Joel Reicher wrote: >Exactly. I haven't tried to solve that problem, but to make it easier to >experiment with solutions. The output routine of the script can be >easily altered to sort the overall folder by a call to refile for each >message (which is very slow) or to create sequences (which would be quick).
The trouble with using sequences is that all the existing nmh programs ignore order within sequences and just handle the messages in ascending numerical order; this is pretty useless for threading and means we can't just tweak sortm and pick. (Unless you want to maintain folders in thread order, which would require enough intelligence in refile to insert new messages without having to rethread the entire folder.) >I believe threading >should be done as dynamically as possible. It's not a computationally slow >procedure anyway. Linear in the number of messages, I think. The bottleneck >is in parsing all the messages for their headers. Agreed. -- PMM _______________________________________________ Nmh-workers mailing list [email protected] http://lists.nongnu.org/mailman/listinfo/nmh-workers
