From: Craig A Summerhill (craig@cni.org)
Date: Tue Apr 20 1999 - 04:04:20 CDT
On Sun, 18 Apr 1999, Daniel Stenberg <dast@sth.frontec.se> wrote:
>
> On Fri, 16 Apr 1999, Paul Haldane wrote:
> >
> > I have a from line that I reckon the parser handles wrongly
>
> Ah, you're right.
>
> I've changed it now to only scan for ()-comments if there was no <>-address
> found in the line.
>
> In a first glance, it seems to make your lines work better and I couldn't see
> any other examples fail because of it either.
>
> As soon as we have something here that might work, I'll add it to the
> hypermail parser and release a new tar-ball. Many problems will be solved
> with this.
>
> (Oh, right, the mime-decoding can be ignored by this parser since that'll be
> performed separatly before the function is called.)
Daniel,
I will send to you directly (and under separate cover) the OUTPUT of the
fromparse.c program for you to review if you would like.
One comment, I noticed in a mailbox of 2765 messages that the program
OUTPUT 5535 entries. Given this, I hope that this code doesn't reflect
the programs ability to parse messages in a mailbox. I suspect that
it picked up a few extra entries along the lines where people had
embedded within a message a forwarded header such as this:
From: Craig A. Summerhill <craig@cni.org>
Date: Tue, 20 Apr 1999 08:30:00 -0500
Subject: something
Just thought I would check about this though. Otherwise, there is
a counter problem in the code...
-- Craig A. Summerhill, Systems Coordinator and Program Officer Coalition for Networked Information 21 Dupont Circle, N.W., Washington, D.C. 20036 Internet: craig@cni.org AT&Tnet (202) 296-5098