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

Re: bug#1894: mutt-1.5.6i: alternates "thinks" a mail is from me which is not



reopen 1894 "Spiro R. Trikaliotis" <trik-news@xxxxxx>
retitle 1894 $index_format %F %f documented behaviour mismatch
thanks


 On Friday, May 28, 2004 at 2:26:01 PM +0200, Spiro R. Trikaliotis wrote:

> I just checked the manual, but I could not find any hint the regex
> should be written with heading ^ and trailing $.

    From manual.txt:

| 4.  Advanced Usage
| 4.1.  Regular Expressions
| All string patterns in Mutt including those in more complex
| ``patterns'' must be specified using regular expressions (regexp) in
| the ``POSIX extended'' syntax (which is more or less the syntax used
| by egrep and GNU awk). For your convenience, we have included below a
| brief description of this syntax.
    [...]
| Note that ``\'' must be quoted if used for a regular expression in an
| initialization command: ``\\''.
    [...]
| The caret ``^'' and the dollar sign ``$'' are metacharacters that
| respectively match the empty string at the beginning and end of a
| line.

    Note that this "line" here means in alternates case: Any address
extracted from given field.


> I do not understand why "%F" in the $index_format uses the From_ or
> the Return-path: instead of the From:.

    Well, because of the same fall down strategy. Mutt searches the
sender, and finds a valid email only in RP. Then it prints corresponding
fullname if available, or the found email if not.

    This strategy works very well in many cases, even very broken.
Granted, here it is fooled by the RP containing a special list bounce
address. But there is no way for Mutt to guess if a given valid RP is
expected true real sender address, or just another junk as "From:".


> If it would use the latter, and that's what I would have thought, than
> it would display the correct string "Ravi", wouldn't it?

    Yes. But here "Ravi" is corresponding to an invalid address. The
name and address pair is rejected.


> I even checked with "%f": Although the manual clearly states:
>| "%f entire From: line (address + real name)"
> mutt does *not* display the From: line, but the Return-path: line.

    The manual is true, but simplified. It doesn't describe here the
fall down strategy and consequences. It doesn't say that what it calls
"entire line" is in fact "name <email>" as reconstructed from infos
extracted from the line, whatever format it had.


> This is surely not the same as the manual tells me, so I believe this
> to be another bug

    On one side, manual should be accurate. On other side it has to be
readable, concise and not losing reader in gory details. After all we're
in a border case. The manual is true with the vast majority of mails
(valid "From:" ==> manual true). And when some more level of details is
needed, there is always the source code.


> what should I do?

    I reopened/retitled #1894 to hear further arguments.


Bye!    Alain.
-- 
Everything about locales on Sven Mascheck's excellent site at new
location <URL:http://www.in-ulm.de/~mascheck/locale/>. The little tester
utility is at <URL:http://www.in-ulm.de/~mascheck/locale/checklocale.c>.