added pd as namespace #12268
added pd as namespace #12268
Conversation
more consistent with code in the rest of the document
Generally, also numpy's randn should used as np.random.randn
| @@ -2801,6 +2801,24 @@ everything in the sub-store and BELOW, so be *careful*. | |||
|
|
|||
| .. _io.hdf5-types: | |||
|
|
|||
| .. warning:: Hierarchical keys cannot be retrieved as dotted (attribute) access as described above for items stored under root node. | |||
There was a problem hiding this comment.
This belongs in a separate PR --- or I would consider renaming this one to make it obvious that this is the big change here.
There was a problem hiding this comment.
Either way, what would you prefer?
There was a problem hiding this comment.
I'd opt for another PR, it just keeps things easier if PRs are smaller and about one thing only.
|
@dacoex yeh, you will need to explain why you want to add a warning like that (in another PR). |
|
@jreback seems that you are not convinced of the usefulness. |
|
I am talking about the keys accessor warning. just take out for now and you can put in another PR. (e.g. just the No I am not convinced its something people should be doing at all. The attribute access is just a convenience feature, string accessing by paths is the preferred method. |
This reverts commit dfac360.
|
OK, now this PR is only about the namespace changes. |
|
can you rebase this |
|
|
||
| .. ipython:: python | ||
|
|
||
| store.foo.bar.bah |
There was a problem hiding this comment.
this sectionneeds to be in a separete PR
There was a problem hiding this comment.
I thought with was already merged. So here you go.
|
pls don't open new pull-requests for the same fix, just force push to the same. see the docs simply
|
more consistent with code in the rest of the document