<<< Date Index >>>     <<< Thread Index >>>

Re: send-hook not setting $from



On Sat, Dec 06, 2003 at 07:14:20PM -0600, Eugene Lee wrote:
> On Fri, Dec 05, 2003 at 07:28:11AM -0500, David Yitzchak Cohen wrote:
> : On Thu, Dec 04, 2003 at 07:07:29PM -0600, Eugene Lee wrote:
> : > On Thu, Dec 04, 2003 at 03:50:48PM +0100, Michael Tatge wrote:
> : > : * On Thu, Dec 04, 2003 Eugene Lee (list-mutt@xxxxxxxx) muttered:

> : > : >
> : > : >       send-hook . 'set from=me@xxxxxxxxxx'
> : > : >       send-hook '~C abuse@xxxxxxx' 'set from=me+aol@xxxxxxxxxx'
> : > : 
> : > :                                       use my_hdr
> : > : 
> : > :         set from = me@xxxxxxxxxx
> : > :         send-hook . 'unmy_hdr from'
> : > :         send-hook '~C abuse@xxxxxxx' 'my_hdr From: me+aol@xxxxxxxxxx'
> : > 
> : > Michael (and Noir), thanks for your comments.  I know that my_hdr works.
> : > But I don't understand why "set $from" does not work in a send-hook.
> : 
> : In plain English, changing $from in a send-hook has no effect on the
> : message being composed because Mutt has already set your From: header
> : before it executes the send-hook (based on what $from was then - i.e.,
> : before your hook was called).
> 
> This behavior is not explained in the normal HTML documentation.

You must be right.  I can't find it now :-(

> : my_hdr works because it simply replaces
> : any existing header by the same name - in this case, the "From: " header.
> : 
> : > The docs mention nothing.
> : 
> : The docs specifically mention that changing $from has no effect on the
> : current message from within send-hooks.
> 
> I can't find anything in the docs that support this statement, even
> though your statement is true.  :-(  In fact, the docs only have a
> statement about using my_hdr:
> 
>       http://www.mutt.org/doc/manual/manual-3.html#send-hook
> 
>       Also note that my_hdr commands which modify recipient headers,
>       or the message's subject, don't have any effect on the current
>       message when executed from a send-hook.

Yup, that's all I was able to find now trying to look through the docs.
That is really annoying. . .

> : > The problem is, since setting $from is unreliable while "my_hdr From:"
> : > always works, what's the point of keeping this configuration variable?
> : > Get rid of what's broken and keep what works.
> : 
> : There, you have a feature request.  I'd rather keep both, but fix
> : evaluation of send-hooks so they happen before the "From: " header is
> : constructed.  The only problem is that doing so would prevent us from
> : being able to match on the "From: " header in a send-hook (since the
> : header wouldn't exist yet).  The "correct" solution is difficult to work
> : out, IMHO.  There are too many overlapping feature requests, and many of
> : them are incompatible by design.
> 
> I can see how it gets complicated.  Without looking into the bowels of
> the source, I can't even imagine what the "correct" solution should be.
> But I do know that setting $from in a send-hook seems to be a recurring
> question.  I think it would be very useful if the official docs would 
> specifically mention this caveat.

...and if they already mention it somewhere (which I _thought_ they did),
could they please make that somewhere a tad easier to find???

> So what's the best way to request and correct a documentation bug?

the flea program, along with a patch in the comments section, or attached

If you mail me the text you want to add, I'll make a diff against the
CVS docs and submit the bug report, and run after the devs to get it
applied ASAP.  I was under the impression that this little tidbit was
already in there.

> : > Just a bit confused since the docs (short of parsing the source code)
> : > doesn't match behavior.  Maybe I should file a bug report?
> : 
> : You can certainly file a wishlist, but since the docs _do_ predict this
> : behavior, it's not technically a "bug" as such.  (It's simply a documented
> : "feature" - and a rather annoying one, at that.)
> 
> I must be blind, because I cannot find anything in the docs that even
> predicts this behavior.  Could you possible provide a link to the blurb
> in the docs that says so?

Sadly, I must be blind too :-(

Sorry,
 - Dave

-- 
Uncle Cosmo, why do they call this a word processor?
It's simple, Skyler.  You've seen what food processors do to food, right?

Please visit this link:
http://rotter.net/israel

Attachment: pgpDkulRvRKEr.pgp
Description: PGP signature