Re: pop3-read-response

[Lars Magne Ingebrigtsen <larsi@gnus.org>]

Daniel Dehennin <daniel.dehennin@baby-gnu.org> writes:

> Sorry, I’m using bc1350c0871a8da4d356861b06020e37692056ef and the
> problem is still there.

Darn.

> If you need more information, I can make some tests.

My suspicion was that there was data in the buffer that the pop3 parser
didn't expect to be there, so it regarded something as a response before
getting the complete

+OK USER ... \r\n

line.  But that may be wrong.

If you apply the following patch, pop3 should message what's in the
buffer, and that might give us a clue what's failing precisely,

pop3-read-point is supposed to be before the "+OK", so waiting for the
\r\n should have given the complete line.  I'm thinking pop3-read-point
is somewhere earlier in the buffer for some reason or other.

diff --git a/lisp/pop3.el b/lisp/pop3.el
index 90e11b3..917a401 100644
--- a/lisp/pop3.el
+++ b/lisp/pop3.el
@@ -345,6 +345,7 @@ Return the response string if optional second argument is non-nil."
 		  (not (search-forward "\r\n" nil t)))
 	(pop3-accept-process-output process)
 	(goto-char pop3-read-point))
+      (message "%d %s" pop3-read-point (buffer-string))
       (setq match-end (point))
       (goto-char pop3-read-point)
       (if (looking-at "-ERR")

-- 
(domestic pets only, the antidote for overdose, milk.)
  bloggy blog http://lars.ingebrigtsen.no/