nem sei pq tantos arquivos

This commit is contained in:
2025-02-11 11:07:58 -03:00
parent 66fb4eb17b
commit 2da09a8a25
1841 changed files with 115867 additions and 77478 deletions

View File

@@ -27,7 +27,6 @@ from pip._internal.utils.misc import (
split_auth_from_netloc,
splitext,
)
from pip._internal.utils.models import KeyBasedCompareMixin
from pip._internal.utils.urls import path_to_url, url_to_path
if TYPE_CHECKING:
@@ -55,25 +54,25 @@ class LinkHash:
name: str
value: str
_hash_re = re.compile(
_hash_url_fragment_re = re.compile(
# NB: we do not validate that the second group (.*) is a valid hex
# digest. Instead, we simply keep that string in this class, and then check it
# against Hashes when hash-checking is needed. This is easier to debug than
# proactively discarding an invalid hex digest, as we handle incorrect hashes
# and malformed hashes in the same place.
r"({choices})=(.*)".format(
r"[#&]({choices})=([^&]*)".format(
choices="|".join(re.escape(hash_name) for hash_name in _SUPPORTED_HASHES)
),
)
def __post_init__(self) -> None:
assert self._hash_re.match(f"{self.name}={self.value}")
assert self.name in _SUPPORTED_HASHES
@classmethod
@functools.lru_cache(maxsize=None)
def split_hash_name_and_value(cls, url: str) -> Optional["LinkHash"]:
def find_hash_url_fragment(cls, url: str) -> Optional["LinkHash"]:
"""Search a string for a checksum algorithm name and encoded output value."""
match = cls._hash_re.search(url)
match = cls._hash_url_fragment_re.search(url)
if match is None:
return None
name, value = match.groups()
@@ -95,6 +94,28 @@ class LinkHash:
return hashes.is_hash_allowed(self.name, hex_digest=self.value)
@dataclass(frozen=True)
class MetadataFile:
"""Information about a core metadata file associated with a distribution."""
hashes: Optional[Dict[str, str]]
def __post_init__(self) -> None:
if self.hashes is not None:
assert all(name in _SUPPORTED_HASHES for name in self.hashes)
def supported_hashes(hashes: Optional[Dict[str, str]]) -> Optional[Dict[str, str]]:
# Remove any unsupported hash types from the mapping. If this leaves no
# supported hashes, return None
if hashes is None:
return None
hashes = {n: v for n, v in hashes.items() if n in _SUPPORTED_HASHES}
if not hashes:
return None
return hashes
def _clean_url_path_part(part: str) -> str:
"""
Clean a "part" of a URL path (i.e. after splitting on "@" characters).
@@ -149,25 +170,38 @@ def _ensure_quoted_url(url: str) -> str:
and without double-quoting other characters.
"""
# Split the URL into parts according to the general structure
# `scheme://netloc/path;parameters?query#fragment`.
result = urllib.parse.urlparse(url)
# `scheme://netloc/path?query#fragment`.
result = urllib.parse.urlsplit(url)
# If the netloc is empty, then the URL refers to a local filesystem path.
is_local_path = not result.netloc
path = _clean_url_path(result.path, is_local_path=is_local_path)
return urllib.parse.urlunparse(result._replace(path=path))
return urllib.parse.urlunsplit(result._replace(path=path))
class Link(KeyBasedCompareMixin):
def _absolute_link_url(base_url: str, url: str) -> str:
"""
A faster implementation of urllib.parse.urljoin with a shortcut
for absolute http/https URLs.
"""
if url.startswith(("https://", "http://")):
return url
else:
return urllib.parse.urljoin(base_url, url)
@functools.total_ordering
class Link:
"""Represents a parsed link from a Package Index's simple URL"""
__slots__ = [
"_parsed_url",
"_url",
"_path",
"_hashes",
"comes_from",
"requires_python",
"yanked_reason",
"dist_info_metadata",
"metadata_file_data",
"cache_link_parsing",
"egg_fragment",
]
@@ -178,7 +212,7 @@ class Link(KeyBasedCompareMixin):
comes_from: Optional[Union[str, "IndexContent"]] = None,
requires_python: Optional[str] = None,
yanked_reason: Optional[str] = None,
dist_info_metadata: Optional[str] = None,
metadata_file_data: Optional[MetadataFile] = None,
cache_link_parsing: bool = True,
hashes: Optional[Mapping[str, str]] = None,
) -> None:
@@ -196,11 +230,10 @@ class Link(KeyBasedCompareMixin):
a simple repository HTML link. If the file has been yanked but
no reason was provided, this should be the empty string. See
PEP 592 for more information and the specification.
