another 2.5.1pre keywords-related coredump

Nix nix at esperi.org.uk
Fri Oct 9 20:48:59 UTC 2009


On 9 Oct 2009, julien at trigofacile.com uttered the following:
> Incidentally, I do not understand why we have if (DO_KEYWORDS) and
> not #if DO_KEYWORDS; we have lots of useless boolean checks (if 0) for
> every header of every article...  Is there a reason not to use
> #if DO_KEYWORDS compilation checks?

It's uglier, and with any even halfway decent compiler, if (0 && ...)
gets its entire body elided at compile time. (Even GCC 2.7.2 could do
that at -O2: GCC 4+ often does it at -O0!)

>> Thankfully the crash is not caused by some complex cross-article arena
>> corruption but is reproducibly caused by the attached article (still in
>> wire format, as an application/octet-stream accordingly):
>
> As you may have found, the problem comes from the fact that your test
> article contains a Keywords: header :)

That's what I thought. They're not at all common...



More information about the inn-workers mailing list