Re: [PATCH] notmuch-new: Eliminate tallocs whilst construct filenames.

[Carl Worth <cworth@cworth.org>]

[-- Attachment #1: Type: text/plain, Size: 1262 bytes --]

On Fri, 27 Nov 2009 14:17:02 +0000, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> > I *know* I composed a reply to this message earlier, but apparently
> > you're right that it never went out. (*sigh*---if only I had a reliable
> > mail client[*]).
> 
> I hear there's one called sup... ;-)

Heh. But seriously, I hit a lot of crashes with sup, and that invariably
led to *lots* of lost tag changes. I'm willing to live with lots of
Xapian-defect-250 pain right now to avoid that lossage.

> The issue I see with the "error, continue" pattern is that we are in
> danger of not reporting the first error but the last one.

OK. That would be a problem, yes.

> Is notmuch ready for fault-injection yet? Maybe once you have a nasty
> testsuite...

It's not "ready" in the sense that there is going to be a huge series of
fixes that fault-injection will find. But it's definitely "ready" in
the sense that I want to start doing this kind of testing.

But yes, we need a test suite.

Oh, and we'll also need to deal with remaining glib usage inside of
notmuch, (and inside of GMime as well), before we can do good testing
for memory-fault injection. Maybe what we'll end up with is a patch to
de-glib-ify GMime? I'm not sure.

-Carl

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]