.

== Questions? Contact us? ==
If you need to contact us for any questions, you can:
* use this [[Talk:SoC_2009| Talk page]],
* use the [http://www.videolan.org/developers/lists.html vlc-devel] mailing list,
* come on IRC irc://irc.videolan.org #videolan
* contact one of the mentors or the backup mentors,
* or contact [[User:J-b| Jean-Baptiste Kempf]]

=== Mentors ===
The proposed mentors are:
* ?

=== Administrators ===
* jb
* xtophe

[[Category:Activities]] [[Category:SoC]]

VLMa/Projects

2009-02-26T03:44:51Z

Jpountz:

This are ideas of projetcs if you want to start coding on VLMa.

== Automated deployment ==

VLMa aims at automating as many operations required to setup a farm of streaming servers as possible. A painful operation which is not covered is the start of VLC on every streaming server.

VLMa is provided with a wrapper script around VLC which provides some services that VLC's telnet interface does not. The aim of this project is to set up an automated deployment capability in VLMa by providing mechanisms to deploy this script and running it on hosts which have been identified as streaming servers.

== Remote diagnostic ==

It can be frustrating to notice that some channels are not broadcasted without knowing why.

Currently, to know wether a channel is broadcasted or not, VLMa either joins the multicast group (in case of multicast streaming) or connects the the builtin server in VLC (HTTP or RTSP) so it has no access to debug information that would be useful to trace problems VLC encountered which could explain the fact that the channel is not broacasted.

For other issues (mainly retrieving metrics concerning the running VLC instance), a wrapper around VLC has been written (in Python). This wrapper is able to:
* restart VLC,
* expose metrics such as Memory usage of VLC / CPU load / etc.,
* get the last lines of logs of VLC,
* get VLC uptime,
* get VLC version.

The idea would be to write a log listener for VLC in this wrapper that would be able to analyze errors reported by VLC and expose some methods related to this analysis. Then VLMa daemon could query the wrapper to try to guess why the channel is not broadcasted.

== See also ==

[[VLMa | VLMa overview]]

VLMa/Projects

2009-02-26T03:43:15Z

Jpountz: New page: This are ideas of projetcs if you want to start coding on VLMa. == Automated deployment == VLMa aims at automating as many operations required to setup a farm of streaming servers as pos...

This are ideas of projetcs if you want to start coding on VLMa.

== Automated deployment ==

VLMa aims at automating as many operations required to setup a farm of streaming servers as possible. A painful operation which is not covered is the start of VLC on every streaming server.

VLMa is provided with a wrapper script around VLC which provides some services that VLC's telnet interface does not. The aim of this project is to set up an automated deployment capability in VLMa by providing mechanisms to deploy this script and running it on hosts which have been identified as streaming servers.

== Remote diagnostic ==

It can be frustrating to notice that some channels are not broadcasted without knowing why.

Currently, to know wether a channel is broadcasted or not, VLMa either joins the multicast group (in case of multicast streaming) or connects the the builtin server in VLC (HTTP or RTSP) so it has no access to debug information that would be useful to trace problems VLC encountered which could explain the fact that the channel is not broacasted.

For other issues (mainly retrieving metrics concerning the running VLC instance), a wrapper around VLC has been written (in Python). This wrapper is able to:
* restart VLC,
* expose metrics such as Memory usage of VLC / CPU load / etc.,
* get the last lines of logs of VLC,
* get VLC uptime,
* get VLC version.

The idea would be to write a log listener for VLC that would be able to analyze errors reported by VLC and expose some methods related to this analysis. Then VLMa daemon could query the wrapper to try to guess why the channel is not broadcasted.

== See also ==

[[VLMa | VLMa overview]]

VLMa

2009-02-26T03:15:25Z

Jpountz:

VLMa/Documentation:Installation

2009-02-14T21:42:17Z

Jpountz:

== Requirements ==

* A JDK version 5.0 or higher
* VLMa binary package

== How to ==

First extract VLMa distribution archive to the folder of your choice, you should see the following folders:
* bin/ Command line wrappers to manage VLMa
* conf/ Configuration
* data/ Data
* lib/ Dependencies
* logs/ Log directory
* webapps/ Webapps to deploy
* work/ Folder needed by the servlet container

=== UNIX specific ===

Ensure that the JAVA_HOME environment variable has been set. Run the following
command:
echo $JAVA_HOME

If it does not print anything to the standard input but whitespaces, this means
that JAVA_HOME has not been set. You can set JAVA_HOME using the following
command:
export JAVA_HOME=/path/to/your/JDK/installation

Then open a terminal and go to the folder where you extracted VLMa. You can run
VLMa daemon by running the following command:
bin/vlmad start

To run the web UI, open a new terminal and run:
bin/vlmaw start

=== Windows specific ===

Ensure that the JAVA_HOME environment variable has been set and that it points
to a valid JDK installation. To do this, have a look at:
* http://support.microsoft.com/kb/310519 for Windows XP,
* http://support.microsoft.com/kb/931715 for Windows Vista.

