CLN: @doc - base.py & indexing.py #31970

HH-MWB · 2020-02-14T02:20:29Z

working on CLN: Update docstring decorator from Appender and Substitution to doc #31942
tests added / passed
passes black pandas
passes git diff upstream/master -u -- "*.py" | flake8 --diff
whatsnew entry

…e-doc-decorator

HH-MWB · 2020-02-14T12:36:36Z

Replacing all docstring decorator at once will be too hard to accomplish. I would prefer to do it step-by-step. Clean a few / one file in a PR, and do things iteratively.

Would you mind help me review the current changes? At the mean time, I will keep working on next part.

WillAyd · 2020-02-14T16:56:00Z

@HH-MWB can you post screenshots of the affected docstrings?

simonjayhawkins

Thanks @HH-MWB generally lgtm, however a few questions.

pandas/core/indexing.py

pandas/core/indexes/datetimelike.py

simonjayhawkins · 2020-02-14T19:56:10Z

pandas/core/arrays/categorical.py

@@ -1352,8 +1353,7 @@ def memory_usage(self, deep=False):
        """
        return self._codes.nbytes + self.dtype.categories.memory_usage(deep=deep)

-    @Substitution(klass="Categorical")
-    @Appender(_shared_docs["searchsorted"])
+    @doc(IndexOpsMixin.searchsorted, klass="Categorical")


should this be inheriting from core.arrays.base.ExtensionArray instead? Also I can't see this docstring in the published docs.

I agree with this comment here - this seems strange to use IndexOpsMixin since it isn't part of this classes hierarchy. Any reason not to address this?

…e-doc-decorator

HH-MWB · 2020-02-15T02:06:48Z

@HH-MWB can you post screenshots of the affected docstrings?

@WillAyd No problem. Should I use help() or .__doc__?

WillAyd · 2020-02-15T02:18:34Z

You can build the doc strings and just post a screenshot of the HTML

HH-MWB · 2020-02-15T02:29:13Z

You can build the doc strings and just post a screenshot of the HTML

Get! Will do, thanks!

…e-doc-decorator

simonjayhawkins

thanks @HH-MWB lgtm. would rather #31970 (comment) was addressed here but could be a follow-up.

would it be feasible to inherit the docstring template from the super class instead of explicitly specifying it? It would address my concern for Categorical array.

HH-MWB · 2020-02-15T19:09:50Z

thanks @HH-MWB lgtm. would rather #31970 (comment) was addressed here but could be a follow-up.

would it be feasible to inherit the docstring template from the super class instead of explicitly specifying it? It would address my concern for Categorical array.

@simonjayhawkins Sorry for the late reply.

First, I agree with you that inherit the docstring from the super class is very intuitive, and also could saves us a lot of time. Then, when I dive deeper for this case, I found that the original docstring is different with it's super class. This difference confused me and I am trying to figure it out what cause this difference and how can we solve it.

Inherit Docstring

I think it is a very good idea to inherit the docstring. Maybe this should not be a default behavior but at least we should be able to do inheritance easily.

A potential quick solution for this is to create another decorator, for example, doc_inhert. I would expect it to bring the first non-empty docstring for it's super class.

For example, B.f should have the docstring inhert from A.f:

class A:
    def f(self):
        """func doc"""

class B(A):
    @doc_inhert
    def f(self):
        pass

Another example, C.f should have the docstring inhert from A.f, because B.f is empty:

class A:
    def f(self):
        """func doc"""

class B(A):
    def f(self):
        pass

class C(B):
    @doc_inhert
    def f(self):
        pass

I am still looking to see if there are other better solutions. For example, one thing in my mind will be create decorator for the class, and do docstring inhert for all methods belong to that class.

Would you mind if I create a new issue about docstring inheritance and propose my idea there? I am not 100% sure, but I think we already have some issues about this.

Open for suggestions and advices. What should I do next?

Case Here

As what I mentioned above, the original docstring is different from it's super class's docstring. Should I keep it as the original one or switch to be same as the super class one?

