importing into a big database causes hangups (refreshBibtexIdLinks)

[kristiank]

Our Aigaion database now holds 10,000+ authors and 80,000+ publications (these values are will grow to double their size when importing new data has been fullfilled).

When importing more than 300 entries (or more than 900 entries with index added as explained in issue #8) the importing stops and loads an empty white page.

When looking at the mysql processlist during import, it seems that the most time consuming query is the data transfer for refreshBibtexIdLinks(). This function is called for every added publication, and must send all the pub_id and bibtex_id from the mysql database over to php (that is 2*80,000+ values in our case). I am not yet sure what the purpose of this function is, but it doesn't seem to affect the database, only keeps all the values in the user's CI session.

Could someone explain what this function does?

[kristiank]

This problem is being dealt with in Issue #10.