Then open a prompt and change directory to where you extracted VLMa. You can
run VLMa daemon by running the following command:
bin\vlmad start

To run the web UI, open a new prompt and run:
bin\vlmaw start

== The end ==

To finish with, go to http://localhost:8080, you should see VLMa dashboard
which tells you which medias are currently streamed and the list of servers
VLMa didn't manage to contact through their telnet interface.

==See also==

* [[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation:Installation

2009-02-14T21:39:16Z

Jpountz: New page: == Requirements == * A JDK version 5.0 or higher * VLMa binary package == How to == First extract VLMa distribution archive to the folder of your choice, you should see the following fo...

VLMa/Documentation:Assignment algorithm

2009-02-14T21:33:49Z

Jpountz: New page: = Orders assignment algorithm = == Acknolegements == * Every adapter has a score. This score is 0 when VLMa starts and then gets incremented by 5 when the order handled by this adapter i...

= Orders assignment algorithm =

== Acknolegements ==

* Every adapter has a score. This score is 0 when VLMa starts and then gets incremented by 5 when the order handled by this adapter is broadcasted and decremented by 1 otherwise.

* The user can assign priorities to programs, so that a program with a higer priority is more likely to be broadcasted.

== Steps of the assignment ==

=== Grouping ===

Programs that can be handled by the same adapter (two DTT channels having the same frequency for example) are put together into what we call "Program groups".

=== Partitionment ===

Not every adapter can read a given program group. So the first step si to put together adapters and program groups by affinity, DVB-T adapters with DTT program groups, stream channels with stream program groups, or DBV-S adapters bound to satellite XXX with Satellite program groups whose media is bound to XXX.

=== Assignment ===

For each partition, adapters are then sorted by score and program groups by priority (the priority of a group being the sum of the priorities of its programs).

Then the first program group is assigned to the first adapter, the 2nd program group with the 2nd adapter, etc.

==See also==

* [[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation

2009-02-14T20:48:29Z

Jpountz:

= VLMa documentation =

== Developer documentation ==

* [[VLMa/Documentation:Build | Build HOWTO]]
* [[VLMa/Documentation:Architecture | VLMa architecture]]
* [[VLMa/Documentation:Exclipse integration | Using Eclipse to develop VLMa]]
* [[VLMa/Documentation:Assignment_algorithm | Algorithm used for the assignment of orders]]

== User documentation ==

* [[VLMa/Documentation:Installation | Installing VLMa on your own computer]]
* [[VLMa/Documentation:Configuration_reference | VLMa configuration variables]]

==See also==

* [[VLMa | VLMa overview]]

VLMa/Documentation:Architecture

2009-02-14T20:47:09Z

Jpountz:

=Architecture=

==The model==

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

==The threads==

The daemon is made of two main threads:
* a thread which monitors the state of the servers (up/down),
* a thread which monitors the state or programmation (whether programs are really streamed or not) and asks for orders to be given again whether there is something wrong.

Most events are triggered by these two threads, the other events being triggered from user interaction.

==Orders==

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

==A programmation state==

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==See also==

* [[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation

2009-02-03T16:52:29Z

Jpountz:

VLMa/Documentation:Exclipse integration

2009-02-03T16:51:14Z

Jpountz:

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project with Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

If you plan to send patches, please ensure to set a 4-spaces indentation (no tabs!).

== See also ==

* [[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation:Build

2009-02-03T16:50:56Z

Jpountz:

= VLMa build HOWTO =

== Requirements ==

You need the following softwares to build VLMa:

* A Java SDK version 5.0 or higher

You can check whether your Java installation is suitable by checking that the ''javac'' executable is available and that the version is higher than 5.0 (''javac -version'').

* [http://maven.apache.org Maven 2]

VLMa uses Maven as a build system. If you are not familiar with this tool, check the [http://maven.apache.org/guides/getting-started/maven-in-five-minutes.html 5 minutes guide].

* VLMa source tree

You can either download the latest tarballs at http://www.videolan.org/vlma/download.html and extract them or check out the development tree using [[Git]]

$ git clone git://git.videolan.org/vlma.git

== Build ==

Once you have downloaded VLMa source code, change directory to the root of the project and type:

$ mvn install

Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

== See also ==

* [[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation:Architecture

2009-02-03T16:50:07Z

Jpountz:

=Architecture=

==The model==

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

==The threads==

The daemon is made of two main threads:
* a thread which monitors the state of the servers (up/down),
* a thread which monitors the state or programmation (whether programs are really streamed or not) and asks for orders to be given again whether there is something wrong.

Most events are triggered by these two threads, the other events being triggered from the user interaction.

==Orders==

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

==A programmation state==

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==See also==

* [[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation:Architecture

2009-02-03T16:49:51Z

Jpountz: New page: =Architecture= ==The model== The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation te...

=Architecture=

==The model==

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

==The threads==

The daemon is made of two main threads:
* a thread which monitors the state of the servers (up/down),
* a thread which monitors the state or programmation (whether programs are really streamed or not) and asks for orders to be given again whether there is something wrong.

Most events are triggered by these two threads, the other events being triggered from the user interaction.

==Orders==

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

==A programmation state==

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==See also==

[[VLMa/Documentation | VLMa documentation index]]

VLMa

2009-02-03T16:42:08Z

Jpountz:

VLMa/Documentation:Exclipse integration

2009-02-03T16:37:41Z

Jpountz: New page: ==Integration with Eclipse== Like any Maven project, there are two ways to integrate the VLMa project with Eclipse: * using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Mave...

VLMa/Documentation

2009-02-03T16:35:14Z

Jpountz:

VLMa/Documentation:Build

2009-02-03T16:32:57Z

Jpountz:

= VLMa build HOWTO =

== Requirements ==

You need the following softwares to build VLMa:

* A Java SDK version 5.0 or higher

You can check whether your Java installation is suitable by checking that the ''javac'' executable is available and that the version is higher than 5.0 (''javac -version'').

* [http://maven.apache.org Maven 2]

VLMa uses Maven as a build system. If you are not familiar with this tool, check the [http://maven.apache.org/guides/getting-started/maven-in-five-minutes.html 5 minutes guide].

* VLMa source tree

You can either download the latest tarballs at http://www.videolan.org/vlma/download.html and extract them or check out the development tree using [[Git]]

$ git clone git://git.videolan.org/vlma.git

== Build ==

Once you have downloaded VLMa source code, change directory to the root of the project and type:

$ mvn install

Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

== Internal links ==

[[VLMa/Documentation | VLMa documentation index]]

VLMa/Documentation:Build

2009-02-03T16:27:46Z

Jpountz: New page: = VLMa build HOWTO = == Requirements == You need the following softwares to build VLMa: * A Java SDK version 5.0 or higher You can check whether your Java installation is suitable by c...

VLMa/Documentation

2009-02-03T16:13:25Z

Jpountz: New page: = VLMa documentation = == Developer documentation == * Build HOWTO * VLMa architecture == User documentation == ...

= VLMa documentation =

== Developer documentation ==

* [[VLMa/Documentation:Build | Build HOWTO]]
* [[VLMa/Documentation:Architecture | VLMa architecture]]

== User documentation ==

* [[VLMa/Documentation:Installation | Installing VLMa on your own computer]]

VLMa

2009-02-03T16:01:06Z

Jpountz:

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLAN's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Usage==

VLMa is not the right tool if what you want is a graphical front-end to VLM. You should only consider using VLMa if you have several streamers to manage simultaneously.

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://trac.videolan.org/vlma VLMa Trac].
* [http://nightlies.videolan.org/vlma/ Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2009-01-01T00:02:14Z

Jpountz:

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Usage==

VLMa is not the right tool if what you want is a graphical front-end to VLM. You should only consider using VLMa if you have several streamers to manage simultaneously.

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://trac.videolan.org/vlma VLMa Trac].
* [http://nightlies.videolan.org/vlma/ Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-12-21T16:39:46Z

Jpountz:

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Usage==

VLMa is not the right tool if what you want is a graphical front-end to VLM. You should only consider using VLMa if you have several streamers to manage simultaneously.

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://trac.videolan.org/vlma VLMa Trac].
* [http://nightlies.videolan.org/vlma/ Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-09-14T20:42:41Z

Jpountz: oops

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://trac.videolan.org/vlma VLMa Trac].
* [http://nightlies.videolan.org/vlma/ Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-09-14T20:41:44Z

Jpountz: Update the nighly builds website URL

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://trac.videolan.org/vlma VLMa Trac].
* [http://nightlies.videolan.org/vlma Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-08-09T11:21:26Z

Jpountz: add a link to the trac

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://trac.videolan.org/vlma VLMa Trac].
* [http://nightlies.jpountz.net/vlma Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-07-16T21:35:08Z

Jpountz: URL correction

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://nightlies.jpountz.net/vlma Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-07-10T19:15:49Z

Jpountz: programmation state

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

===A programmation state===

A media can have 3 states:
* not programmed,
* programmed but not streamed,
* programmed and streamed.

To check whether programs are ''really'' streamed or not, there are stream watchers. Currently, there are two main implementations : the <code>DirectMulticastStreamWatcher</code> which joins the multicast group of the program (in case of an UDP multicast streaming strategy) and tries to receive data. If not enough data is received, the stream is known as not streamed. The other watcher, <code>HttpStreamWatcher</code> connects on the VLC http interface and verifies that no 404 error is encountered.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://nightlies.jpountz.net Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-07-10T17:52:59Z

Jpountz: precision

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread which exports the RMI registry and waits for method calls,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not, runs periodically too).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://nightlies.jpountz.net Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-07-06T00:23:56Z

Jpountz: typos

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in École Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge of managing the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://nightlies.jpountz.net Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

VLMa

2008-07-06T00:18:01Z

Jpountz: how does it work

[[image:VLMa.png|right]]

==What is VLMa?==

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in Ecole Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge to manage the streamers:
* give orders using the telnet interface,
* fallback whenever a server fails,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://en.wikipedia.org/wiki/Remote_Method_Invocation RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==How does it work?==

===The model===

The 2 key objects of the model are the media and the programmation. Basically, a media stands for something that can be streamed whereas the programmation tells you how it should be streamed.

===The threads===

There are 3 important threads:
* the main thread,
* a thread which monitors the state of the servers (runs periodically),
* a thread which monitors the state or programmation (whether programs are really streamed or not).

===Orders===

When VLMa notices that some programs are not really streamed, orders are computed and then sent to the servers. Computing orders means:
* identify available adapters,
* identify medias that should be streamed,
* put together medias into groups of medias that should be streamed by the same server (for example, because of their frequency in case of a [[DVB]] input),
* map every media group to an available adapter,
* send orders to the servers.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

==Related links==

* [http://nightlies.jpountz.net Nightly builds].
* [http://www.ohloh.net/projects/vlma Project details] at Ohloh.

[[Category:VideoLAN projects]]

JVLC

2008-06-27T11:51:56Z

Jpountz: Redirecting to Java bindings

#REDIRECT [[Java_bindings]]

VLMa

2008-06-26T07:57:41Z

Jpountz:

User:Jpountz

2008-06-21T17:21:58Z

Jpountz: typo

Adrien Grand

jpountz on #videolan@irc.freenode.net

VLMa developer.

VLMa

2008-06-19T12:27:27Z

Jpountz: related links

VLMa

2008-06-19T12:20:08Z

Jpountz: History of VLMa

[[image:VLMa.png|right]]

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in Ecole Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge to manage the streamers:
* give orders using the telnet interface,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://fr.wikipedia.org/wiki/Remote_method_invocation_(Java) RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

==History==

VLMa development started in January 2006 under the impulsion of Sylvain Cadilhac, who was network administrator at [http://www.via.ecp.fr VIA Centrale Réseaux] and [[VideoLAN]] treasurer. There were around 10 servers in charge of streaming television and radio over the campus network, and the aim of the project was to provide a web interface to manage every server.

[[Category:VideoLAN projects]]

VideoLAN projects

2008-06-19T11:42:03Z

Jpountz: add VLMa

== VideoLAN projects ==
=== Active projects ===

Here is a short list of the main [[VideoLAN projects|projects]] hosted by the VideoLAN project:

* [[VLC media player]], the core and main program, is a complete video solution that is a player, a live transcoder and a streamer.
* [[x264]], a free H264 / AVC encoder.
* [[LibVLC]] Python bindings.
* [[LibVLC]] Java bindings.
* [[LibVLC]] Common Intermediary Language bindings, i.e. the LibVLC .Net bindings.
* [[MiniSAPServer]] is a SAP sending server.
* [[VLMa]] a web application to manage several VLC at the same time.

