Valentina is cross-platform SQL and non-SQL columnar database that allows development of client-server [Web] solutions and applications with an embedded local database using the same sources. Valentina DB provides an Object-Relational model, but you can also mix Relational and Extended Navigational. It introduces a revolutionary model abstraction "Link" that greatly simplifies db schema, and therefore simplifies SQL queries. It supports disk and in-memory databases, and field types including Bit, BLOB, Pictures, and Enum. It provides advanced features such as triggers, views, stored procedures, regular expressions, XML, full-text search, and calculated fields. It exists as Valentina Server, Valentina Studio, and a set of Valentina ADKs for all major programming languages.
GraphicsMagick is a robust collection of tools and libraries which support reading, writing, and manipulating an image in over 90 major formats including popular formats like DPX, DICOM, BMP, GIF, JPEG, JPEG-2000, PDF, PNG, PNM, SVG, and TIFF. A high-quality 2D renderer is included, which provides a subset of SVG capabilities. C, C++, Perl, Tcl, and Ruby are supported. Originally based on ImageMagick, GraphicsMagick focuses on performance, minimizing bugs, and providing stable APIs and ABIs. It runs on all modern variants of Unix, Windows, and Mac OS X.
ProcPile is a pure Ruby distributed computing cluster environment. Its goal is to allow fast prototyping on smaller datasets, especially those where serialization and deserialization are minor expenses relative to the core processing task. Once the algorithms are settled, you should be able to scale out to full production usage. At that point, you can replace the core processes with Ruby modules written in C (or Inline::C), or move the design to a more "grown-up" clustering environment.
Virtual Print Engine Community Edition is a report generator, print engine, and PDF library. It allows you to create documents like reports, forms, drawings, and diagrams on-the-fly by placing objects like text, lines, and bitmaps in any position using function calls. Flexible dynamic layouts are supported. Documents containing tens of thousands of pages can be created with a very small memory footprint. New documents can be assembled from several single documents.
Winnow efficiently trains and operates any number of unique Bayesian (Naive Bayes) classifiers on large sets of content. It has very high performance and works with very small training and unbalanced training sets. It has been used to power an innovative Web feed reader that uses smart tags, which learn and find the content you want to see, from more sources than you can follow with traditional feed readers. It works particularly well with Ruby and Ruby on Rails.