Working with the zipped file itself

[brihijoshi]

Hey! Thanks for such a great README.md, there is literally no good documentation for Freebase.

A lot of us who want to work with freebase but cannot extract the KB due to its huge size have to resort to working with the compressed file. Could you extend the scope of the script to also do ETL operations from the zipfile instead?

Also, if this feature also exists, could you kindly guide me as to how I can run the scripts?

[nchah]

Hi Brihi,

Thank you for the kind comments!

Creating versions of these scripts that can work with the compressed file is actually something I would like to implement in an updated version of this repository. In the meantime, it's possible to use the z* commands (zless, zmore, zgrep) to work with the zipped Freebase data dump file. The output could then be piped to another file (e.g. using zgrep to get the triples for say /people/person/place_of_birth and loading it into a triplestore or other database).

All the best!