Toby Johnson schrieb:
Dirk wrote:
Ok, now I found the last one with 17 seconds difference.....
Wow, not good... now that I look at that code again, I can't see the
harm in upping that threshold to, say, 60 seconds... we're already
matching on physical name, action type, and author.
While talking about it, I found incredible 25 seconds ;-). I can only
capture this type of error, if I later reference the item again. If the
mapping doesn't work, I map the items to orphaned and exclude them from
the conversion.
In fact, in MergeParentData(), we could find the timestamp of the
*next* record, and use that as an upper bound instead of an arbitrary
one. If it took 17 seconds between recording that information in the
parent and in the child, surely nothing else happened with that child
record in the meantime, correct?
I thought about something similar, too. Could be a feasible solution.
But I don't know how to write this in SQL syntax ;-)
Dirk
_______________________________________________
vss2svn-users mailing list
Project homepage:
http://www.pumacode.org/projects/vss2svn/
Subscribe/Unsubscribe/Admin:
http://lists.pumacode.org/mailman/listinfo/vss2svn-users-lists.pumacode.org