:param dist_info_metadata: the metadata attached to the file, or None if no such
metadata is provided. This is the value of the "data-dist-info-metadata"
attribute, if present, in a simple repository HTML link. This may be parsed
into its own `Link` by `self.metadata_link()`. See PEP 658 for more
information and the specification.
:param metadata_file_data: the metadata attached to the file, or None if
no such metadata is provided. This argument, if not None, indicates
that a separate metadata file exists, and also optionally supplies
hashes for that file.
:param cache_link_parsing: A flag that is used elsewhere to determine
whether resources retrieved from this link should be cached. PyPI
URLs should generally have this set to False, for example.
@@ -208,6 +241,10 @@ class Link(KeyBasedCompareMixin):
determine the validity of a download.
"""
# The comes_from, requires_python, and metadata_file_data arguments are
# only used by classmethods of this class, and are not used in client
# code directly.
# url can be a UNC windows share
if url.startswith("\\\\"):
url = path_to_url(url)
@@ -216,8 +253,10 @@ class Link(KeyBasedCompareMixin):
# Store the url as a private attribute to prevent accidentally
# trying to set a new value.
self._url = url
# The .path property is hot, so calculate its value ahead of time.
self._path = urllib.parse.unquote(self._parsed_url.path)
link_hash = LinkHash.split_hash_name_and_value(url)
link_hash = LinkHash.find_hash_url_fragment(url)
hashes_from_link = {} if link_hash is None else link_hash.as_dict()
if hashes is None:
self._hashes = hashes_from_link
@@ -227,9 +266,7 @@ class Link(KeyBasedCompareMixin):
self.comes_from = comes_from
self.requires_python = requires_python if requires_python else None
self.yanked_reason = yanked_reason
self.dist_info_metadata = dist_info_metadata
super().__init__(key=url, defining_class=Link)
self.metadata_file_data = metadata_file_data
self.cache_link_parsing = cache_link_parsing
self.egg_fragment = self._egg_fragment()
@@ -247,12 +284,28 @@ class Link(KeyBasedCompareMixin):
if file_url is None:
return None
url = _ensure_quoted_url(urllib.parse.urljoin(page_url, file_url))
url = _ensure_quoted_url(_absolute_link_url(page_url, file_url))
pyrequire = file_data.get("requires-python")
yanked_reason = file_data.get("yanked")
dist_info_metadata = file_data.get("dist-info-metadata")
hashes = file_data.get("hashes", {})
# PEP 714: Indexes must use the name core-metadata, but
# clients should support the old name as a fallback for compatibility.
metadata_info = file_data.get("core-metadata")
if metadata_info is None:
metadata_info = file_data.get("dist-info-metadata")
# The metadata info value may be a boolean, or a dict of hashes.
if isinstance(metadata_info, dict):
# The file exists, and hashes have been supplied
metadata_file_data = MetadataFile(supported_hashes(metadata_info))
elif metadata_info:
# The file exists, but there are no hashes
metadata_file_data = MetadataFile(None)
else:
# False or not present: the file does not exist
metadata_file_data = None
# The Link.yanked_reason expects an empty string instead of a boolean.
if yanked_reason and not isinstance(yanked_reason, str):
yanked_reason = ""
@@ -266,7 +319,7 @@ class Link(KeyBasedCompareMixin):
requires_python=pyrequire,
yanked_reason=yanked_reason,
hashes=hashes,
dist_info_metadata=dist_info_metadata,
metadata_file_data=metadata_file_data,
)
@classmethod
@@ -283,17 +336,42 @@ class Link(KeyBasedCompareMixin):
if not href:
return None
url = _ensure_quoted_url(urllib.parse.urljoin(base_url, href))
url = _ensure_quoted_url(_absolute_link_url(base_url, href))
pyrequire = anchor_attribs.get("data-requires-python")
yanked_reason = anchor_attribs.get("data-yanked")
dist_info_metadata = anchor_attribs.get("data-dist-info-metadata")
