monkee.ch/SickGear
1
0
Fork 0
mirror of https://github.com/SickGear/SickGear.git synced 2024-12-19 01:03:37 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'feature/UpdateFeedparser' into dev

Browse source
This commit is contained in:
JackDandy 2023-04-27 12:37:26 +01:00
commit 0794ca330f
33 changed files with 3280 additions and 2349 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
CHANGES.md
View file

@ -2,6 +2,7 @@
* Update attr 22.2.0 (a9960de) to 22.2.0 (683d056)
* Update diskcache 5.4.0 (1cb1425) to 5.6.1 (4d30686)
* Update feedparser 6.0.10 (5fcb3ae) to 6.0.10 (6d032b8)
* Update filelock 3.9.0 (ce3e891) to 3.11.0 (d3241b9)
* Update Msgpack 1.0.4 (b5acfd5) to 1.0.5 (0516c2c)
* Update Requests library 2.28.1 (ec553c2) to 2.29.0 (87d63de)

33
lib/feedparser/__init__.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -27,12 +27,18 @@
from .api import parse
from .datetimes import registerDateHandler
from .exceptions import *
from .exceptions import (
CharacterEncodingOverride,
CharacterEncodingUnknown,
FeedparserError,
NonXMLContentType,
UndeclaredNamespace,
)
from .util import FeedParserDict
__author__ = 'Kurt McKee <contactme@kurtmckee.org>'
__license__ = 'BSD 2-clause'
__version__ = '6.0.10'
__author__ = "Kurt McKee <contactme@kurtmckee.org>"
__license__ = "BSD 2-clause"
__version__ = "6.0.10"
# HTTP "User-Agent" header to send to servers when downloading feeds.
# If you are embedding feedparser in a larger application, you should
@ -46,3 +52,20 @@ RESOLVE_RELATIVE_URIS = 1
# If you want feedparser to automatically sanitize all potentially unsafe
# HTML content, set this to 1.
SANITIZE_HTML = 1
# If you want feedparser to use only a prefix of the feed to detect encodings
# (uses less memory), set this to 1.
OPTIMISTIC_ENCODING_DETECTION = 1
__all__ = (
"parse",
"registerDateHandler",
"FeedParserDict",
"FeedparserError",
"CharacterEncodingOverride",
"CharacterEncodingUnknown",
"NonXMLContentType",
"UndeclaredNamespace",
)

360
lib/feedparser/api.py
View file

@ -1,5 +1,5 @@
# The public API for feedparser
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -26,29 +26,23 @@
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
import datetime
import io
import time
from typing import Dict, List, Union
import urllib.error
import urllib.parse
import xml.sax
from typing import IO, Dict, Optional, Union
import sgmllib3k as sgmllib
from .datetimes import registerDateHandler, _parse_date
from .encodings import convert_to_utf8
from .html import BaseHTMLProcessor
from . import http
from .encodings import MissingEncoding, convert_file_to_utf8
from .html import BaseHTMLProcessor
from .mixin import XMLParserMixin
from .parsers.json import JSONParser
from .parsers.loose import LooseXMLParser
from .parsers.strict import StrictXMLParser
from .parsers.json import JSONParser
from .sanitizer import replace_doctype
from .urls import convert_to_idn, make_safe_absolute_uri
from .urls import make_safe_absolute_uri
from .util import FeedParserDict
# List of preferred XML parsers, by SAX driver name. These will be tried first,
# but if they're not installed, Python will keep searching through its own list
# of pre-installed parsers until it finds one that supports everything we need.
@ -57,27 +51,30 @@ PREFERRED_XML_PARSERS = ["drv_libxml2"]
_XML_AVAILABLE = True
SUPPORTED_VERSIONS = {
'': 'unknown',
'rss090': 'RSS 0.90',
'rss091n': 'RSS 0.91 (Netscape)',
'rss091u': 'RSS 0.91 (Userland)',
'rss092': 'RSS 0.92',
'rss093': 'RSS 0.93',
'rss094': 'RSS 0.94',
'rss20': 'RSS 2.0',
'rss10': 'RSS 1.0',
'rss': 'RSS (unknown version)',
'atom01': 'Atom 0.1',
'atom02': 'Atom 0.2',
'atom03': 'Atom 0.3',
'atom10': 'Atom 1.0',
'atom': 'Atom (unknown version)',
'cdf': 'CDF',
'json1': 'JSON feed 1',
"": "unknown",
"rss090": "RSS 0.90",
"rss091n": "RSS 0.91 (Netscape)",
"rss091u": "RSS 0.91 (Userland)",
"rss092": "RSS 0.92",
"rss093": "RSS 0.93",
"rss094": "RSS 0.94",
"rss20": "RSS 2.0",
"rss10": "RSS 1.0",
"rss": "RSS (unknown version)",
"atom01": "Atom 0.1",
"atom02": "Atom 0.2",
"atom03": "Atom 0.3",
"atom10": "Atom 1.0",
"atom": "Atom (unknown version)",
"cdf": "CDF",
"json1": "JSON feed 1",
}
def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, handlers, request_headers, result):
def _open_resource(
url_file_stream_or_string,
result,
):
"""URL, filename, or string --> stream
This function lets you define parsers that take any input source
@ -86,43 +83,44 @@ def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, h
to have all the basic stdio read methods (read, readline, readlines).
