Issue 2 solved?

[Yoni Rabkin]

I tested issue 2 [http://code.google.com/p/montezuma/issues/detail?
id=2] against montezuma-0.1.1 and montezuma-0.1.3b with a corpus of
2001 files. I couldn't reproduce it under 0.1.3b but could easily do
so with 0.1.1. Has this been officially solved? If so, let's mark the
issue as resolved.

[Leslie P. Polzer]

I didn't do anything specifical to solve that issue but it might
be a collateral to the UTF8 patch.

It would be nice to have this case added to the test suite.

After that I suppose we could mark it as resolved.

Leslie

[Yoni Rabkin]

> It would be nice to have this case added to the test suite.
>
> After that I suppose we could mark it as resolved.

Since I don't think that this happens with memory-based indexes it
would mean generating a 2000 doc test index on disk, indexing,
searching, and finally removing the test dir.

I've added writing the test to my montezuma TODO.

[Yoni Rabkin]

How about this: http://montezuma-dev.googlegroups.com/web/m2k-test.patch?gsc=E6BgegsAAADN1ToO-cI8SH_di_PIvCIF

I "back-ported" the patch to 0.1.1 and it dutifully failed. 0.1.3b
passes of course.

[Leslie P. Polzer]

Looks good, thank you very much for your efforts!

In turn I will apply it and release 0.1.3 this week. :)

Leslie