Hi --

On Thu, 3 Jun 2004, Robert Klemme wrote:

> 
> "David A. Black" <dblack / wobblini.net> schrieb im Newsbeitrag
> news:Pine.LNX.4.44.0406030437540.1134-100000 / wobblini...
> > Hi --
> >
> > On Thu, 3 Jun 2004, Robert Klemme wrote:
> >
> > >
> > > "David Alan Black" <dblack / wobblini.net> schrieb im Newsbeitrag
> > > news:m3vfi98ggu.fsf / wobblini.net...
> > > >
> > > > Note (from David Black): This was a message sent by Botp to
> ruby-talk
> > > > that didn't make it through.  I resent it, with a small change, to
> see
> > > > if it would make it, and it did.
> > > >
> > > > The change was to strip "Re:" and put "about" in the subject
> line....
> > >
> > > I assume this is a stupid question, but the GW doesn't contain a
> header
> > > detection regexp that looks like this:
> > >
> > > /^(.*):\s+(.*)$/
> > >
> > > >> /^(.*):\s+(.*)$/ =~ 'Subject: Re: foo bar'
> > > => 0
> > > >> $1
> > > => "Subject: Re"
> > > >> $2
> > > => "foo bar"
> > >
> > > Really stupid question...
> >
> > Which part is the question? :-) If it's whether that regexp appears in
> > the code, the answer is no.  But I'm not understanding what suggests
> > that it might.
> 
> You said you fixed the message by removing "Re:" from the subject line and
> replaced it by "about".  The wild guess was just that - since "Re:" ends
> with a colon - there might be a regexp somewhere that wrongly identifies
> the header name as "Subject: Re:", which in turn might have had other
> unwanted consequences.
> 
> But I barely dared to ask that because I didn't assume that anyone
> involved in the GW software would use such a regexp...  :-)

Yeah, hopefully not :-) I think what's happening is that the "Re:"
flags the message as a reply, so then the lack of a Reference: header
results in rejection of the message (since it then appears to be a
reply to nothing).  In the case of Botp's message, I saved the
message, manually removed "Re:", and then resent it to the gateway.


David

-- 
David A. Black
dblack / wobblini.net