Free solver software for linux

Sudoku solver application was created for solving a Sudoku with a backtracking algorithm. Instead of using a 9 x 9 matrix, it extends the matrix to 10 x 36 (10 rows, 36 columns), storing information in the extra cells.

The last row is used for keeping track of how many cells, for the current column, are still available.

The columns 9-17 (0-based) are used for storing the numbers which are still available in rows 0-8.

The columns 18-26 are used for storing the numbers which are still available in columns 0-8.

The columns 27-35 are used for storing the numbers which are still available in each square (counting from left to right and from top to bottom).

Gecode/R is a Ruby interface to Gecode, allowing constraint programming in Ruby. The interface is intended for people with no previous experience of constraint programming, aiming to be easy to pick up and use in practice.
Constraint programming
Constraint programming is a declarative programming paradigm, you describe what kind of solution you want rather than how you want it computed. When using constraint programming you try to model the problem and then feed that model to the solver. The solver then searches for a solution by exploring the space of all possible assignments while using the constraints in the model to prune parts without having to visit them.
A popular example would be Soduko, to solve a Soduko with constraint programming you feed the rules (all numbers in each row must be distinct etc) to the solver, which then searches for a solution satisfying all the constraints.
Enhancements:
- A new Website has been launched, and a lot of documentation has been added (both code and user documentation).

PHAMLs primary initial goal of the PHAML project was to produce a parallel version of MGGHAT. This has subsequently been extended to include additional capabilities. The target architecture is distributed memory multiprocessors, such as Beowolf-type PC clusters.
MGGHAT is a sequential program for the solution of 2D elliptic partial differential equations using low or high order finite elements, adaptive mesh refinement based on newest node bisection of triangles, and multigrid. All aspects of the method are based on the hierarchical basis functions.
Adaptive refinement, multigrid and parallel computing have each been shown to be effective means of vastly reducing the time required to solve differential equations. However, effectively combining all three techniques is not easy and is the subject of current research. PHAML is an attempt to solve this problem.
Enhancements:
- Native conjugate gradients and GMRES solvers were added, along with support for DRUM and ZOLTAN_FILE in partition methods. -
- The tests have been modified to run in parallel, and assorted bugs have been fixed.

DiaCanvas2 is a flexible diagramming widget. It is based on GnomeCanvas. It uses its own constraint solver to maintain relationships between items on the canvas.
DiaCanvas2 is Model, View and Controller-based and makes full use of GnomeCanvas features.
Main features:
- Model/View/Controller based design: The DiaCanvas class only holds abstract data (using DiaShape objects), the data is rendered by one or more DiaCanvasViews.
- The widely used GnomeCanvas for visualization. This makes it easy to let DiaCanvas2 display anti-aliased diagrams with translucency (alpha) support.
- Export facilities for GnomePrint and SVG.
- Objects can be rotated/sheared/resized/etc. without the need to recalculate shapes. DiaCanvas relies heavily on the LibArt library (which is a standard GNOME library).
- Objects can connect to each other with handles. The connection is represented as a mathematical equation, which is solved using a linear constraint solver (see the reference documentation for more info). Handles do not need predefined connection points, but can connect to each other in a more generic way.
- Of course DiaCanvas2 has all the features a modern application needs, including undo/redo functionality.
- DiaCanvas2 is tested using Unit-tests. The unit test is a set of tests that ensure a consistent (application programming) interface.
Note that DiaCanvas is not related to the diagramming tool DIA, except for its look and feel.

GOBLIN is a C++ class library focussed on graph optimization and network programming problems. GOBLIN Graph Library deals with all of the standard graph optimization problems discussed by textbooks and in courses on combinatorial optimization.
This software package also consists of a shell interpreter which extends the well-known Tcl/Tk language to graph objects and a graph browser and editor tool. Executable solvers are available for practical optimization problems. The graph browser applies for teaching and scientific documentation purposes.
GOBLIN is open source software and licenced by the GNU Lesser Public License (LGPL). That is, GOBLIN may be downloaded, compiled and used for scientific, educational and other purposes free of charge. For details, in particular the statements about redistribution and changes of the source code, observe the LGPL document which is attached to the package.
Main features:
- The gosh interpreter extends the Tcl/Tk scripting language to graph objects in a natural way.
- The goblet graph browser and editor tool. Graphical front end to the library.
- An open class hierarchy which strictly separates between abstract classes (all mathematical algorithms are defined as methods of abstract classes), implementations (i.e. by incidence lists, adjacency matrices) and logical views (problem transformations).
- A generic branch and bound module with several applications to graph optimization.
- Logging and tracing functionality which allows to study the various algorithms by examples.
- A runtime configuration module controls the selection of mathemetical methods, logging information, and the tracing of data objects.
- Compile time configuration module for code optimization.
- A file interface which can be easily extended to new problem classes.
- Source code for executable solver programs.
Today, GOBLIN provides strongly polynomial algorithms for the following graph optimization problems:
- Shortest paths in graphs and digraphs with negative lengths.
- Negative cycles and minimum mean cycles.
- Strong and 2-connected components.
- Minimum spanning trees, arborescences and 1-trees.
- Maximum st-flows, feasible circulations and b-flows.
- Min-cost st-flows, b-flows and circulations.
- Assignment problems of any kind.
- 1-matchings, b-matchings, capacitated b-matchings, f-factors and degree-constrained subgraphs.
- Directed and undirected Chinese postman problems, T-joins.
Enhancements:
- This release comes with explicit code for orthogonal drawing of trees.
- There are also new generator methods for regular planara graphs.
- A lot of doxygen comments has been added, but dexcription are still far from complete.

