Index CSV file data separation

[commonsense]

Index CSV file data separation

[rspeer]

I've been working on 5.3 for a while, but I could try going back and doing a build on 5.2 if it's clear what to fix. Can you tell me where the extra tab occurs?

Also, out of curiosity, what kind of index are you building? Trying to find a suitable way to index the data is the main task I have remaining for 5.3.

[DaveGerrard]

Hi - it's in the middle of some source data for the edge with the hash: 4ac44fb6ea699cd6861c26b716fdba15eb736806 - the source data looks like: /or/[/and/[/s/activity/omcs/omcs1,_possibly_free_text/,/s/contributor/omcs/<> /]/,/and/[/s/activity/omcs/vote/,/s/contributor/omcs/Alexi/]/]

(Apologies if I've put it there by mistake, but I don't think I have...)

I'm just building a quick and dirty index in MySQL for the time being - my plan is to mash up bits of ConceptNet graph with a graph of Twitter follower data I've built, but I only want to hydrate the bits of ConceptNet that are directly referred to by the folllowers (and those concepts' hypernyms / holonyms)... Hence wanting to put the CSVs in a relational data store, as (if indexed properly) it's the quickest way of accessing it, I think. (Certainly quicker than hitting it several tens of thousands of times over the web would be, anyway...)

[rspeer]

I can at least confirm that this is fixed in 5.3.