Just .close() the object when you're done with it.
If the etag argument is supplied, it will be used as the value of an
If-None-Match request header.
If the modified argument is supplied, it can be a tuple of 9 integers
(as returned by gmtime() in the standard Python time module) or a date
string in any format supported by feedparser. Regardless, it MUST
be in GMT (Greenwich Mean Time). It will be reformatted into an
RFC 1123-compliant date and used as the value of an If-Modified-Since
request header.
If the agent argument is supplied, it will be used as the value of a
User-Agent request header.
If the referrer argument is supplied, it will be used as the value of a
Referer[sic] request header.
If handlers is supplied, it is a list of handlers used to build a
urllib2 opener.
if request_headers is supplied it is a dictionary of HTTP request headers
that will override the values generated by FeedParser.
:return: A bytes object.
:return: A seekable, readable file object.
"""
if hasattr(url_file_stream_or_string, 'read'):
return url_file_stream_or_string.read()
# Some notes on the history of the implementation of _open_resource().
#
# parse() might need to go over the feed content twice:
# if the strict parser fails, it tries again with the loose parser.
#
# In 5.2.0, this returned an open file, to be read() by parse().
# By 6.0.8, this returned bytes directly.
#
# Since #296 (>6.0.8), this once again returns an open file
# (to reduce memory usage, see convert_file_to_utf8() for details).
# However, to accommodate parse() needing the content twice,
# the returned file is guaranteed to be seekable.
# (If the underlying resource is not seekable,
# the content is read and wrapped in a io.BytesIO/StringIO.)
if isinstance(url_file_stream_or_string, str) \
and urllib.parse.urlparse(url_file_stream_or_string)[0] in ('http', 'https', 'ftp', 'file', 'feed'):
return http.get(url_file_stream_or_string, etag, modified, agent, referrer, handlers, request_headers, result)
if callable(getattr(url_file_stream_or_string, "read", None)):
if callable(getattr(url_file_stream_or_string, "seekable", None)):
if url_file_stream_or_string.seekable():
return url_file_stream_or_string
return _to_in_memory_file(url_file_stream_or_string.read())
looks_like_url = isinstance(
url_file_stream_or_string, str
) and urllib.parse.urlparse(url_file_stream_or_string)[0] in (
"http",
"https",
)
if looks_like_url:
data = http.get(url_file_stream_or_string, result)
return io.BytesIO(data)
# try to open with native open function (if url_file_stream_or_string is a filename)
try:
with open(url_file_stream_or_string, 'rb') as f:
data = f.read()
except (IOError, UnicodeEncodeError, TypeError, ValueError):
return open(url_file_stream_or_string, "rb")
except (OSError, TypeError, ValueError):
# if url_file_stream_or_string is a str object that
# cannot be converted to the encoding returned by
# sys.getfilesystemencoding(), a UnicodeEncodeError
@ -131,33 +129,32 @@ def _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, h
# (such as an XML document encoded in UTF-32), TypeError will
# be thrown.
pass
else:
return data
# treat url_file_stream_or_string as string
if not isinstance(url_file_stream_or_string, bytes):
return url_file_stream_or_string.encode('utf-8')
return url_file_stream_or_string
# treat url_file_stream_or_string as bytes/string
return _to_in_memory_file(url_file_stream_or_string)
def _to_in_memory_file(data):
if isinstance(data, str):
return io.StringIO(data)
else:
return io.BytesIO(data)
class LooseFeedParser(LooseXMLParser, XMLParserMixin, BaseHTMLProcessor):
pass
class StrictFeedParser(StrictXMLParser, XMLParserMixin, xml.sax.handler.ContentHandler):
pass
def parse(
url_file_stream_or_string,
etag: str = None,
modified: Union[str, datetime.datetime, time.struct_time] = None,
agent: str = None,
referrer: str = None,
handlers: List = None,
request_headers: Dict[str, str] = None,
response_headers: Dict[str, str] = None,
resolve_relative_uris: bool = None,
sanitize_html: bool = None,
url_file_stream_or_string,
response_headers: Optional[Dict[str, str]] = None,
resolve_relative_uris: Optional[bool] = None,
sanitize_html: Optional[bool] = None,
optimistic_encoding_detection: Optional[bool] = None,
) -> FeedParserDict:
"""Parse a feed from a URL, file, stream, or string.
