From: Greg A. Woods (woodsweird.com)
Date: Mon May 03 2004 - 18:59:47 CDT

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

[ On Sunday, May 2, 2004 at 16:27:19 (-0500), Noel Jones wrote: ]
> Subject: Re: Help with header_checks
>
> The way header_checks (and mime_header_checks) are used, any header
> may span multiple lines.

Of course.

> To have /m turned off by default, *as used
> everywhere else* would require that EVERY (mime_)header_check rule
> have the /m flag added to it.

_IFF_ that's what the RE author intends, of course.

Consider the possibility that the user only wants to match the first
line of a multi-line header....

> This would undoubtedly cause far more
> problems for the masses than the few cases where someone who knows
> regexp and doesn't read the postfix docs does what they think is the
> right thing.

It is undoubtable that you are mistaken.

If folks are not reading the Postfix docs then they're going to read, or
have read, the PCRE docs (or, e.g., the "Mastering Regular Expressions"
book, Perl or Python docs, etc., etc., etc.) and thus they're going to
assume that Postfix follows the otherwise universally accepted de facto
standards and that the "/m" flag will turn _on_ PCRE_MULTILINE.

Indeed Postfix treatment of PCRE flags contractdicts the PCRE internal
flags (e.g. "(?m)"), causing yet further confusion.

> While it may not be consistent with similar flags elsewhere,
> it is unquestionably the right thing in postfix.

Inconsistency of otherwise universally accepted de facto standards and
documented features is _never_ a good thing in this sort of scenario.

--
                                                Greg A. Woods

+1 416 218-0098 VE3TCP RoboHack <woodsrobohack.ca>
Planix, Inc. <woodsplanix.com> Secrets of the Weird <woodsweird.com>

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]