Zebra_Database is an advanced, compact (one-file only), lightweight, object-oriented MySQL database wrapper built upon PHP’s MySQL extension. It provides methods for interacting with MySQL databases that are more powerful and intuitive to use than PHP’s default ones. It supports transactions and provides ways for caching query results either by saving cached data on the disk, or by using memcache.The class provides a comprehensive debugging interface with detailed information about the executed queries: execution time, returned/affected rows, excerpts of the found rows, error messages, etc. It also automatically EXPLAIN‘s each SELECT query (so you don’t miss those keys again). It encourages developers to write maintainable code and provides a better default security layer by automatically escaping strings. Zebra_Database‘s code is heavily commented and generates no warnings, errors, or notices when PHP’s error reporting level is set to E_ALL.
TbsSQL is a small wrapper class for working with SQL databases within PHP scripts. It has few methods and two properties and it’s only one file per database for installation. Its most interesting feature is that it can protect and format values before merging them into the SQL statement. It has a cache feature, a trace mode, a debug console, and it can return data as arrays, standard objects, specific class objects, or clone objects. It can also delete PHP connection variables after making the connection.
PersistClass provides access to relational databases for PHP. It attempts to focus on simplicity and convenience by eliminating the need for boilerplate, meaningless code. It can establish database connections (MySQL, Oracle, etc.), execute SQL queries, parse query results, provide convenience methods (read a single row or value, etc.), escape data (against SQL injection, HTML escaping for outputs), handle transactions, and provide strict error handling and object persistence (storing and retrieving PHP objects in db tables).
SDLbits is a very lightweight SDL wrapper for Java. This library was designed to be very small and simple. It does not use Swig, but goes directly from "native" Java definitions to the sdlbits C wrapper library to SDL or OpenGL (in C). All symbols are exposed as they would be in C, except the "SDL_" prefix has been removed, since all names are inside the SDL class. The SDL structures from C are translated to very simple Java classes, such as SDL.VideoInfo. Most fields are exposed as read-only "get" methods, but "set" methods for writable members are supported as well. These classes simply interface to the actual C data with a ByteBuffer "pointer".
The LogicBoxes API is a full-control set of programming interfaces that allow you and your resellers to develop your own Website and control panels and integrate them into LogicBoxes OrderBox. Every function of LogicBoxes OrderBox is exposed via the OrderBox API. You can duplicate any or all of the functionality at your end by integrating into the API. The API comes with a specification, client side kits, sample forms, sample code, and complete documentation. Integrating with the API is a breeze. The API is based on SOAP/WSDL. Client side kits are available in PHP, Perl, ASP.NET, and Java. Additionally, since the protocol is standards-based, the API is truly platform independent. Monitoring systems prevent API abuse and any denial of service attacks on the API servers. The API is scalable and secure.
Multicrush is a wrapper around pngcrush that distributes the work of brute-force compressing a single image over several pngcrush processes. This gets you a slightly less than a linear speedup, caused by different compression methods having slightly different durations. For example, the author has measured a speedup of 1.92 on a 2-core Intel T5300, and 3.86 on a 4-core Intel Q6600. The only requirements are a copy of pngcrush and at least Python 2.4. If you have version 2.6 or higher, multicrush can automatically detect how many CPUs to use.
Libfairydust is a small wrapper library intended for use with GPU clusters that 'hijacks' CUDA and OpenCL calls. It can be used to 're-route' calls to a certain GPU, so a process requesting GPU#0 might end up running on GPU#4 without knowing (or caring) about it. This works completely transparently and does not need any sort of 'cooperation' from the application, changes to code, or relinking.