Here is the original docstring:

Find indices where elements should be inserted to maintain order.

Find the indices into a sorted {klass} `self` such that, if the
corresponding elements in `value` were inserted before the indices,
the order of `self` would be preserved.

.. note::

    The {klass} *must* be monotonically sorted, otherwise
    wrong locations will likely be returned. Pandas does *not*
    check this for you.

Parameters
----------
value : array_like
    Values to insert into `self`.
side : {{'left', 'right'}}, optional
    If 'left', the index of the first suitable location found is given.
    If 'right', return the last such index.  If there is no suitable
    index, return either 0 or N (where N is the length of `self`).
sorter : 1-D array_like, optional
    Optional array of integer indices that sort `self` into ascending
    order. They are typically the result of ``np.argsort``.

Returns
-------
int or array of int
    A scalar or array of insertion points with the
    same shape as `value`.

    .. versionchanged:: 0.24.0
        If `value` is a scalar, an int is now always returned.
        Previously, scalar inputs returned an 1-item array for
        :class:`Series` and :class:`Categorical`.

See Also
--------
sort_values
numpy.searchsorted

Notes
-----
Binary search is used to find the required insertion points.

Examples
--------
>>> x = pd.Series([1, 2, 3])
>>> x
0    1
1    2
2    3
dtype: int64

>>> x.searchsorted(4)
3

>>> x.searchsorted([0, 4])
array([0, 3])

>>> x.searchsorted([1, 3], side='left')
array([0, 2])

>>> x.searchsorted([1, 3], side='right')
array([1, 3])

>>> x = pd.Categorical(['apple', 'bread', 'bread',
                        'cheese', 'milk'], ordered=True)
[apple, bread, bread, cheese, milk]
Categories (4, object): [apple < bread < cheese < milk]

>>> x.searchsorted('bread')
1

>>> x.searchsorted(['bread'], side='right')
array([3])

If the values are not monotonically sorted, wrong locations
may be returned:

>>> x = pd.Series([2, 1, 3])
>>> x.searchsorted(1)
0  # wrong result, correct would be 1

Here is the docstring from it's super class:

Find indices where elements should be inserted to maintain order.

.. versionadded:: 0.24.0

Find the indices into a sorted array `self` (a) such that, if the
corresponding elements in `value` were inserted before the indices,
the order of `self` would be preserved.

Assuming that `self` is sorted:

======  ================================
`side`  returned index `i` satisfies
======  ================================
left    ``self[i-1] < value <= self[i]``
right   ``self[i-1] <= value < self[i]``
======  ================================

Parameters
----------
value : array_like
    Values to insert into `self`.
side : {'left', 'right'}, optional
    If 'left', the index of the first suitable location found is given.
    If 'right', return the last such index.  If there is no suitable
    index, return either 0 or N (where N is the length of `self`).
sorter : 1-D array_like, optional
    Optional array of integer indices that sort array a into ascending
    order. They are typically the result of argsort.

Returns
-------
array of ints
    Array of insertion points with the same shape as `value`.

See Also
--------
numpy.searchsorted : Similar method from NumPy.

HH-MWB · 2020-02-15T20:54:12Z

You can build the doc strings and just post a screenshot of the HTML

Hi @WillAyd, I might mis-understand what you said.

I create HTML via python make.py html, but because all files changed is under pandas.core and pandas.util, the html doc might not cover those things.

The pandas.core, pandas.compat, and pandas.util top-level modules are PRIVATE. Stable functionality in such modules is not guaranteed.

simonjayhawkins · 2020-02-15T21:09:37Z

Would you mind if I create a new issue about docstring inheritance and propose my idea there? I am not 100% sure, but I think we already have some issues about this.

Open for suggestions and advices. What should I do next?