@ -174,20 +171,6 @@ def parse(
When a URL is not passed the feed location to use in relative URL
resolution should be passed in the ``Content-Location`` response header
(see ``response_headers`` below).
:param etag:
HTTP ``ETag`` request header.
:param modified:
HTTP ``Last-Modified`` request header.
:param agent:
HTTP ``User-Agent`` request header, which defaults to
the value of :data:`feedparser.USER_AGENT`.
:param referrer:
HTTP ``Referer`` [sic] request header.
:param handlers:
A list of handlers that will be passed to urllib2.
:param request_headers:
A mapping of HTTP header name to HTTP header value to add to the
request, overriding internally generated values.
:param response_headers:
A mapping of HTTP header name to HTTP header value. Multiple values may
be joined with a comma. If a HTTP request was made, these headers
@ -201,20 +184,14 @@ def parse(
Should feedparser skip HTML sanitization? Only disable this if you know
what you are doing! Defaults to the value of
:data:`feedparser.SANITIZE_HTML`, which is ``True``.
:param optimistic_encoding_detection:
Should feedparser use only a prefix of the feed to detect encodings
(uses less memory, but the wrong encoding may be detected in rare cases).
Defaults to the value of
:data:`feedparser.OPTIMISTIC_ENCODING_DETECTION`, which is ``True``.
"""
# Avoid a cyclic import.
if not agent:
import feedparser
agent = feedparser.USER_AGENT
if sanitize_html is None:
import feedparser
sanitize_html = bool(feedparser.SANITIZE_HTML)
if resolve_relative_uris is None:
import feedparser
resolve_relative_uris = bool(feedparser.RESOLVE_RELATIVE_URIS)
result = FeedParserDict(
bozo=False,
entries=[],
@ -223,50 +200,110 @@ def parse(
)
try:
data = _open_resource(url_file_stream_or_string, etag, modified, agent, referrer, handlers, request_headers, result)
file = _open_resource(
url_file_stream_or_string,
result,
)
except urllib.error.URLError as error:
result.update({
'bozo': True,
'bozo_exception': error,
})
result.update(
{
"bozo": True,
"bozo_exception": error,
}
)
return result
if not data:
# at this point, the file is guaranteed to be seekable;
# we read 1 byte/character to see if it's empty and return early
# (this preserves the behavior in 6.0.8)
initial_file_offset = file.tell()
if not file.read(1):
return result
file.seek(initial_file_offset)
# overwrite existing headers using response_headers
result['headers'].update(response_headers or {})
result["headers"].update(response_headers or {})
data = convert_to_utf8(result['headers'], data, result)
use_json_parser = result['content-type'] == 'application/json'
use_strict_parser = result['encoding'] and True or False
try:
_parse_file_inplace(
file,
result,
resolve_relative_uris=resolve_relative_uris,
sanitize_html=sanitize_html,
optimistic_encoding_detection=optimistic_encoding_detection,
)
finally:
if not hasattr(url_file_stream_or_string, "read"):
# the file does not come from the user, close it
file.close()
if not use_json_parser:
result['version'], data, entities = replace_doctype(data)
return result
def _parse_file_inplace(
file: Union[IO[bytes], IO[str]],
result: dict,
*,
resolve_relative_uris: Optional[bool] = None,
sanitize_html: Optional[bool] = None,
optimistic_encoding_detection: Optional[bool] = None,
) -> None:
# Avoid a cyclic import.
import feedparser
if sanitize_html is None:
sanitize_html = bool(feedparser.SANITIZE_HTML)
if resolve_relative_uris is None:
resolve_relative_uris = bool(feedparser.RESOLVE_RELATIVE_URIS)
if optimistic_encoding_detection is None:
optimistic_encoding_detection = bool(feedparser.OPTIMISTIC_ENCODING_DETECTION)
stream_factory = convert_file_to_utf8(
result["headers"], file, result, optimistic_encoding_detection
)
