C&C/Boxer HTTP API

Looking for the C&C source code? I'm not affiliated with the developers of C&C and Boxer. However, one of my software packages contains it as a dependency. You can find it here: KNEWS.

The C&C/Boxer HTTP API provides an easy access to the C&C tools without the need to install software on your machine other than cURL or similar libraries for HTTP request (that usually come with the operating system anyway).

A few reasons to use the API:

  1. integrating the C&C tools into other projects and building upon its technology is now an HTTP request away.
  2. in a teaching environment, it's just much easier to give one command line to execute to the students, rather than having them install the whole pipeline on their machines.
  3. even microcomputers like the Raspberry-PI will be able to get a full semantic parse in a split second.

I mantain an experimental installation of the HTTP API at this endpoint:
http://gingerbeard.alwaysdata.net/candcapi/
An example call is:

$ curl -d 'Every man loves a woman' 'http://gingerbeard.alwaysdata.net/candcapi/proxy.php/raw/pipeline?semantics=fol'
fol(1,not(some(A,and(n1man(A),not(some(B,some(C,and(r1patient(B,C),and(r1agent(B,A),and(v1love(B),n1woman(C))))))))))).
There is a limit of 5,000 requests/hour per client. The source code is on github.

Disclaimer

The service is provided as is. I do not guarantee its continuity nor the quality of the output. Please do not rely on this API to perform scientific experiments, I cannot guarantee the reproducibility of the results obtained with this API (use instead the stable version of the C\&C tools. By using this Web service you are also agreeing to the C&C System Licence Agreement and to the Boxer licence agreement (in the main directory of the C&C tools, LICENCE-BOXER.txt).

To learn about the URLs and options, read the documentation.

back to my homepage.