"""Session language policy, independent of workspace document content."""
import re

_LANGUAGE_TAG = re.compile(
    r"(?P<language>[A-Za-z]{2,8})(?:-(?P<script>[A-Za-z]{4}))?"
    r"(?:-(?P<region>[A-Za-z]{2}|[0-9]{3}))?"
    r"(?P<rest>(?:-(?:[A-Za-z0-9]{5,8}|[0-9][A-Za-z0-9]{3}))*"
    r"(?:-[0-9A-WY-Za-wy-z](?:-[A-Za-z0-9]{2,8})+)*"
    r"(?:-[xX](?:-[A-Za-z0-9]{1,8})+)?)"
)


def validate_language_tag(value: str) -> str:
    """Validate syntax and normalize casing without depending on UI catalogs."""
    match = _LANGUAGE_TAG.fullmatch(value) if isinstance(value, str) else None
    if match is None:
        raise ValueError("interaction_language must be a BCP-47 language tag")
    parts = [match["language"].lower()]
    if match["script"]:
        parts.append(match["script"].title())
    if match["region"]:
        parts.append(match["region"].upper())
    return "-".join(parts) + match["rest"].lower()