> > Looking at things ... it may be a simple fix, actually. I wasn't
> > envisioning any changes to m_getfld(), that's for sure.
> Alright, I think I've got a fix in there, and I've got a test which exercises
> the bug. I was right that the fix was simple, but I had to take into
> account all of the ways that scansbr.c was used. The only downside is that
> scan may sometimes do an extra read() when it doesn't have to, but that would
> only happen if your headers approach the 4k mark (like Steve's).
Ken, sorry, hate to say it, but...
$ git clone git://git.savannah.nongnu.org/nmh.git
$ mv nmh nmh-`date +%Y%m%d`
$ cd nmh-`date +%Y%m%d`
$ uip/scan +scan-bug
1 Fri Mar 16 09:13 Ken Hornstein Re: [Nmh-workers] temporary files | >Rather than pr
2 Fri Mar 16 10:35 Ken Hornstein Re: [Nmh-workers] temporary files | >If we pr
3 Thu Mar 15 17:43 Jon Steinhart Re: [Nmh-workers] temporary files | Jerra
4 Wed Mar 14 19:32 Nico Kadel-Garcia Re: wine 1.4 is out | --
6 Wed Mar 14 10:22 REDACTED Resolving dependency problems | For the last