GNATS' handling of long mail subject lines

GNATS' handling of long mail subject lines

[Yngve Svendsen]

The MIME script I submitted has one very positive side-effect: It works 
around GNATS' inability to handle long mail subject headers that are split 
over several lines.

The problem is as follows: The RFCs describing e-mail and e-mail headers 
(822 and 2047) allow for the possibility of splitting long headers across 
several lines. The splitting is done by adding CRLF and then a space on the 
next line, before the header itself continues.

A long "Subject:" header might thus look like this:

Subject: This is a PR with an extremely long subject line.
  It keeps going on and on and on and on and on and on and
  on and on and on and on and on and on and on and on and
  on and on and on and on and on and on and on and on all
  the way until it stops

The MIME script I wrote removes the extra CRLF and space, placing the 
entire header text on a single line. GNATS seems perfectly happy with that.

If this kind of header is fed directly into GNATS, as in a standard 
plain-vanilla installation, all kinds of strangeness results. For instance, 
the index is messed up, so that queries return garbage. See 
http://sources.redhat.com/cgi-bin/gnatsweb.pl?cmd=view&pr=161&database=gnats 
for more details.

The upshot of this is that mail parsing in GNATS itself needs to be 
improved. This shouldn't be too much of a challenge for someone with a bit 
of experience with C (I am, unfortunately NOT one of those). It should be 
done independently of implementation of MIME handling and other 
internationalization.

Anyone feel up to the challenge? I may be able to provide a little bit of 
further help if someone wants to take it on. In the meantime, the MIME 
script provides a temporary workaround.

Yngve Svendsen
IS Engineer
Clustra AS, Trondheim, Norway
yngve.svendsen@clustra.com

Re: GNATS' handling of long mail subject lines

[Milan Zamazal]

>>>>> "YS" == Yngve Svendsen <yngve.svendsen@clustra.com> writes:

    YS> The MIME script I submitted has one very positive side-effect:
    YS> It works around GNATS' inability to handle long mail subject
    YS> headers that are split over several lines.

    YS> The problem is as follows: The RFCs describing e-mail and e-mail
    YS> headers (822 and 2047) allow for the possibility of splitting
    YS> long headers across several lines. The splitting is done by
    YS> adding CRLF and then a space on the next line, before the header
    YS> itself continues.

[...]

    YS> If this kind of header is fed directly into GNATS, as in a
    YS> standard plain-vanilla installation, all kinds of strangeness
    YS> results. For instance, the index is messed up, so that queries
    YS> return garbage.

I can't reproduce the problem and I don't know what the Subject line
could have common with index.  Could you please describe the exact way
how to reproduce the problem and what exactly becomes wrong?

Thanks,

Milan Zamazal

-- 
Free software is about freedom, not about free beer.  If you care only about
the latter, you'll end up with no freedom and no free beer.