Thanks for the detail. It looks like we should definitely have a separate issue/PR for this discussion.

simonjayhawkins · 2020-02-15T21:13:39Z

I create HTML via python make.py html, but because all files changed is under pandas.core and pandas.util, the html doc might not cover those things.

am I correct in thinking the only published docstring affected by this PR is https://dev.pandas.io/docs/reference/api/pandas.Series.searchsorted.html? If so the screenshot of that should be OK.

HH-MWB · 2020-02-16T07:45:44Z

am I correct in thinking the only published docstring affected by this PR is https://dev.pandas.io/docs/reference/api/pandas.Series.searchsorted.html? If so the screenshot of that should be OK.

Thanks for your help @simonjayhawkins! Here is the screen shots.

simonjayhawkins · 2020-02-16T10:47:45Z

The screenshots are from locally generated documentation?

pep8speaks · 2020-02-29T23:10:21Z

Hello @HH-MWB! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2020-03-11 23:21:34 UTC

simonjayhawkins · 2020-03-02T11:33:27Z

pandas/util/_decorators.py

+                arg.format(**kwargs)
+                if isinstance(arg, str)
+                else dedent(arg.__doc__)  # type: ignore
+                for arg in wrapper._doc_args  # type: ignore


can you confirm that these # type: ignore stem from the fact that we are adding attributes to the wrapper function?

I guess this was discussed originally, but would using a class for the doc decorator help?

Good point here! Most # type: ignore is because we are adding attributes to function.

The only one not for this reason is else dedent(arg.__doc__) # type: ignore (line 288). This one is because theoretically, arg.__doc__ can be str or None. However, we have checked arg.__doc__ is not None before putting it in, so that we are safe here.

Back to the most # type: ignore cases. I strongly agree with you that we could consider solving this issue without using # type: ignore, because this looks like an abuse. I would like to try putting it under a class. Thanks for this advice!

The only one not for this reason is else dedent(arg.__doc__) # type: ignore (line 288). This one is because theoretically, arg.__doc__ can be str or None. However, we have checked arg.__doc__ is not None before putting it in, so that we are safe here.

i think ok to remove

$ mypy pandas --warn-unused-ignores pandas\util\_decorators.py:288: error: unused 'type: ignore' comment

@simonjayhawkins Good catch here! Thanks. Fixing is here.

HH-MWB · 2020-03-03T01:30:01Z

Hello @simonjayhawkins. It seems very hard to remove all # type: ignore because we might have to bind a variable to the function instead of the decorator.

For use case of @doc(f, ...), we should be able to get original templates from function f. If there are any other way might work, please let me know, and I will be more than happy to give it a try.

However, we can reduce using # type: ignore from 6 times to 3 times.

Here is my proposal by using class:

class doc:
    """
    A decorator take docstring templates, concatenate them and perform string
    substitution on it.

    This decorator will add a variable "_docs" to the wrapped function to keep
    track the original docstring template for potential usage. If it should be
    consider as a template, it will be saved as a string. Otherwise, it will be
    saved as callable, and later user __doc__ and dedent to get docstring.

    Parameters
    ----------
    *args : str or callable
        The string / docstring / docstring template to be appended in order
        after default docstring under function.
    **kwags : str
        The string which would be used to format docstring template.
    """

    def __init__(self, *args: Union[str, Callable], **kwargs: str) -> None:
        self._docs: List[Union[str, Callable]] = []
        for arg in args:
            if hasattr(arg, "_docs"):
                self._docs.extend(arg._docs)  # type: ignore
            elif isinstance(arg, str) or arg.__doc__:
                self._docs.append(arg)

        self.parameters = kwargs

    def __call__(self, func: F) -> F:
        if func.__doc__:
            self._docs.insert(0, dedent(func.__doc__))

        func.__doc__ = "".join(
            [
                arg.format(**self.parameters)
                if isinstance(arg, str)
                else dedent(arg.__doc__)  # type: ignore
                for arg in self._docs
            ]
        )

        func._docs = self._docs  # type: ignore

        return func

We can use same idea to reduce # type: ignore from using function:

def doc(*args: Union[str, Callable], **kwargs: str) -> Callable[[F], F]:
    """
    A decorator take docstring templates, concatenate them and perform string
    substitution on it.

