Re: hypermail-2a14-pl2

From: Daniel Stenberg <Daniel.Stenberg_at_sth.frontec.se_at_hypermail-project.org>
Date: Thu, 4 Mar 1999 23:59:46 +0100 (MET)
Message-ID: <Pine.GSO.4.10.9903042339410.15573-100000_at_metal.sth.frontec.se>

On Thu, 4 Mar 1999, Craig A Summerhill wrote:

> I pulled down the source code and compiled it on Digital Unix 4.0a. It
> compiled without a hitch, and seems to run.

I trust that was the tar-package then?

> From: "DANIEL SCHAEFFER"<daniel_schaeffer_at_kirkland.com>

Yeps. To qoute myself from the parse.c log:

date: 1999/03/01 22:42:14; author: daniel; state: Exp; lines: +15 -0 Corrected a problem in getname(). It occured when a From: line had a construct like "name"<addres_at_host.domain>. I.e no space between the quote and the less-than character.

I hope this have corrected those problems of yours.

> o Secondly, in the options.h I opted for:

I know I haven't stressed this properly, and that the docs are outdated: Don't edit options.h. Not at all. Ever.

I've now removed the file from the archive. It won't be there in next tar-ball.

> #define SHOWHEADERS 1
> #define SHOW_HEADERS "Date,From,To,Subject,Message-ID,In-Reply-To,References"

So, no matter what you edit that, it won't take effect. You should instead write a line like:

 show_headers = Date,From,To,Subject,Message-ID,In-Reply-To,References

in the config file.

> Problem is, hypermail included everything in the headers when it
> was marking up the message. I got all the Received lines, X- line,
> MIME version, Content-Type, etc. Everything...

Yeah, default will then be all headers. If that is good or bad I don't know...

> #define SHOW_HEADERS seems to be a new feature to this release of
> the code. Is the coding to support it complete? It clearly didn't
> limit itself to the fields I defined above.

If set in the config file, I believe it works as supposed. Can't say for sure of course.

> Oh, I was calling hypermail with the -c flag to point at a different
> configuration file. Would this affect the settings above?

No.

> configuration file in question does not have values in it overriding
> these pre-compiled settings. Am I correct in believing that you only
> need to include settings in the the .hmrc file you want to override
> the settings defined at compile time in options.h? Thought that is
> how it always used to be...

In order to 1 - make hypermail easier to debug and 2 - more easily use one compiled hypermail be able to produce different looking archives, I have step-by-step moved from that system. Hypermail will use its internal defaults unless other specified. Environment variables, command line flags and the config files are places to specify other behaviours.

> That's it for now. I'll be working on this all evening, and probably all
> night. Expect a few more messages...

I've got a few ;-) I may reply to some of them soon, unless I get stuck in debugging...

> Oh, BTW... the mark up seems considerably faster than the 2.0b2 software
> I last used.

That's good news! One would fear it would have gotten slower otherwise, since all strings are dynamic lengths now. Of course, more stuff than that have been altered...

--
             Daniel Stenberg - http://www.fts.frontec.se/~dast
   ech`echo xiun|tr nu oc|sed 'sx\([sx]\)\([xoi]\)xo un\2\1 is xg'`ol

Received on Fri 05 Mar 1999 01:03:43 AM GMT