Hi thanks for the nice update, interesting. The thinking sounds very reasonable to me.
Ola and I will work a bit on the optimization of SQL view. One issue here is that when writing "generic" sql queries for tables like orgunitgroupsetstructure, one needs to select * (all) columns in order to be portable because the table has each column per groupset, ie. its dynamic. But I guess we can fix these things when generating the SQL from Java / DHIS. Re size, a good thing is that a typical pivot table won't contain the whole database, rather the data for a district and its facilities. We want to be able to compare with other districts (maybe inside the same province only) but for those we only need the aggregated data at the district level (not facility data) - which means a lot less data. Its also interesting to see how excel uses a "minimal" xml (<r><x v="0"/></r> etc). We could consider that for DXF 2 datavalues. Size/performance vs readability. Lars
_______________________________________________ Mailing list: https://launchpad.net/~dhis2-devs Post to : dhis2-devs@lists.launchpad.net Unsubscribe : https://launchpad.net/~dhis2-devs More help : https://help.launchpad.net/ListHelp
