Function _normalize_target

openkb/lint.py:30–47

Normalize a wikilink target for fuzzy comparison. Applies, in order: - NFKC unicode normalization (e.g. full-width '）' → ASCII ')') - Lowercase - Underscore → hyphen - Collapse repeated hyphens - Strip leading/trailing hyphens (per segment when path-like) Path separator

(target: str)

def _normalize_target(target: str) -> str:
    """Normalize a wikilink target for fuzzy comparison.

    Applies, in order:
    - NFKC unicode normalization (e.g. full-width '）' → ASCII ')')
    - Lowercase
    - Underscore → hyphen
    - Collapse repeated hyphens
    - Strip leading/trailing hyphens (per segment when path-like)

    Path separators are preserved so ``concepts/Gist_Memory`` normalizes to
    ``concepts/gist-memory``.
    """
    s = unicodedata.normalize("NFKC", target)
    s = s.lower().replace("_", "-")
    # Normalize each path segment independently to avoid collapsing the '/'
    parts = [re.sub(r"-+", "-", p).strip("-") for p in s.split("/")]
    return "/".join(parts)


def build_norm_index(known_targets: set[str]) -> dict[str, str]: