Free modular software for linux

App::Modular is a Perl module with modularization framework for perl programs.
SYNOPSIS
package App::Modular::Module::Test;
use base qw(App::Modular::Module);
sub say_hello {
print "Hello, dear user!";
};
package main;
use App::Modular;
my $modul = instance App::Modular;
$modul->module(Test)->say_hello();
exit;
App::Modular aims to provide a framework which should it make very easy to programmes to create any kind of modular program.
It supports:
- module dependency solving
- autoloading of modules
- event handling (implemented as a contributed App::Modular module)

Modular Google Talk Bot (gtalkbot) is a modular bot system for Google Talk and other Jabber IM networks.

It uses simple Python plug-ins that register verbs, which appear as the first word of an IM and result in another IM being sent back to the user. An example MythTV bot is included.

To run mbot:

- Write a plugin! There is a sample on in plugins/mythbot.py
- Create a password file at /usr/local/share/mbot/passwd

The format is:

< username >< tab >< password >

For example:

exampleuser@gmail.com banana

- Create a jabber or Google Talk account for the bot
- Start up mbot:

./mbot plugins < botaccount >@gmail.com < botpassword >

- Now you can IM the bot account, and off you go. Youll need to start off with a "auth < password >" to login to the bot

F Modular Synthesizer (FMS) is basically a tool to generate all kinds of sounds. It should work on any up-to-date Linux system.
Main features:
Playing sounds
- with any frequency and volume
- one after another or at the same time (mixer) or both
- stereo (different sounds on different channels)
- with any balance between the two channels
- with built-in sweep that even follows frequency modulation
- like speech synthesis [listen]
- both on-the-fly playing with direct user access and asynchronous, pre-computed output; also combining the two, taking the best of both
- with all modulation options described below
Making noise
- now although it tends to get noisy in any way, we mean real noise here
- in 4 different ways
- nearly-white noise
- narrow band noise
Modulating
- amplitude (volume)
- frequency
- minima, maxima, amplitude and frequency of the modulation (here the "Fully Modular" comes to existence)
- balance, moving the sound from left to right and back in stereo mode
- narrowness of the noise frequency band in many different ways
Making music
- some-kind-of-midi-mapping mode (sound + envelope => instrument, FMS midi file format)
- auto-converter and player for MUS file format, listen [here] (Musplay)
- auto-composer for those who like a rather modern kind of music
- chords
Displaying
- spectrum of a sound
- oscillogram of a sound
- mixes between different sounds with different volumes / frequencies
Saving
- sounds as oscillograms in FMS file format (mathematical description, not complete wavetables)
- sounds as Fourier overtone amplitudes
- music in FMS midi format
- FMS output in wave format
- wave file sounds in FMS file format (auto-converter) - experimental
GUI
- a somewhat ugly and extremely limited tcl/tk gui
- FROCOR, an installation that connects the FMS sound backend to graphical interfaces and outputs
- a fully modular Qt GUI - experimental, but Ill give you a [screenshot]
Enhancements:
- This release fixes all bugs discovered via user feedback, and should be quite stable.
- Sweep frequency modulation is now implemented, and using both blur and frequency modulation on the same sound at the same time is possible.

AlsaModularSynth project is a realtime modular synthesizer and effect processor.
Main features:
- MIDI controlled modular software synthesis
- Realtime effect processing with capture from e.g. "Line In" or "Mic In".
- Full control of all synthesis and effect parameters via MIDI.
- Integrated LADSPA Browser with search capability
- JACK Support
Modular synthesis is one of the most flexible and fascinating ways of making electronic music. The principles of voltage-controlled modular synthesis have been stated in the famous paper Voltage-Controlled Electronic Music Modules by Robert A. Moog , presented in 1964 at the Annual Convention of the Audio Engineering Society.
AlsaModularSynth is a digital implementation of a classical analog modular synthesizer system. It uses virtual control voltages to control the parameters of the modules. The control voltages which control the frequency e.g. of the VCO (Voltage Controlled Oscillator) and VCF (Voltage Controlled Filter) modules follow the convention of 1V / Octave. Read the first two pages of the Moog paper to understand why this logarithmic scale is so important.
A real analog synthesizer has a lot of knobs and switches which give immediate access to all important parameters of the generated sound. AlsaModularSynth can be controlled by MIDI controller boxes, like e.g. the Doepfer Pocket Dial . MIDI events can be assigned to module parameters in a very flexible way by using the Control Center dialog of AlsaModularSynth.

Graph::ModularDecomposition is a Perl module for modular decomposition of directed graphs.

SYNOPSIS

use Graph::ModularDecomposition qw(pairstring_to_graph tree_to_string);
my $g = new Graph::ModularDecomposition;

my $h = $g->pairstring_to_graph( ab,ac,bc );
print "yesn" if check_transitive( $h );
print "yesn" if $h->check_transitive; # same thing
my $m = $h->modular_decomposition_EGMS;
print tree_to_string( $m );

This module extends Graph::Directed by providing new methods related to modular decomposition.

The most important new method is modular_decomposition_EGMS(), which for a directed graph with n vertices finds the modular decomposition tree of the graph in O(n^2) time. Method tree_to_string() may be useful to represent the decomposition tree in a friendlier format; this needs to be explicitly imported.

If you need to decompose an undirected graph, represent it as a directed graph by adding two directed edges for each undirected edge.

The method classify() uses the modular decomposition tree to classify a directed graph as non-transitive, or for transitive digraphs, as series-parallel (linear or parallel modules only), decomposable (not series-parallel, but with at least one non-primitive module), indecomposable (primitive), decomposable but consisting of primitive or series modules only (only applies to graphs of at least 7 vertices), or unclassified (should never apply).

Modular Audio Recognition Framework is an open-source research platform and a collection of voice/sound/speech/text and natural language processing (NLP) algorithms written in Java and arranged into a modular and extensible framework facilitating addition of new algorithms. MARF can run distributedly over the network and may act as a library in applications or be used as a source for learning and extension. A few example applications are provided to show how to use the framework. There is also a reasonbly detailed manual and the API references in the javadoc format as the project tends to be well-documented.
Enhancements:
- This is the first release of the Distributed MARF proof-of-concept (PoC) implementation.
- MARFs pipeline stages can run on multiple computers using 3 different communication technologies: CORBA, RMI, and Web Services.
- The six implemented services include SpeakerIdent Front-end Service (plus the client), MARF Pipeline, Sample Loading, Preprocessing, Feature Extraction, Training and Classification.
- This is based on the snapshot-0.3.0-devel-20060811.

Modell (Modular Extension Lisp Language) is an object-oriented, modular variant of the LISP programming language, specially suitable for embedding in applications using a C++ API.

Modell is implemented as a C++ library (libmodell) and a command-line interactive interpreter for testing and rapid development, and includes modules for diverse tasks such as math, strings processing, interaction with the operating system, and others.

Installation:

Just the usual:

$ ./configure
$ make
$ make install

MetaRuby contains miscellaneous libraries (useful now) for a future Ruby-in-Ruby interpreter including Array/Hash/String as abstract ("Hollow") classes, an undo queue, a statistical time-profiler.

It also has an event loop, a modular marshaller ("ToSource"), a specification for a modular+reflexive+homoiconic remote call system ("LGRAM"), a declarative type system, a schema for expressing Ruby source code as proper (non-special) Ruby objects, etc.