Description
Pylint scan found some Python code defects in ray data, such as:
ray/python/data/datasource/datasink.py:13:0: C0103: Type variable name "WriteReturnType" doesn't conform to predefined naming style (invalid-name)
ray/python/ray/data/datasource/file_based_datasource.py:276:20: R1730: Consider using 'num_threads = min(num_threads, len(read_paths))' instead of unnecessary if block (consider-using-min-builtin)
ray/python/ray/data/read_api.py:3214:4: R1705: Unnecessary "elif" after "return", remove the leading "el" from "elif" (no-else-return)
/python/ray/data/dataset.py:845:20: R1701: Consider merging these isinstance calls to isinstance(column, (pd.DataFrame, pd.Index, pd.Series)) (consider-merging-isinstance)
ray/python/ray/data/dataset.py:867:16: R1705: Unnecessary "else" after "return", remove the "else" and de-indent the code inside it (no-else-return)
All these issues do not affect the logic of program execution, but we can fix these defects, improve code readability, reduce loop load, or remove useless code, and so on.
For example, the corresponding source code for ray/python/ray/data/dataset.py: 845:20 above is:
“if (
isinstance(column, pd.Series)
or isinstance(column, pd.DataFrame)
or isinstance(column, pd.Index)
):
”
We can completely merge the above three judgments into one isinstance (column, (pd. DataFrame, pd. Index, pd. Series)) without affecting the logic, and the code is more concise.