Selecting columns from a HDFStore with a MultiIndex fails #3748
Comments
|
this was a bug thanks for the report, should be in upcoming 0.11.1, was not handling this case as a workaround, you can do:
|
|
also...what version of |
|
Master from last Friday, on top of an Anaconda Python 3.3 environment |
|
what happens when you do:
|
|
Nothing pretty... Probably should have checked before, but I never was so sure about how to run Pandas' test suite. So that's something I learned now! (Output below truncated due to Github limit) |
|
ok....just found this out myself! (as pytables JUST (like yesterday) came out officially for py3k; problem is that everything is stored as bytes, |
|
I have opened an issue for this #3750, and am working to get it in the 0.11.1 release. The version you have 'works' as long as you don't use a |
|
if you are adventurous, pls give this PR a shot: #3762 |
|
@hmgaudecker I merged into master....all py3k should work now.....(and numpy >=1.6.1) pls give me a try and let me know...thanks |
|
Works for me, thanks a lot!!! |
|
great! do let me know if you find anything...this is pretty new also....didn't REALLY test actually saving a hdf in py3 and then reading in py2, I think it should work |
|
I'm duplicating this on Gholke's Pytables downloaded today (8/10/13) , and producing the following error: Array objects cannot currently deal with void, unicode or object arrays. This thread reads like this was to have been fixed by now and I'm wondering if any of you could point me at binary Pytables install set for Windows 7 64-bit Python 3.3. Thx. |
|
u r using pandas 0.12? show me the code u r using as well |
|
pandas 0.11 -- Is pandas 0.12 required now? Here's the code : import pandas as pd This is ex1.csv: a b c d message This is from Python for Data Analysis, page 141 on my Nook for PC version, chapter 6, Binary Data Formats, Using HDF5 Format. I had this snippet tested Saturday morning and then it not only stopped, but my eclipse stopped finding builtins like print() and could not bring up the debugger to duplicate the error message. So, I can't duplicate this problem right now but I thought I would pass this along for your consideration. Thanks for your interest. |
|
yes 0.12 is required if on py 3k with pytables 3.0.0 if not on py3k then < 0.12 is ok |
|
I am going to put a note in the docs about this |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Using the example from the docs here: http://pandas.pydata.org/pandas-docs/dev/io.html#storing-multi-index-dataframes but specifying the 'columns' in
[1219]keyword when selecting:leads to the failure below. The original example works, as does selecting columns with a standard index. I'm on a fairly recent master (~1 week) and Python 3.3. Hope this is not expected and I did not miss an issue, I did search for a bit.
The text was updated successfully, but these errors were encountered: