chrzyki / candc

A repository containing the original C&C CCG parser sources and newer modifications

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

What is this?

From the (now unavailable) project page:

The first full version of the C&C tools, version 1.00, has been released! The tools consist of the C&C CCG parser -- including the computational semantics tool, Boxer -- and the C&C taggers. (The tools also use the morphological analyser morpha.)

The tools have been developed by James Curran and Stephen Clark. Boxer has been developed by Johan Bos. Both the parser and taggers are written in C++ and are efficient enough to be used for large-scale Natural Language Processing tasks.

The tools are released under an academic (non-commercial) licence. If you are interested in a commercial licence then please email candc@it.usyd.edu.au.

In order to download the tools you will need to register. The documentation pages contain descriptions of the tools, but please email us if there is something missing. The FAQ page contains some frequently asked questions.

Availability

The project page went down sometime in early 2016. Ever since then, here and there a number of people provided download links to the original 1.0 release of C&C. Most notably, probably, chbrown here.

This repository

In this repository I want to provided access to the original release sources and all the changes that were pushed to the SVN trunk but never made it into an official release. Also, potentially, I'd like to include my own changes to the C&C sources. Here, with much gratitude, many thanks to texttheater, who provided the sources to the 2614M SVN version of C&C.

For the push to the 2614M SVN version I also included all *.out and *.o files. These were in the original SVN tree and I decided to add them for the sake of completeness.

See the releases section for a preserved original state of the sources. You can safely go there and completely ignore all my small 'update' commits which I mostly just did while working through the scripts and sources to get a better understanding of how everything in here plays together.

License

The original license is included. I'd be very grateful for any help with making it compatible (or checking compatibility in the first place) with a modern (and not hand-tailored) license.

About

A repository containing the original C&C CCG parser sources and newer modifications


Languages

Language:C 57.2%Language:Prolog 27.6%Language:C++ 9.0%Language:Python 1.8%Language:JavaScript 1.4%Language:Perl 1.1%Language:Lex 0.9%Language:Shell 0.5%Language:Makefile 0.2%Language:TeX 0.2%Language:Perl 6 0.1%Language:CSS 0.0%Language:HTML 0.0%Language:sed 0.0%