Eek, of all the plays to quote, Racine's _Phèdre_! Horrible memories of high
school French class (and of falling asleep at a performance thereof) resurface.
When Hippolytus said «je pars, cher Théramènes» I should have said it with him
and done it.
Louis-David Mitterrand wrote,
| Yes but how come the error only happens for a small minority of messages
| on a 150 message mailbox processed with "formail -s procmail < mbox"?
Procmail writes a bufferful at a time when it feeds a message to a program.
If head exits somewhere during the last bufferful and procmail has no more
left to write, there's no write error. If head exits when procmail still
has some text to send it, there's a write error. So it all depends on whe-
ther a buffer boundary falls somewhere between head's exit point and the end
of the message; in your case, whether a buffer boundary falls in the signa-
ture. Apparently, as big and annoying as the signature is as signatures go,
it's a lot smaller than procmail's write buffer, so only for a small minority
of messages does the boundary between buffers fall within the signature.
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail