The tricky one is consolidating multiple edits to a page into one entry on the Recent Changes page, the way UseMod and Phase II did. This is a big performance hit.
Do we have some measurements on this? Would it be worth the effort if I tried to come up with some SQL that did this for the new script?
I have only compared the actual query used in Phase II (with that ugly LEFT OUTER JOIN and GROUP BY) with my present query (actually two queries, because MySQL doesn't have UNION yet). If you want to try other queries on the new database schema that might work with comparable speed to the present one, I can try those out on my server and time them using a version of Neil's bot.
lcrocker@nupedia.com wrote:
The tricky one is consolidating multiple edits to a page into one entry on the Recent Changes page, the way UseMod and Phase II did. This is a big performance hit.
Do we have some measurements on this? Would it be worth the effort if I tried to come up with some SQL that did this for the new script?
I have only compared the actual query used in Phase II (with that ugly LEFT OUTER JOIN and GROUP BY) with my present query (actually two queries, because MySQL doesn't have UNION yet). If you want to try other queries on the new database schema that might work with comparable speed to the present one, I can try those out on my server and time them using a version of Neil's bot.
...and I actually _prefer_ the presentation of the new software, without the filtering.
Neil
wikipedia-l@lists.wikimedia.org