Francesc Alted

Results 320 comments of Francesc Alted

That's a nice suggestion. Could you send the profile? As always, a PR is very welcome.

Hmm, this does not seem related with numexpr limitations (vm="python"), so this should be qualifyied as bug inside bcolz.

Nope, bee8b1e is not a proper fix. I have access to a POWER8 machine (Big Endian), and tests are not passing. Probably this needs more consideration.

I think most of what you are referring as errors are just warnings. The last two are actual errors, and only seems to happen in the second run. Looks like...

This is interesting. Is it possible to get a dump of the segfault so that we can get some hints on the root of the problem?

@cgohlke Could you please try to add the fix in commit e39dcc1 and see if that fixes the issue? Thanks in advance.

Cool. Fixed the [upstream C-Blosc](https://github.com/Blosc/c-blosc/commit/2856403c1d6ed78714fe086a3a4ca5d0134b5310) too. Thanks for the detailed dump; it helped a lot.

_From [[email protected]](https://code.google.com/u/117706089323896263204/) on December 13, 2009 19:21:34_ Hmm. Looking again, I now see that "and" is not a supported operator. So I guess this bug report now morphs into "I...

_From [[email protected]](https://code.google.com/u/116605672876360875462/) on September 27, 2011 06:18:23_ fixed in r5591638b532f . Numexpr now raises an exception in that case. **Status:** Fixed

_From [[email protected]](https://code.google.com/u/101083734699402587936/) on November 06, 2011 02:16:57_ Unfortunately, PyTables uses `ExpressionNode` comparisons via `__nonzero__` a lot internally (the failures can be seen in, for example, PyTables test unit `test_queries.IndexedTableUsage13`), so...