# PEP 714: Indexes must use the name data-core-metadata, but
# clients should support the old name as a fallback for compatibility.
metadata_info = anchor_attribs.get("data-core-metadata")
if metadata_info is None:
metadata_info = anchor_attribs.get("data-dist-info-metadata")
# The metadata info value may be the string "true", or a string of
# the form "hashname=hashval"
if metadata_info == "true":
# The file exists, but there are no hashes
metadata_file_data = MetadataFile(None)
elif metadata_info is None:
# The file does not exist
metadata_file_data = None
else:
# The file exists, and hashes have been supplied
hashname, sep, hashval = metadata_info.partition("=")
if sep == "=":
metadata_file_data = MetadataFile(supported_hashes({hashname: hashval}))
else:
# Error - data is wrong. Treat as no hashes supplied.
logger.debug(
"Index returned invalid data-dist-info-metadata value: %s",
metadata_info,
)
metadata_file_data = MetadataFile(None)
return cls(
url,
comes_from=page_url,
requires_python=pyrequire,
yanked_reason=yanked_reason,
dist_info_metadata=dist_info_metadata,
metadata_file_data=metadata_file_data,
)
def __str__(self) -> str:
@@ -302,15 +380,26 @@ class Link(KeyBasedCompareMixin):
else:
rp = ""
if self.comes_from:
return "{} (from {}){}".format(
redact_auth_from_url(self._url), self.comes_from, rp
)
return f"{redact_auth_from_url(self._url)} (from {self.comes_from}){rp}"
else:
return redact_auth_from_url(str(self._url))
def __repr__(self) -> str:
return f"<Link {self}>"
def __hash__(self) -> int:
return hash(self.url)
def __eq__(self, other: Any) -> bool:
if not isinstance(other, Link):
return NotImplemented
return self.url == other.url
def __lt__(self, other: Any) -> bool:
if not isinstance(other, Link):
return NotImplemented
return self.url < other.url
@property
def url(self) -> str:
return self._url
@@ -346,7 +435,7 @@ class Link(KeyBasedCompareMixin):
@property
def path(self) -> str:
return urllib.parse.unquote(self._parsed_url.path)
return self._path
def splitext(self) -> Tuple[str, str]:
return splitext(posixpath.basename(self.path.rstrip("/")))
@@ -377,10 +466,10 @@ class Link(KeyBasedCompareMixin):
project_name = match.group(1)
if not self._project_name_re.match(project_name):
deprecated(
reason=f"{self} contains an egg fragment with a non-PEP 508 name",
reason=f"{self} contains an egg fragment with a non-PEP 508 name.",
replacement="to use the req @ url syntax, and remove the egg fragment",
gone_in="25.0",
issue=11617,
gone_in="25.1",
issue=13157,
)
return project_name
@@ -395,22 +484,13 @@ class Link(KeyBasedCompareMixin):
return match.group(1)
def metadata_link(self) -> Optional["Link"]:
"""Implementation of PEP 658 parsing."""
# Note that Link.from_element() parsing the "data-dist-info-metadata" attribute
# from an HTML anchor tag is typically how the Link.dist_info_metadata attribute
# gets set.
if self.dist_info_metadata is None:
"""Return a link to the associated core metadata file (if any)."""
if self.metadata_file_data is None:
return None
metadata_url = f"{self.url_without_fragment}.metadata"
# If data-dist-info-metadata="true" is set, then the metadata file exists,
# but there is no information about its checksum or anything else.
if self.dist_info_metadata != "true":
link_hash = LinkHash.split_hash_name_and_value(self.dist_info_metadata)
else:
link_hash = None
if link_hash is None:
if self.metadata_file_data.hashes is None:
return Link(metadata_url)
return Link(metadata_url, hashes=link_hash.as_dict())
return Link(metadata_url, hashes=self.metadata_file_data.hashes)
def as_hashes(self) -> Hashes:
return Hashes({k: [v] for k, v in self._hashes.items()})