Re: [Nmh-workers] m_getfld branch merged

[Ken Hornstein]

David,

Thanks for all your hard work!

>There is a small performance penalty, I see 10% in my scans
>of big folders.  It's due to m_getfld() now having to keep
>track of the file stream read position in case the caller
>changes it.  If necessary, we could replace the uses of
>fseek()/ftell() by callers with more efficient bookkeeping.
>But at this point, I think correctness is paramount and
>performance is acceptable.

Ok, I'm wondering about this ... I see that the big difference at least
from a system call trace is the calls to lseek(), which I guess are
resulting from calls from fseek() or ftell().  So do a lot of nmh programs
use fseek() to change the stream pointer mid-stream?  I ask because maybe
programs like scan could tell m_getfld() that they don't need to keep
track of that stuff (or the programs that care about that could tell m_getfld()
that they do care about that).

--Ken