ENH: Read entire group of an HDF5 file in a single call

[VGonPa]

See this question in SO.

I guess that something like store.get_group('df') might do the job:

>>> import pandas as pd
>>> store = pd.HDFStore('test.h5',mode='w')
>>> store.append('df/foo1',DataFrame(np.random.randn(10,2)))
>>> store.append('df/foo2',DataFrame(np.random.randn(10,2)))
>>> store.get_group('df')
...           0         1
... 0 -0.495847 -1.449251
... 1 -0.494721  1.572560
... 2  1.219985  0.280878
... 3 -0.419651  1.975562
... 4 -0.489689 -2.712342
... 5 -0.022466 -0.238129
... 6 -1.195269 -0.028390
... 7 -0.192648  1.220730
... 8  1.331892  0.950508
... 9 -0.790354 -0.743006
... 0 -0.761820  0.847983
... 1 -0.126829  1.304889
... 2  0.667949 -1.481652
... 3  0.030162 -0.111911
... 4 -0.433762 -0.596412
... 5 -1.110968  0.411241
... 6 -0.428930  0.086527
... 7 -0.866701 -1.286884
... 8 -0.649420  0.227999
... 9 -0.100669 -0.205232
... 
... [20 rows x 2 columns]

I never did a pull request before, nor tried to dive into pandas code, but I guess that a possible implementation (inspired in this answer) would be something like:

class HDFStore():
    ...
    def get_group(self, group_name):
        return pd.concat([ self.select(node._v_pathname) for node in self.get_node(group_name) ])
    ...

[cpcloud]

I think this would be better if it returned this:

{'df': {'foo1': store.select('foo1'), 'foo2': store.select('foo2')}}

or

{'foo1': store.select('foo1'), 'foo2': store.select('foo2')}

Users can do the concat afterwards if they want to. This shouldn't assume that the frames are the same shape, dtype etc. For example, I have different metrics related to one "thing" all under one key, like

eeg/blocked
eeg/filtered
eye/blocked
eye/samples
eye/events

I don't want to concatenate the result but I often want to get just the eye or just the eeg datasets. This could even be a context manager where the context is the node. Something like:

with store.get_group('df') as group:
    foo1 = group['foo1']

[jreback]

the issue I have with this is that its tricky to pass a where to this as you can't assume anything about the tables. I like the returning a dict though.

[VGonPa]

I like the idea of returning a dict as well.

What about something like this?:

def get_group(self, group_name):
    return {node._v_name: self.select(node._v_pathname) for node in store.get_node(group_name) }