DOC: Add examples to .get_loc methods #17563
Conversation
topper-123
force-pushed
the
get_loc_examples
branch
2 times, most recently
from
944aa9b to
46159be
Compare
Codecov Report
@@ Coverage Diff @@
## master #17563 +/- ##
==========================================
- Coverage 91.25% 91.23% -0.02%
==========================================
Files 163 163
Lines 49617 49617
==========================================
- Hits 45277 45268 -9
- Misses 4340 4349 +9
Continue to review full report at Codecov.
|
Codecov Report
@@ Coverage Diff @@
## master #17563 +/- ##
==========================================
- Coverage 91.25% 91.23% -0.02%
==========================================
Files 163 163
Lines 49617 49617
==========================================
- Hits 45277 45268 -9
- Misses 4340 4349 +9
Continue to review full report at Codecov.
|
Codecov Report
@@ Coverage Diff @@
## master #17563 +/- ##
==========================================
- Coverage 91.19% 91.18% -0.02%
==========================================
Files 163 163
Lines 49626 49626
==========================================
- Hits 45258 45249 -9
- Misses 4368 4377 +9
Continue to review full report at Codecov.
|
jreback
added
Docs
Indexing
pandas/core/indexes/base.py
Outdated
| @Appender(_index_shared_docs['get_loc']) | ||
| Examples | ||
| --------- | ||
| >>> unique_index = pd.%(klass)s(list('abc')) |
There was a problem hiding this comment.
I don't think 'klass' substitution is needed for the example, just show it with base Index, logic of behaviour should be the same for all types (for other types of Index subclasses the 'abc' values might not be valid anyhow, and you get a 'wrong' example)
pandas/core/indexes/interval.py
Outdated
| Parameters | ||
| ---------- | ||
| key : label | ||
| method : {None, 'pad'/'ffill', 'backfill'/'bfill', 'nearest'}, optional |
There was a problem hiding this comment.
I wouldn't tlist the other methods, as you say they are not implemented (like you did in the categorical docstring)
| >>> i3 = pd.Interval(0, 2) | ||
| >>> overlapping_index = pd.IntervalIndex.from_intervals([i2, i3]) | ||
| >>> overlapping_index.get_loc(1.5) | ||
| array([0, 1], dtype=int64) |
There was a problem hiding this comment.
We are discussing to change this behaviour, but the above is still correct on master. However, I would change the example to already add get_loc(Interval(..)) cases (the case we intent to keep working, the scalar numerical value might raise in the future). So that way we can just remove those case that will not be valid anymore, but keep the exact match for actual Interval objects examples.
There was a problem hiding this comment.
Added an interval example
pandas/core/indexes/multi.py
Outdated
| -------- | ||
| Index.get_loc : get_loc method for (single-level) index. | ||
| get_locs : Given a tuple of slices/lists/labels/boolean indexer to a | ||
| level-wise spec, produce an indexer to extract those locations |
There was a problem hiding this comment.
extra level of indentation here
pandas/core/indexes/category.py
Outdated
| @@ -354,7 +354,7 @@ def _to_safe_for_reshape(self): | |||
|
|
|||
| def get_loc(self, key, method=None): | |||
| """ | |||
| Get integer location for requested label | |||
| Get integer location for requested label. | |||
There was a problem hiding this comment.
I actually like the "Get integer location, slice or boolean mask for requested label" from the multi-index docstring, as this seems to be more complete description?
There was a problem hiding this comment.
I agree, Will change
pandas/core/indexes/base.py
Outdated
| >>> monotonic_index = pd.%(klass)s(list('abbc')) | ||
| >>> monotonic_index.get_loc('b') | ||
| slice(1, 3, None) | ||
| >>> non_monotonic_index = pd.%(klass)s(list('abcb')) |
There was a problem hiding this comment.
put a blank line in-between cases
pandas/core/indexes/multi.py
Outdated
| -------- | ||
| Index.get_loc : get_loc method for (single-level) index. | ||
| get_locs : Given a tuple of slices/lists/labels/boolean indexer to a | ||
| level-wise spec, produce an indexer to extract those locations |
There was a problem hiding this comment.
I think you need to indent this (to the :)
topper-123
force-pushed
the
get_loc_examples
branch
from
46159be to
8c9839f
Compare
|
I think you still need to use the |
topper-123
force-pushed
the
get_loc_examples
branch
from
8c9839f to
ed372a2
Compare
|
Hello @topper-123! Thanks for updating the PR. Cheers ! There are no PEP8 issues in this Pull Request. 🍻 Comment last updated on September 20, 2017 at 14:58 Hours UTC |
|
I've made various changes.The @appender is still there, though the examples are in it as well. The EDIT: Are you sure it's ok to force-push changes to the same commit? To me it seems that it makes reviewting more difficult, as the connection between code and comments seems to get lost. |
topper-123
force-pushed
the
get_loc_examples
branch
2 times, most recently
from
0c9094d to
0ab5488
Compare
|
The error is at So the error seems to be unrelated to my pull request... |
if commits are logical / organized and have good messages, yes. however most commits I see should just be combined into 1 and promote even more confusion. rebasing is also much easier with fewer commits. so force pushing is fine as its your own fork. |
|
looks fine. can you rebase and push again to force ci to run. |
For that reason in general I prefer to add commits instead of rebasing/force push, but to clean up messed up github state, it is perfectly fine (as long as it is in a branch). |
topper-123
force-pushed
the
get_loc_examples
branch
from
0ab5488 to
28c9c89
Compare
|
Green @jreback |
|
@topper-123 Thanks! |
git diff upstream/master -u -- "*.py" | flake8 --diffThis pull request adds some examples to
.get_locdoc string ofpd.Index,pd.CategoricalIndex,pd.MultiIndex&pd.IntervalIndex. Also clarifies the return value in some cases.I previously proposed pull request #17380, but the commit history there became a mess. I've deleted that pull request and added this instead.