pandas read_csv() parse_dates parameter has incorrect type

[Dr-Irv]

Environment data

• Language Server version: 2020.11.0
• OS and version: Windows 10
• Python version (& distribution if applicable, e.g. Anaconda): 3.7.8 (Anaconda)

Expected behaviour

No bug

Actual behaviour

df = pd.read_csv("foo.csv", parse_dates=["Start", "Finish"])

reports

No overloads for "pd.read_csv("foo.csv", parse_dates=["Start", "Finish"])" match parameters
  Argument types: (Literal['foo.csv'], List[str])

The issue here is that the parse_dates argument can be a number of things, including a list of column names.

Code Snippet / Additional information

Partial fix to pandas/io/parsers.pyi

 diff parsers.pyi.ORIG parsers.pyi
37c37
<     parse_dates: bool = ...,
---
>     parse_dates: Union[bool, List[int], List[str]] = ...,
374c374
<     parse_dates: bool = ...,
---
>     parse_dates: Union[bool, List[int], List[str]] = ...,
427c427
<     parse_dates: bool = ...,
---
>     parse_dates: Union[bool, List[int], List[str]] = ...,
764c764
<     parse_dates: bool = ...,
---
>     parse_dates: Union[bool, List[int], List[str]] = ...,

NOTE: This is not all the possible argument types for parse_dates, but the above solves this problem.

[jakebailey]

Thanks again for the patch.

This issue has been fixed in version 2020.11.1, which we've just released. You can find the changelog here: https://github.com/microsoft/pylance-release/blob/master/CHANGELOG.md#2020111-11-november-2020