cvs2unicode, an 8-bit to Unicode converter with automatic codepage detection for CVS

Please follow this link for project documentation.

Features

Support for :local: CVS protocol
Automatic detection of text/binary contents based on CVS substitution modes. Only text fiels are converted.
Support for all standard Cyrillic charsets (KOI8-R, windows-1251, ISO-8859-5, IBM866). Files are read line-by-line, so a single versioned file can have differently encoded lines (which is usually the case if the file was re-encoded between revisions).
Support for double-encoded files (e.g: KOI-windows-KOI (KWK), Unicode-KOI-Unicode (UKU)). The idea is borrowed from Andrzej Novosiolov, who originally added this feature to the FAR Manager.
Automatic codepage detection if external dictionary (aspell/ispell/myspell/hunspell) is available.
Integration with Hunspell via HunspellBridJ to spell different forms of the same word. Automatic codepage detection works best with Hunspell (Russian and/or Ukrainian dictionaries should be installed so that Hunspell can find them).
Support for interactive charset selection where existing dictionary is not enough (e.g. for misspelled words). New words are automatically added to the user dictionary.
Vim integration. If you're unsure which charset to pick, you can jump to exactly the same line with Vim, and examine the context.

mvn exec:java