    This decorator will add a variable "_docs" to the wrapped function to keep
    track the original docstring template for potential usage. If it should be
    consider as a template, it will be saved as a string. Otherwise, it will be
    saved as callable, and later user __doc__ and dedent to get docstring.

    Parameters
    ----------
    *args : str or callable
        The string / docstring / docstring template to be appended in order
        after default docstring under function.
    **kwags : str
        The string which would be used to format docstring template.
    """

    def decorator(func: F) -> F:
        @wraps(func)
        def wrapper(*args, **kwargs) -> Callable:
            return func(*args, **kwargs)

        # collecting docstring and docstring templates
        docs: List[Union[str, Callable]] = []
        if func.__doc__:
            docs.append(dedent(func.__doc__))

        for arg in args:
            if hasattr(arg, "_docs"):
                docs.extend(arg._docs)  # type: ignore
            elif isinstance(arg, str) or arg.__doc__:
                docs.append(arg)

        # formatting templates and concatenating docstring
        wrapper.__doc__ = "".join(
            [
                arg.format(**kwargs)
                if isinstance(arg, str)
                else dedent(arg.__doc__)  # type: ignore
                for arg in docs
            ]
        )

        # saving docstring templates list
        wrapper._docs = docs  # type: ignore

        return cast(F, wrapper)

    return decorator

Open for feed backs here. Which one is better? Any other improvement suggestions please?

simonjayhawkins · 2020-03-03T14:51:18Z

Open for feed backs here. Which one is better? Any other improvement suggestions please?

probably best to leave as a follow-up and have more detailed discussion and just keep the changes to the decorator in this PR limited to the changes needed for using a docstring from a method that is not decorated.

HH-MWB · 2020-03-04T03:54:19Z

Open for feed backs here. Which one is better? Any other improvement suggestions please?

probably best to leave as a follow-up and have more detailed discussion and just keep the changes to the decorator in this PR limited to the changes needed for using a docstring from a method that is not decorated.

Sounds very reasonable. I will follow-up latter.

datapythonista · 2020-03-07T20:36:22Z

pandas/util/_decorators.py

@@ -268,17 +269,24 @@ def decorator(func: F) -> F:
        def wrapper(*args, **kwargs) -> Callable:
            return func(*args, **kwargs)

-        templates = [func.__doc__ if func.__doc__ else ""]
+        # collecting and docstring templates
+        wrapper._doc_args: List[Union[str, Callable]] = []  # type: ignore


I think the name _doc_args is misleading. In the context of this function makes sense, since it's the value of the args parameters of the doc decorator, but if I see a _doc_args attribute of for example pandas.Series.head, I don't think it really tells what's in it. Something like _docstring_components could be clearer?

Also, when can it contain a callable? Even when we have something like @doc(pandas.Series.head) we're extending it with the content templates of pandas.Series.head, not with the function.

Last thing, what about using a docstring_components variable to avoid all the type: ignore, and just set wrapper._docstring_components = _docstring_components at the end? Probably cleaner.

@datapythonista Thanks for this suggestion, especially for the new variable name. I strongly agree with you here.

datapythonista

There is a trailing space and a mypy error, other than that looks good.

datapythonista · 2020-03-08T00:00:38Z

pandas/util/_decorators.py

-    This decorator is robust even if func.__doc__ is None. This decorator will
-    add a variable "_docstr_template" to the wrapped function to save original
-    docstring template for potential usage.
+    This decorator will add a variable "_docstring_components" to the wrapped 


There is a trailing whitespace in this line breaking the CI.

HH-MWB · 2020-03-08T01:36:01Z

There is a trailing space and a mypy error, other than that looks good.

@datapythonista, @simonjayhawkins. Sorry to bother you. This mypy error makes me very confused.

