From owner-freebsd-ports Mon May 29 19:07:28 1995 Return-Path: ports-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id TAA01471 for ports-outgoing; Mon, 29 May 1995 19:07:28 -0700 Received: from eikon.regent.e-technik.tu-muenchen.de (root@eikon.regent.e-technik.tu-muenchen.de [129.187.42.3]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id MAA06151 ; Mon, 29 May 1995 12:57:27 -0700 Received: from vector.eikon.e-technik.tu-muenchen.de ([129.187.142.36]) by eikon.regent.e-technik.tu-muenchen.de with SMTP id <43104>; Mon, 29 May 1995 21:56:59 +0200 Received: from localhost (localhost [127.0.0.1]) by vector.eikon.e-technik.tu-muenchen.de (8.6.11/8.6.9) with SMTP id BAA18094; Mon, 29 May 1995 01:35:40 +0200 Message-Id: <199505282335.BAA18094@vector.eikon.e-technik.tu-muenchen.de> X-Authentication-Warning: vector.eikon.e-technik.tu-muenchen.de: Host localhost didn't use HELO protocol To: "Rodney W. Grimes" cc: asami@FreeBSD.org, ports@FreeBSD.org, gj@FreeBSD.org Subject: Re: Hylafax - a patch to fix the broken port, tested against ctm444. Date: Mon, 29 May 1995 01:35:40 +0200 From: Julian Howard Stacey Sender: ports-owner@FreeBSD.org Precedence: bulk Hi Rod, Re. > > CC Readers: > > If you don't want to wait for the CVS + CTM delay, grabs it now. > > My unbroken original Hylafax port also remains in freefall:~jhs > > (this never got sent to ctm recipients, as between my original commit > > & the ctm 0438 transmission, someone broke it. > > > > I see a real problem here, the file patch-aa will never contain the > $Header$ strings you commit due to the way cvs works!! Yes, I realise that patch-aa should not have access to its's own cvs header string. Sam Leffler's patch.sh upgrades about 7 of his RCS header lines, as well as his code, all in files which are extracted from the distfile. My patch-aa treated Sam's RCS headers as plain text, (to show where we got the distribution from ) in the same way it changed the code. > I will probably > have to nuke the current patch-aa,v file. Import a correct one and then > do a cvs admin -ko on it so that all RCS strings are left untouched by > cvs on check out. It sounds like you think our CVS spots RCS strings embedded in the middle of patch-aa & changes them, when we need them unchanged ? I'm not sure if this is a bug, a feature, or even desirable (I guess: bug ?). > This is neither Julian's or Satashi's fault that this problem arrose, it > was a misunderstanding of just what CVS is going to do to this patch > file and requires a repository manager (mangler if you prefer) to make > it work correctly. > > > *** /usr3/ports/comms/hylafax.ctm444/patches/patch-aa Mon May 22 15:21:15 1995 > > --- /usr3/ports/comms/hylafax.jhs/patches/patch-aa Sun May 21 19:36:59 1995 > > *************** > ... > > DO NOT ATTEMPT TO COMMIT THIS PATCH, it will not do what you think it will > do. The $Header$ and $Id$ strings will change to 1.2, not what you want > them to!! > > I will look closely at the patch and see if I can do the patch in such > a way that it applies cleanly without looking at the RCS strings (either > narrow the context diff with a -c 2 or -c 1, or possible use a unidiff > format). Thanks Rod, It'll be great to see Hylafax resolved :-) > Can you give me the original ``clean'' full version of the patch-aa file > as it should be to apply to the fetched port files. Sure, My original is in something like : freefall:~jhs/hylafax.as.committed/patches/patch-aa I have also mailed you a personal copy (12K). I generated patch-aa by uudecoding the commented out preamble email in patch-aa (which is from Sam's post to the flexfax list). Uudecoding creates patch.sh which applies some sort of non context diff. I applied patch.sh on the extracted distfile, then generated patch-aa with diff -c. The uuencoded original mail comes from Sam Leffler, author of Hylafax. patch-aa is dedicated to reproducing what Sam released, there's none of my stuff in there, so if you need to, you can work from Sam's original. > This will take me some time to resolve, as I will have to fetch the > bits over a 28.8K line :-(. Sorry it'll clog your line a bit, the distfile is > 1.06 M byte. But Hylafax is a nice package to have around :-) (PS I run my V32bis at 9K6 to my IP provider currently, (as 14.4 fails)). If it help to save you downloading, I have a directory freefall:~jhs/hylafax.compile (or something like that) where I was test compiling today, the make was nohup'ed & cc'ing when I left. Change what you want in there, The Makefile has a couple of slight diference from the distrib. (eg I suppressed check on ghostscript exec. temporarily). Thanks :-) Julian S. jhs@freebsd.org