plexpy/lib/packaging/utils.py

# This file is dual licensed under the terms of the Apache License, Version
# 2.0, and the BSD License. See the LICENSE file in the root of this repository
# for complete details.

from __future__ import annotations

import functools
import re
from typing import NewType, Tuple, Union, cast

from .tags import Tag, parse_tag
from .version import InvalidVersion, Version, _TrimmedRelease

BuildTag = Union[Tuple[()], Tuple[int, str]]
NormalizedName = NewType("NormalizedName", str)


class InvalidName(ValueError):
    """
    An invalid distribution name; users should refer to the packaging user guide.
    """


class InvalidWheelFilename(ValueError):
    """
    An invalid wheel filename was found, users should refer to PEP 427.
    """


class InvalidSdistFilename(ValueError):
    """
    An invalid sdist filename was found, users should refer to the packaging user guide.
    """


# Core metadata spec for `Name`
_validate_regex = re.compile(
    r"^([A-Z0-9]|[A-Z0-9][A-Z0-9._-]*[A-Z0-9])$", re.IGNORECASE
)
_canonicalize_regex = re.compile(r"[-_.]+")
_normalized_regex = re.compile(r"^([a-z0-9]|[a-z0-9]([a-z0-9-](?!--))*[a-z0-9])$")
# PEP 427: The build number must start with a digit.
_build_tag_regex = re.compile(r"(\d+)(.*)")


def canonicalize_name(name: str, *, validate: bool = False) -> NormalizedName:
    if validate and not _validate_regex.match(name):
        raise InvalidName(f"name is invalid: {name!r}")
    # This is taken from PEP 503.
    value = _canonicalize_regex.sub("-", name).lower()
    return cast(NormalizedName, value)


def is_normalized_name(name: str) -> bool:
    return _normalized_regex.match(name) is not None


@functools.singledispatch
def canonicalize_version(
    version: Version | str, *, strip_trailing_zero: bool = True
) -> str:
    """
    Return a canonical form of a version as a string.

    >>> canonicalize_version('1.0.1')
    '1.0.1'

    Per PEP 625, versions may have multiple canonical forms, differing
    only by trailing zeros.

    >>> canonicalize_version('1.0.0')
    '1'
    >>> canonicalize_version('1.0.0', strip_trailing_zero=False)
    '1.0.0'

    Invalid versions are returned unaltered.

    >>> canonicalize_version('foo bar baz')
    'foo bar baz'
    """
    return str(_TrimmedRelease(str(version)) if strip_trailing_zero else version)


@canonicalize_version.register
def _(version: str, *, strip_trailing_zero: bool = True) -> str:
    try:
        parsed = Version(version)
    except InvalidVersion:
        # Legacy versions cannot be normalized
        return version
    return canonicalize_version(parsed, strip_trailing_zero=strip_trailing_zero)


def parse_wheel_filename(
    filename: str,
) -> tuple[NormalizedName, Version, BuildTag, frozenset[Tag]]:
    if not filename.endswith(".whl"):
        raise InvalidWheelFilename(
            f"Invalid wheel filename (extension must be '.whl'): {filename!r}"
        )

    filename = filename[:-4]
    dashes = filename.count("-")
    if dashes not in (4, 5):
        raise InvalidWheelFilename(
            f"Invalid wheel filename (wrong number of parts): {filename!r}"
        )

    parts = filename.split("-", dashes - 2)
    name_part = parts[0]
    # See PEP 427 for the rules on escaping the project name.
    if "__" in name_part or re.match(r"^[\w\d._]*$", name_part, re.UNICODE) is None:
        raise InvalidWheelFilename(f"Invalid project name: {filename!r}")
    name = canonicalize_name(name_part)

    try:
        version = Version(parts[1])
    except InvalidVersion as e:
        raise InvalidWheelFilename(
            f"Invalid wheel filename (invalid version): {filename!r}"
        ) from e

    if dashes == 5:
        build_part = parts[2]
        build_match = _build_tag_regex.match(build_part)
        if build_match is None:
            raise InvalidWheelFilename(
                f"Invalid build number: {build_part} in {filename!r}"
            )
        build = cast(BuildTag, (int(build_match.group(1)), build_match.group(2)))
    else:
        build = ()
    tags = parse_tag(parts[-1])
    return (name, version, build, tags)


def parse_sdist_filename(filename: str) -> tuple[NormalizedName, Version]:
    if filename.endswith(".tar.gz"):
        file_stem = filename[: -len(".tar.gz")]
    elif filename.endswith(".zip"):
        file_stem = filename[: -len(".zip")]
    else:
        raise InvalidSdistFilename(
            f"Invalid sdist filename (extension must be '.tar.gz' or '.zip'):"
            f" {filename!r}"
        )

    # We are requiring a PEP 440 version, which cannot contain dashes,
    # so we split on the last dash.
    name_part, sep, version_part = file_stem.rpartition("-")
    if not sep:
        raise InvalidSdistFilename(f"Invalid sdist filename: {filename!r}")

    name = canonicalize_name(name_part)

    try:
        version = Version(version_part)
    except InvalidVersion as e:
        raise InvalidSdistFilename(
            f"Invalid sdist filename (invalid version): {filename!r}"
        ) from e

    return (name, version)