In this commit, I removed type: ignore but everything works fine. However, now it shows the type error. I am confused why the original code can pass without error but the new code can't.

I know we can easily fix this by adding type: ignore, but I would like to know what case this problem and if there are a better way to solve that.

datapythonista · 2020-03-08T01:48:03Z

pandas/util/_decorators.py

+        # formatting templates and concatenating docstring
+        wrapper.__doc__ = "".join(
+            [
+                arg.format(**kwargs) if isinstance(arg, str) else dedent(arg.__doc__)


The decorator can be used like this:

@doc def foo(): pass @doc(foo) def bar(): pass

If that happens, in this line arg will be the function foo, and arg.__doc__ will be None. When dedent is called with None as parameter we get:

>>> dedent(None) Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/home/mgarcia/miniconda3/lib/python3.7/textwrap.py", line 430, in dedent text = _whitespace_only_re.sub('', text) TypeError: expected string or bytes-like object

And mypy is warning you about it. There are different ways of fixing this. May be this?

Suggested change

arg.format(**kwargs) if isinstance(arg, str) else dedent(arg.__doc__)

arg.format(**kwargs) if isinstance(arg, str) else dedent(arg.__doc__ or "")

Thanks! That make sense.

In line 281 and 282, there are codes to make sure arg.__doc__ won't be None.

elif isinstance(arg, str) or arg.__doc__: docstring_components.append(arg)

In this case, maybe we can use # type: ignore?

I think that code ensures _docstring_components is not set to None (in the parent). And the exception won't happen as far as _docstring_components is always set with the @doc decorator. But technically speaking there is no guarantee that something like this can be done:

def foo(): pass foo._docstring_components = [None] @doc(foo) def bar(): pass

So, my preference is an actual fix, like the one I proposed, than a typing ignore.

I think that code ensures _docstring_components is not set to None (in the parent). And the exception won't happen as far as _docstring_components is always set with the @doc decorator. But technically speaking there is no guarantee that something like this can be done:

def foo(): pass foo._docstring_components = [None] @doc(foo) def bar(): pass

So, my preference is an actual fix, like the one I proposed, than a typing ignore.

That is a good point! Error fixed. Thanks!

datapythonista · 2020-03-08T01:48:25Z

pandas/util/_decorators.py

+            ]
+        )
+
+        # saving docstring components list


Probably not worth having this comment.

datapythonista

lgtm, just one idea that could make sense.

Thanks for the work on this @HH-MWB, it's a great improvement.

datapythonista · 2020-03-10T01:23:26Z

pandas/util/_decorators.py

+            if hasattr(arg, "_docstring_components"):
+                docstring_components.extend(arg._docstring_components)  # type: ignore


Not sure if this could be a better fix for this one:

Suggested change

if hasattr(arg, "_docstring_components"):

docstring_components.extend(arg._docstring_components) # type: ignore

if hasattr(arg, "_docstring_components") and isinstance(arg._docstring_components, list):

docstring_components.extend(arg._docstring_components)

Hi @datapythonista. I would consider this as two parts. One is to add type check for _docstring_components and one for remove # type: ignore.

For remove # type: ignore

After I made the change, I still get the errors below and not sure if we can remove that.

pandas/util/_decorators.py:280: error: Item "str" of "Union[str, Callable[..., Any]]" has no attribute "_docstring_components" pandas/util/_decorators.py:280: error: Item "function" of "Union[str, Callable[..., Any]]" has no attribute "_docstring_components"

I guess my settings are not the same as checks here. I even the last passed build, I still get mypy errors on local. So, the change might fix it and just I don't know.

For checking if _docstring_components is list

I think this is a very good point, we do as defensive programming. But I am a little bit curious about how far we should go to protect it. Also, since this function is internal use for pandas only, would it be better to let it failed? Then the developer would get notified there is a conflict immediately?

Just to be clarified, I would be happy to make this change if you (or anyone) feels needed. Because I see both pros and cons for this, I am not sure which one will be better.

I'm happy to merge this as is, was wondering if the suggested change would fix the problem.

