Hdf5 handler

[wasade]

Added support so we can automatically and implicitly resize hdf5 datasets. This is in support of the pending updates to the split libraries workflow

[wasade]

retest this please

[jairideout]

Travis doesn't listen to retest comments unfortunately. To have this retested, click "Details" and hit the refresh button that's towards the top right of the screen.

[wasade]

ahh, thanks

[wasade]

there isn't a refresh there

[wasade]

@ElBrogrammer, travis is confused over the repo name change. this is a low prio pull right now as we don't yet have any hdf5 code in use though, just trying to clean out PRs

[jairideout]

probably b/c pyqi was moved from the bipy org to biocore

[jairideout]

Found the new link to this PR on Travis. I went ahead and restarted the tests:

https://travis-ci.org/biocore/pyqi/builds/15857676

[wasade]

Thanks. One failure, which isn't surprising as HDF5 should be considered an
optional dependency here. Do you know how to get nosetests to not execute
tests if an optional dependency is missing?

On Wed, Mar 19, 2014 at 11:34 AM, Jai Ram Rideout
notifications@github.comwrote:

Found the new link to this PR on Travis. I went ahead and restarted the
tests:

https://travis-ci.org/biocore/pyqi/builds/15857676

Reply to this email directly or view it on GitHubhttps://github.com//pull/244#issuecomment-38082313
.

[jairideout]

Not sure for nose, but you might add conditional unit tests that are skipped if h5py isn't present:

http://docs.python.org/3.1/library/unittest.html#skipping-tests-and-expected-failures

[wasade]

this PR got really weird conflicts that ive never seen before. right now, through mechanisms i don't understand, this PR is now blocked by py33

[jairideout]

Want to close this one and reissue?

[wasade]

suspect it'll be fine once py33 is merged

we really need to avoid long outstanding PRs, though i know it is difficult
to get some of these merged at times and to get effort on them

On Wed, Mar 19, 2014 at 12:22 PM, Jai Ram Rideout
notifications@github.comwrote:

Want to close this one and reissue?

Reply to this email directly or view it on GitHubhttps://github.com//pull/244#issuecomment-38088056
.