repo: inline filter_file(..., **kwargs)

[haampie]

Type checkers have issue with kwargs = {...} + f(**kwargs) due to type
erasure. The pattern is often somewhat redundant and seems to have been
copied to match style of other packages.

• Inline **kwargs
• Remove kwargs that repeat defaults

Before: 1762 diagnostics, after: 1525 diagnostics; from ty check --extra-search-path ~/spack/lib/spack --extra-search-path repos --ignore unresolved-reference --output-format concise

was done automatically

#!/usr/bin/env python3

import ast
import sys

filter_file_defaults = {
    "string":False,
    "backup":False,
    "ignore_absent":False,
    "start_at":None,
    "stop_at":None,
    "encoding":"utf-8",
}



def check_filter_file_calls(filename):
    with open(filename, "r") as f:
        lines = f.readlines()

    tree = ast.parse("".join(lines), filename=filename)

    for node in ast.walk(tree):
        # find the last assignment to `kwargs` in the current scope
        if isinstance(node, ast.Assign) and isinstance(node.value, ast.Dict):
            for target in node.targets:
                if isinstance(target, ast.Name) and target.id == "kwargs":
                    kwargs_value = {}
                    for key, value in zip(node.value.keys, node.value.values):
                        if isinstance(key, ast.Constant):
                            kwargs_value[key.value] = ast.literal_eval(value)
                    print(f"{filename}:{node.lineno}:{node.col_offset}:{node.end_col_offset}: kwargs = {kwargs_value}")
                    # remove all defaults from kwargs_value
                    for k in filter_file_defaults:
                        if k in kwargs_value and kwargs_value[k] == filter_file_defaults[k]:
                            del kwargs_value[k]

        # either the funcion call is `filter_file(..., **kwargs)`
        # or its obj.filter(..., **kwargs)

        if not isinstance(node, ast.Call):
            continue
        
        if (
            getattr(node.func, "id", None) == "filter_file"
            or isinstance(node.func, ast.Attribute) and node.func.attr == "filter"
        ):
            if node.keywords:
                for kw in node.keywords:
                    if isinstance(kw.value, ast.Name) and kw.value.id == "kwargs":
                        if not kwargs_value:
                            # kwargs is empty, we delete this part from the function call
                            lines[kw.lineno - 1] = lines[kw.lineno - 1][:kw.col_offset] + lines[kw.lineno - 1][kw.end_col_offset:]
                        else:
                            # we inline the kwargs values directly into the function call
                            lines[kw.lineno - 1] = lines[kw.lineno - 1][:kw.col_offset] + ", ".join(f"{k}={repr(v)}" for k, v in kwargs_value.items()) + lines[kw.lineno - 1][kw.end_col_offset:]
                        print(lines[kw.lineno - 1], end="")

    with open(filename, "w") as f:
        f.writelines(lines)

            

if __name__ == "__main__":
    for filename in sys.argv[1:]:
        check_filter_file_calls(filename)

[climbfuji]

I mainly looked at eccodes. Thanks for cleaning this up.