Hello,

as I got access to VM with big debian mailinglist archives only 
yesterday, so this week I was only playing with my small 
archives. This week I spent on measuring time - how fast is my 
current implementation, specially generating threads/trees of 
emails. And I got conclusion that also for smaller archives (=~ 
5000 emails) it is not so fast. It was about 3-5 seconds to 
generate tree from sqlite database. Biggest problem is to join 
emails without in-reply-to header to thread with similar/same 
subject. So I started writing new code (and changing format of 
tables in database) for email threading to have fast access to 
all emails in specific thread in cost of decreasing speed of 
adding new emails. Also I'm changing algorithm for generating 
tree of emails from DAG which comes from database (from in-reply-
to and references headers). I decided to use slightly modified 
topological sorting from bottom of DAG to deal with missing 
emails in threading view.

-- 
Pali Rohár
[email protected]

Attachment: signature.asc
Description: This is a digitally signed message part.

_______________________________________________
Soc-coordination mailing list
[email protected]
http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/soc-coordination

Reply via email to