Spurious line inserted if mm-uu-hide-markers is nil

[Ralf Angeli <angeli@caeruleus.net>]

Hi,

if you set `mm-uu-hide-markers' to nil as to not lose stuff like
verbatim marks when replying to a message, then a spurious empty line is
inserted in front of a region marked with `message-mark-inserted-region'
such as this:

--8<---------------cut here---------------start------------->8---
foo
--8<---------------cut here---------------end--------------->8---

The empty line is fontified with the face `mm-uu-extract'.

The issue goes away if you apply the following patch:

--- lisp/gnus/mm-uu.el	2010-09-02 01:42:32 +0000
+++ lisp/gnus/mm-uu.el	2010-11-20 20:36:07 +0000
@@ -165,7 +165,7 @@
      ;; dependency on `message.el'.
      "^-+[8<>]*-\\{9,\\}[a-z ]+-\\{9,\\}[a-z ]+-\\{9,\\}[8<>]*-+$"
      "^-+[8<>]*-\\{9,\\}[a-z ]+-\\{9,\\}[a-z ]+-\\{9,\\}[8<>]*-+$"
-     (lambda () (mm-uu-verbatim-marks-extract -1 0 1 -1))
+     (lambda () (mm-uu-verbatim-marks-extract 0 0 1 -1))
      nil)
     ;; Omitting [a-z8<] leads to false positives (bogus signature separators
     ;; and mailing list banners).


Does anybody know if the -1 for the `start-offset' parameter of
`mm-uu-verbatim-marks-extract' has been used on purpose, i.e. would the
change have any negative effects?  If not and if there are no
objections, I'd apply it in Emacs' Bazaar repository.

-- 
Ralf