We can surely go back to it in the future if needed, but let's merge this first, since this already adds a lot of value, and we can start replacing Appender by doc at a larger scale after this.

WillAyd · 2020-03-11T02:09:36Z

pandas/core/arrays/categorical.py

@@ -51,7 +52,7 @@
    _extension_array_shared_docs,
    try_cast_to_ea,
 )
-from pandas.core.base import NoNewAttributesMixin, PandasObject, _shared_docs
+from pandas.core.base import IndexOpsMixin, NoNewAttributesMixin, PandasObject


I believe I am in the minority on this view and don't want to be overly difficult, but can you refactor the IndexOpsMixin as a pre-cursor to this, or leave this module separate from the rest of changes (which look good btw)

Yes, that is a good idea. I would convert this back to using _shared_docs. This PR is already too large, and I should really put them separately.

datapythonista

Can you merge master. I guess the CI problems are unrelated.

datapythonista · 2020-03-11T13:12:02Z

pandas/util/_decorators.py

+            if hasattr(arg, "_docstring_components"):
+                docstring_components.extend(arg._docstring_components)  # type: ignore


I'm happy to merge this as is, was wondering if the suggested change would fix the problem.

We can surely go back to it in the future if needed, but let's merge this first, since this already adds a lot of value, and we can start replacing Appender by doc at a larger scale after this.

WillAyd · 2020-03-17T00:10:49Z

Great thanks @HH-MWB ! If you can do a follow up to fix the IndexOpsMixin import would be very much appreciated!

HH-MWB · 2020-03-17T00:49:26Z

Great thanks @HH-MWB ! If you can do a follow up to fix the IndexOpsMixin import would be very much appreciated!

Yes. I am thinking what is the best way of doing that. Definitely want to keep making contributing on this! Thanks.

HH-MWB added 2 commits February 13, 2020 21:19

CLN: @doc - base.py

06a5d9e

CLN: @doc - indexing.py

9a4abae

HH-MWB force-pushed the use-doc-decorator branch from b7d7ac9 to 9a4abae Compare February 14, 2020 03:57

Merge branch 'master' of https://github.com/pandas-dev/pandas into us…

e580a96

…e-doc-decorator

HH-MWB force-pushed the use-doc-decorator branch from 405589b to e580a96 Compare February 14, 2020 05:17

HH-MWB requested a review from WillAyd February 14, 2020 12:36

simonjayhawkins added Clean Docs labels Feb 14, 2020

simonjayhawkins reviewed Feb 14, 2020

View reviewed changes

Merge branch 'master' of https://github.com/pandas-dev/pandas into us…

ede988c

…e-doc-decorator

HH-MWB force-pushed the use-doc-decorator branch from 020c24f to ede988c Compare February 15, 2020 02:03

HH-MWB added 2 commits February 15, 2020 10:44

Merge branch 'master' of https://github.com/pandas-dev/pandas into us…

41a7552

…e-doc-decorator

ENH: auto convert __doc__ to be format ready

6cd8082

HH-MWB force-pushed the use-doc-decorator branch from a002400 to 6cd8082 Compare February 15, 2020 15:46

HH-MWB added 2 commits February 15, 2020 11:48

CLN: remove _shared_docs

1534dad

DOC: update klass in docstring

703f7a6

HH-MWB force-pushed the use-doc-decorator branch from 4139d59 to 703f7a6 Compare February 15, 2020 16:49