Frepple is a Free Production Planning Library.
Plenty of planning tools are available for various production problems, different subprocess, different solver techniques.... A bit suprising, no proper solutions exist in the open source domain. Users often are left with expensive and inflexible tools and resort to spreadsheets to meet their needs.
This project aims at building an open source lightweight framework that easily and quickly delivers a solution for production planning problems. The initial focus is on small to medium sized master planning problems for the discrete manufacturing industry.
For the developer community, the project is also trying to establish a common ground framework for planning applications. Rather than building the basic building blocks (e.g. item, buffer, bill of material, ...) from scratch over and over again, developers can leverage a proven framework to extend with their own extension modules.
Main features:
- Modeling and solving framework for discrete manufacturing environments.
Key modeling constructs are:
- Item
- Buffer
- Resource
- Operation
- Demand
See here for a short domain model (requires Acrobat Reader).
A complete class diagram and API reference are available to developers.
- Heuristic, "MRP-like" solving algorithm respecting capacity, material and leadtime constraints.
- XML-based data input and output.
- Extensible and customizable architecture.
New modeling constructs and solving algorithms can be developed in C++ and loaded as a plugin module.
- The browser-based user interface and underlying database schema are based on the powerful Django framework.
Both can very easily be extended and customized.
- Embeds Python as scripting language.
Allows for flexible and easy integration and customization.
- Frepple comes as a library developed in C++.
It has no graphical user interface as such and requires to be deployed in a application.
Different applications are envisioned:
- Standalone application for use at the command line
- Microsoft Excel frontend
- Accessable from programming languages such as java, python, perl or visual basic
- Can be linked into your own C or C++ application
- Apache module for browser based user interface and network based access
- Supported on 32-bit linux and Windows environments, with Linux being the primary development platform.
Porting to other platforms is discussed under the developer section of the documentation.
- Licenced under the GNU lesser general public license.
Learn here how this license protects my and your work.
- To finish, a pretty bold statement
Frepple will be the worlds first true open source advanced planning system.
Enhancements:
- A major update in the user interface: all-in-one installation of the user interface, Python, Web server, and database.
- Big performance improvements in the user interface.
- Support for PostgreSQL, MySQL, and SQLite databases.
- The forecast module has been greatly improved, and allows for forecast distribution and netting of orders against forecast.
- There are important bugfixes.

Taskjuggler is a project management tool for Linux and UNIX system-based operating systems. Whether you want to plan your colleges shifts for the next month or want to build a skyscraper - Taskjuggler is the tool for you.
Instead of clicking yourself painfully through hundreds of dialog boxes you specify your Taskjuggler project in a simple text format. You simply list all your tasks and their dependencies. The resulting schedule is then presented as GANTT chart, Table, HTML, XML or CSV report.
Taskjuggler does not only honor the task interdependencies but also takes resource constrains into account. Using Taskjugglers powerful filtering and reporting algorithms you can create task lists, resource usage tables, status reports, project calendars and project accounting statements.
Main features:
- Automatic scheduling of interdependent tasks with resource conflict solver.
- Powerful project description syntax with macro support.
- Flexible working hours and vacation handling.
- Support for shifts.
- Multiple time zone support.
- Flexible resource grouping.
- Project accounting support.
- Task may have initial costs, finishing costs.
- Resource may have running costs.
- Support for simple profit/loss analysis.
- HTML and XML report generation.
- Support for plan and actual scenario comparisons.
- Project tracking support.
- Groupware support by using a revision control system such as CVS or RCS on the project description files.
- Support for central resource allocation database.