Re: Dealing with long lines

Ken suggests a summary of the "inc long lines" threads of last
fall would be useful:

 - Andy's patch has a memory bug, and Andy said he wanted to
   re-think the design anyway.

 - My patch I still like the design of, but it was complicated
   to continue to support netsec_readline's de facto interface
   of NUL-terminating the returned byte array.

I have been working since then to update callers of
netsec_readline to not require the NUL string terminator.

I have two patches in this effort waiting to be committed (or
rejected :-).  One of them ("imap_sasl_callback protocol check")
you have already reviewed but not yet committed.

When this background work is all committed, I will rework my
"long lines" patch, without needing the messiest parts.

 < Stephen