simonjayhawkins approved these changes Feb 15, 2020

View reviewed changes

simonjayhawkins added this to the 1.1 milestone Feb 15, 2020

HH-MWB added 2 commits March 1, 2020 14:02

organize the code in doc decorator

8e320ce

fix test case for doc

1ee2cf7

HH-MWB requested a review from datapythonista March 1, 2020 22:19

simonjayhawkins reviewed Mar 2, 2020

View reviewed changes

remove unused ignores

7bf242e

datapythonista reviewed Mar 7, 2020

View reviewed changes

HH-MWB added 2 commits March 7, 2020 18:35

avoid type ignore in doc

61e87af

avoid misleading var name

224cec8

datapythonista reviewed Mar 8, 2020

View reviewed changes

remove trailing space

e01e108

datapythonista reviewed Mar 8, 2020

View reviewed changes

HH-MWB added 2 commits March 7, 2020 22:10

remove extra comment

48cfcda

fix edge case for dedent

20f96f2

datapythonista approved these changes Mar 10, 2020

View reviewed changes

WillAyd requested changes Mar 11, 2020

View reviewed changes

DOC: using _shared_docs for categorical.py

76ed611

datapythonista approved these changes Mar 11, 2020

View reviewed changes

Merge remote-tracking branch 'upstream/master' into use-doc-decorator

95a86ee

HH-MWB force-pushed the use-doc-decorator branch from c2c4931 to 95a86ee Compare March 11, 2020 23:21

HH-MWB requested a review from WillAyd March 12, 2020 02:03

WillAyd merged commit 871e3af into pandas-dev:master Mar 17, 2020

SeeminSyed pushed a commit to CSCD01-team01/pandas that referenced this pull request Mar 22, 2020

CLN: @doc - base.py & indexing.py (pandas-dev#31970)

091cd7e

HH-MWB deleted the use-doc-decorator branch March 23, 2020 21:42

	arg.format(**kwargs) if isinstance(arg, str) else dedent(arg.__doc__)
	arg.format(**kwargs) if isinstance(arg, str) else dedent(arg.__doc__ or "")

		if hasattr(arg, "_docstring_components"):
		docstring_components.extend(arg._docstring_components) # type: ignore

CLN: @doc - base.py & indexing.py #31970

CLN: @doc - base.py & indexing.py #31970

Conversation

HH-MWB commented Feb 14, 2020 • edited Loading

HH-MWB commented Feb 14, 2020

WillAyd commented Feb 14, 2020

simonjayhawkins left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

HH-MWB commented Feb 15, 2020

WillAyd commented Feb 15, 2020

HH-MWB commented Feb 15, 2020

simonjayhawkins left a comment

Choose a reason for hiding this comment

HH-MWB commented Feb 15, 2020

Inherit Docstring

Case Here

HH-MWB commented Feb 15, 2020

simonjayhawkins commented Feb 15, 2020

simonjayhawkins commented Feb 15, 2020

HH-MWB commented Feb 16, 2020

simonjayhawkins commented Feb 16, 2020

pep8speaks commented Feb 29, 2020 • edited Loading

Comment last updated at 2020-03-11 23:21:34 UTC

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

HH-MWB Mar 4, 2020 • edited Loading

Choose a reason for hiding this comment

HH-MWB commented Mar 3, 2020 • edited Loading

simonjayhawkins commented Mar 3, 2020 • edited Loading

HH-MWB commented Mar 4, 2020

Choose a reason for hiding this comment

Choose a reason for hiding this comment

datapythonista left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

HH-MWB commented Mar 8, 2020

Choose a reason for hiding this comment

HH-MWB Mar 8, 2020 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

datapythonista left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

HH-MWB Mar 11, 2020 • edited Loading

Choose a reason for hiding this comment

For remove # type: ignore

For checking if _docstring_components is list

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

datapythonista left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

WillAyd commented Mar 17, 2020

HH-MWB commented Mar 17, 2020

HH-MWB commented Feb 14, 2020 •

edited

Loading

pep8speaks commented Feb 29, 2020 •

edited

Loading

HH-MWB Mar 4, 2020 •

edited

Loading

HH-MWB commented Mar 3, 2020 •

edited

Loading

simonjayhawkins commented Mar 3, 2020 •

edited

Loading

HH-MWB Mar 8, 2020 •

edited

Loading

HH-MWB Mar 11, 2020 •

edited

Loading

For remove `# type: ignore`

For checking if `_docstring_components` is `list`