I would like to request the installation of the python package csvsort on stat1005.eqiad.wmnet. We generate a lot of different .tsv-files while analysing the SPARQL-Logs and they are sometimes too large to sort in memory.
Description
Description
Event Timeline
Comment Actions
I'd have to create a python debian package for this.
Q: can you used Hadoop? You wouldn't have memory problems then! :)