nmh-workers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Nmh-workers] Features for 1.6


From: Jeffrey Honig
Subject: Re: [Nmh-workers] Features for 1.6
Date: Wed, 16 Oct 2013 22:36:28 -0400

On Wed, Oct 16, 2013 at 10:20 PM, Ken Hornstein <address@hidden> wrote:
>>One problem we have run into with mh-e and nmh 1.5 is that mh-e is no
>>longer able to read the comp formfile directly.  That's because of
>>some dynamically generated content, isn't it?
>
> Well, I suspect you're able to read it, but it's not very useful, is it?
> Because the default components file now supports mh-format(5) and the actual
> draft is generated by having comp process the components file via 
> mh-format(5).
>
> I guess I didn't think this would be a problem, since the front-end program
> I am used to (exmh) handled it fine.  Sorry about that; the change wasn't
> a secret, and I guess I thought if there was a problem then people would
> speak up about it!

Not your fault.  Several of us contribute to both projects and should
have caught
this during development of 1.5.  *blush*  I was aware of this change,
but did not
look into how it would affect mh-e.

>>Right now we are changing the code to use comp to generate a draft and
>>then reading that in.  That requires creating a temporary directory
>>and running comp there.  That insures that we can read message 1.
>>
>>What can we do to simplify this? The obvious answer would be to add a
>>-build flag to comp as is supported on repl and forw.
>
> It looks like .... what exmh does, at least, is simply invokes the
> command "comp -nowhatnowproc", and then gets the message out of the drafts
> folder by seeing what "cur" is.  Is that not sufficient?

I worry about concurrency problems doing that.  Scripts could be
running that would change cur.

What the -build option does is only slightly better. It causes a file
called <draft-folder>/drafts, or <draft-folder>/reply to be created.
There could be collisions from other programs calling [comp|repl|forw]
-build, but that is less likely than something messing with cur.

Thanks

Jeff

--
Jeffrey C. Honig <address@hidden>
http://www.honig.net/jch
GnuPG ID:14E29E13 <http://www.honig.net/jch/key.shtml>



reply via email to

[Prev in Thread] Current Thread [Next in Thread]