Creating the take home database: Difference between revisions

From Knot Atlas
Jump to navigationJump to search
No edit summary
Line 6: Line 6:
#Get a coffee, or perhaps go to sleep, travel to a conference etc.
#Get a coffee, or perhaps go to sleep, travel to a conference etc.
#Eventually, check that [http://katlas.org/data/katlas.rdf.gz] has been created.
#Eventually, check that [http://katlas.org/data/katlas.rdf.gz] has been created.
#Then re-make the smaller data files - type "<code>make all</code>".


==What does this do?==
==What does this do?==

Revision as of 13:34, 18 July 2007

This page is really only for Scott and Dror. It documents the procedure for creating RDF dumps from the Knot Atlas. If you're only interested in the actual data, go to The Take Home Database.

Quick start

  1. Go to /www/html/data/
  2. As root, run ./create-rdf.sh
  3. Get a coffee, or perhaps go to sleep, travel to a conference etc.
  4. Eventually, check that [1] has been created.
  5. Then re-make the smaller data files - type "make all".

What does this do?

XML dump

First, we dump the current version of every page in the Knot Atlas to XML, using the /w/maintenance/dumpBackup.php script from mediawiki. This produces the file katlas.xml.

Convert to RDF

This is actually a two step process. First, we create RDF statements in an on-disk RDF repository. Second, we dump these statements back to the file katlas.rdf

Cleaning up

The file katlas.xml gets deleted, as it's not so useful, and the file katlas.rdf gets gzipped to katlas.rdf.gz. It's then available at [2].

Retrieved from "https://katlas.org/index.php?title=Creating_the_take_home_database&oldid=1688889"