=== Libraries ===
* [[libdvdcss]], a C library to access DVDs without having to bother about the decryption.
* [[libdvbpsi]], a C library to decode and generate MPEG TS and DVB PSI tables.
* [http://en.wikipedia.org/wiki/Libdca libdca], a DTS Coherent Acoustics decoder.

=== Deprecated, old projects and libraries ===
* [[VLS]] is a streaming server, used for professional streaming solutions. The {{VLC}} can now do the same. Please use the VLC media player.
* ''VLCS'', the mini VideoLAN Channel Server, can tell the client what is the multicast address associated to a channel.
* ''VLMS'', the VideoLAN Mini Server.

* [http://www.videolan.org/developers/libdvdplay.html libdvdplay], is a simple library designed for DVD navigation. It depends on libdvdread (which can optionally use libdvdcss). Use libdvdnav.

== SVN Hosted by VideoLAN projects ==
* [http://liba52.sourceforge.net/ liba52]
* [http://libmpeg2.sourceforge.net/ libmpeg2]

== See also ==
* '''See also [[:Category:VideoLAN projects]]'''

[[Category:About VideoLAN]] [[Category:VideoLAN projects]]

VLMa

2008-06-19T11:22:27Z

Jpountz: image on the right

VLMa

2008-06-18T22:47:01Z

Jpountz:

[[image:VLMa.png]]

[http://www.videolan.org/vlma VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in Ecole Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge to manage the streamers:
* give orders using the telnet interface,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://fr.wikipedia.org/wiki/Remote_method_invocation_(Java) RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

[[Category:VideoLAN projects]]

File:VLMa.png

2008-06-18T22:42:08Z

Jpountz: VLMa logo

VLMa logo

User:Jpountz

2008-06-16T18:23:04Z

Jpountz:

Adrien Grand

jpountz on #videolan@irc.freenode.net

VLMa developper.

VLMa

2008-06-16T14:37:17Z

Jpountz: integration with eclipse

[http://vlma.videolan.org/ VLMa] means VideoLAN Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in Ecole Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge to manage the streamers:
* give orders using the telnet interface,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://fr.wikipedia.org/wiki/Remote_method_invocation_(Java) RMI].

==Building VLMa==

VLMa uses Maven as a build system. Building VLMa should be as easy as downloading the source tree and running Maven.

$ git clone git://git.videolan.org/vlma.git
$ cd vlma
$ mvn install

Then Maven will download VLMa dependencies and build the project according to the configuration located in the pom.xml files.

==Integration with Eclipse==

Like any Maven project, there are two ways to integrate the VLMa project within Eclipse:
* using the [http://maven.apache.org/plugins/maven-eclipse-plugin/ Maven eclipse plugin] which is a maven plugin,
* using [http://m2eclipse.codehaus.org/ M2Eclipse] which is an Eclipse plugin.

Please refer to the project websites for more details.

[[Category:VideoLAN projects]]

VLMa

2008-06-16T14:31:31Z

Jpountz: building VLMa

VLMa

2008-04-26T10:03:00Z

Jpountz: add relevant category

[http://vlma.videolan.org/ VLMa] means VideoLan Manager

The VLMa is a Java application which provides a daemon and a web interface to manage several VLC streamers.

It is being used in Ecole Centrale Paris to broadcast Satellite and TNT channels over the network.

Its source code is available under GPLv2 or later in VideoLan's [[Git]], browsable on the web via [http://git.videolan.org/?p=vlma.git;a=summary gitweb].

$ git clone git://git.videolan.org/vlma.git

The daemon is in charge to manage the streamers:
* give orders using the telnet interface,
* monitor streamers' state using SNMP,
* draw RRD graphs.

The web module provides a user-friendly interface to communicate with the daemon. The communication between the daemon and the web interface is done using [http://fr.wikipedia.org/wiki/Remote_method_invocation_(Java) RMI].

A demo version of the web interface is available at http://vlma-demo.videolan.org.

[[Category:VideoLAN projects]]

VLMa

2008-04-22T21:42:25Z

Jpountz: demo version