>>22
No, I just fixed a couple of outstanding bugs, like archival of capcoded posts, threads without images in the OP and some other crap the mods can get away with that broke your regexps, the timezone problem when parsing 4chan dates (nice usage of timelocal there) and gave up on others, like proper archival of thumbnails without height= and width= (but I think I know what's wrong with it now that I think about it), $text needs to be uri_escape'd on searches (seemed a really easy change until I looked at how you did the template system). I also never figured out why the archiver broke on posts below 10 (didn't have enough time). Fixing the 15 newlines limit on posts that are archived without entering Reply mode is also going to require touching stuff I really don't feel like touching.
Another fun thing is that I now have a database with 3 million rows that contains fields with UTF-8 in it, but MySQL thinks it's latin1, because DBI->connect() didn't have the mysql_enable_utf8=>1 parameter in it. And the only way to fix it is to dump the entire database, change a couple of strings, reimport and hope MySQL doesn't feel like repairing the indexes through the "Repair with keycache" method, which literally means “I'm going to hog 99% of your CPU for at least 48 hours.” MySQL fucking sucks.
The auto-drop most likely has to do with the fact that I'm running two processes, one for /a/ and another for /jp/.
I should probably bind the /a/rchive to one IP and the /jp/-archive to another IP to prevent this sort of stuff from happening.