i wrote: |"Stephen Isard" <ksaexz5...@snkmail.com> wrote: ||I have an imap account to which I can connect with mailx but not s-nail.
||Anyway, mailx is doing the right thing for that imap server, but s-nail ||is not. You know, Gunnar Ritter's IMAP code has served many people pretty well, it seems to me. My problem is purely technical, i want to remove any longjmp(3) from the codebase, i want non-blocking I/O, i want object-based mailboxes. I want to have a better error recovery. E.g., if an error occurs during deletion of 50 messages, the operation should stop and not stupidly work the last 49 messages even if we're in error state. Or try to reconnect if the connection was lost. In case of errors, it should be possible to save messages somewhere. Whatever. The IMAP cache of Gunnar Ritter is a neat idea. (It simply records a log and works that in order, but nonetheless, who else has this?) I have just said this yesterday or so in a private message, but Gunnar Ritter never had either of 24/7 nor unexperienced users in mind. Both of which are decisions that i don't like. Also i want to point out that the comment of Dr. Werner Fink from SUSE, in Heirloom mailx, main.c, which says "Nevertheless, I do not like seg faults.", has been removed from S-nail. --steffen ------------------------------------------------------------------------------ __________________________________ S-nail-users@lists.sourceforge.net
