Re: pkg/29026

["John D. Baker" <jdbaker%mylinuxisp.com@localhost>]

The following reply was made to PR pkg/29026; it has been noted by GNATS.

From: "John D. Baker" <jdbaker%mylinuxisp.com@localhost>
To: Bernd Ernesti <veego%netbsd.org@localhost>
Cc: gnats-bugs%netbsd.org@localhost
Subject: Re: pkg/29026
Date: Thu, 30 Jun 2005 01:29:47 -0500 (CDT)

 On Thu, 30 Jun 2005, Bernd Ernesti wrote:
 
 > > I replaced the comms/minicom/patches/patch-ae file with the one you
 > > attached.  I didn't bother generating a new SHA1 signature to put in
 > > the 'distinfo' file, since none of the other patch files' SHA1 signatures
 > > match what's in the 'distinfo' file anyway.
 >
 > They have to match, maybe you use something which doesn't produce the
 > correct SHA1 signature, because your build output shows that only the
 > new patch-ae doesn't match the stored signature.
 
 I was using 'digest' (/usr/pkg/bin/digest) that was installed as part
 of the bootstrap procedure.
 
 I've figured it out.  The CVS/RCS header is added to the patch file
 after the signature is calculated and recorded in 'distinfo'.  The
 integrity check of the patch files must strip the RCS header before
 piping it through digest.
 
 > But then, the build should not continue because of that.
 
 Stranger and stranger.
 
 > I change the patch file and then close this and the other pr.
 
 When my current operation finishes, I'll blow away my comms/minicom and
 update to pick everything up fresh and see how it goes.
 
 -- 
 John D. Baker, KN5UKS                    NetBSD     Darwin/MacOS X
 jdbaker(at)mylinuxisp(dot)com                 OpenBSD            FreeBSD
 BSD -- It just sits there and _works_!