Use str.splitlines() to split lines

[eumiro]

No description provided.

[amoghrajesh]

Changes look good but the CI is failing with the error:

dev/prepare_release_issue.py:304: error:
"Callable[[str, str, SupportsIndex], str]" has no attribute "splitlines" 
[attr-defined]
                    body = " ".join(pr.body.replace.splitlines())
                                    ^~~~~~~~~~~~~~~~~~~~~~~~~~
Found 1 error in 1 file (checked 117 source files)

Have you installed the pre commit? Makes it easier to track static checks..

[eumiro]

@amoghrajesh thanks, I have it installed, but for some reason disabled yesterday. Now it is fixed.

[uranusjr]

I would keep the with open style and rewrite the reading part to

[re2.sub(r"\s*#.*", "", line) for line in ifile]

This would allow the input to be read line by line instead of creating a temporary string before splitting.

[eumiro]

How about

        ignore_file_path = Path(root) / ignore_file_name
        if ignore_file_path.is_file():
            uniq_patterns = {}
            with ignore_file_path.open() as f:
                for line in f:
                    line = line.strip()
                    if line and not line.startswith("#"):
                        pattern = ignore_rule_type.compile(line, Path(base_dir_path), ignore_file_path)
                        if pattern:
                            uniq_patterns[pattern] = None
            # evaluation order of patterns is important with negation
            # so that later patterns can override earlier patterns
            patterns = list(uniq_patterns)

This probably describes the steps even better, and shows that the ignore_rule_type.compile is called before the patterns are filtered via uniq (dict).

[uranusjr]

Yeah I wouldn’t object.