# We're done with file, all access must happen through stream_factory.
del file
# Some notes about the stream_factory.get_{text,binary}_file() methods:
#
# Calling them a second time will raise io.UnsupportedOperation
# if the underlying file was not seekable.
#
# Calling close() on the returned file is ignored
# (that is, the underlying file is *not* closed),
# because the SAX parser closes the file when done;
# we don't want that, since we might try again with the loose parser.
use_json_parser = False
if result["content-type"] in {"application/json", "application/feed+json"}:
use_json_parser = True
use_strict_parser = bool(result["encoding"])
result["version"], stream_factory.prefix, entities = replace_doctype(
stream_factory.prefix
)
# Ensure that baseuri is an absolute URI using an acceptable URI scheme.
contentloc = result['headers'].get('content-location', '')
href = result.get('href', '')
baseuri = make_safe_absolute_uri(href, contentloc) or make_safe_absolute_uri(contentloc) or href
contentloc = result["headers"].get("content-location", "")
href = result.get("href", "")
baseuri = (
make_safe_absolute_uri(href, contentloc)
or make_safe_absolute_uri(contentloc)
or href
)
baselang = result['headers'].get('content-language', None)
baselang = result["headers"].get("content-language", None)
if isinstance(baselang, bytes) and baselang is not None:
baselang = baselang.decode('utf-8', 'ignore')
baselang = baselang.decode("utf-8", "ignore")
if not _XML_AVAILABLE:
use_strict_parser = False
feed_parser: Union[JSONParser, StrictFeedParser, LooseFeedParser]
if use_json_parser:
result['version'] = None
feed_parser = JSONParser(baseuri, baselang, 'utf-8')
try:
feed_parser.feed(data)
except Exception as e:
result['bozo'] = 1
result['bozo_exception'] = e
elif use_strict_parser:
if use_strict_parser and not use_json_parser:
# Initialize the SAX parser.
feed_parser = StrictFeedParser(baseuri, baselang, 'utf-8')
feed_parser = StrictFeedParser(baseuri, baselang, "utf-8")
feed_parser.resolve_relative_uris = resolve_relative_uris
feed_parser.sanitize_html = sanitize_html
saxparser = xml.sax.make_parser(PREFERRED_XML_PARSERS)
@ -279,27 +316,62 @@ def parse(
saxparser.setContentHandler(feed_parser)
saxparser.setErrorHandler(feed_parser)
source = xml.sax.xmlreader.InputSource()
source.setByteStream(io.BytesIO(data))
# If an encoding was detected, decode the file on the fly;
# otherwise, pass it as-is and let the SAX parser deal with it.
try:
source.setCharacterStream(stream_factory.get_text_file())
except MissingEncoding:
source.setByteStream(stream_factory.get_binary_file())
try:
saxparser.parse(source)
except xml.sax.SAXException as e:
result['bozo'] = 1
result['bozo_exception'] = feed_parser.exc or e
result["bozo"] = 1
result["bozo_exception"] = feed_parser.exc or e
use_strict_parser = False
# The loose XML parser will be tried if the JSON parser was not used,
# and if the strict XML parser was not used (or if it failed).
if not use_json_parser and not use_strict_parser:
feed_parser = LooseFeedParser(baseuri, baselang, 'utf-8', entities)
# The loose XML parser will be tried if the strict XML parser was not used
# (or if it failed to parse the feed).
if not use_strict_parser and not use_json_parser:
feed_parser = LooseFeedParser(baseuri, baselang, "utf-8", entities)
feed_parser.resolve_relative_uris = resolve_relative_uris
feed_parser.sanitize_html = sanitize_html
feed_parser.feed(data.decode('utf-8', 'replace'))
result['feed'] = feed_parser.feeddata
result['entries'] = feed_parser.entries
result['version'] = result['version'] or feed_parser.version
# If an encoding was detected, use it; otherwise, assume utf-8 and do your best.
# Will raise io.UnsupportedOperation if the underlying file is not seekable.
data = stream_factory.get_text_file("utf-8", "replace").read()
# As of 6.0.8, LooseFeedParser.feed() can be called exactly once
# with the entire data (it does some re.sub() and str.replace() on it).
#
# SGMLParser (of which LooseFeedParser is a subclass)
# *can* be fed in a streaming fashion,
# by calling feed() repeatedly with chunks of text.
#
# When/if LooseFeedParser will support being fed chunks,
# replace the read() call above with read(size)/feed() calls in a loop.
feed_parser.feed(data)
# If parsing with the loose XML parser resulted in no information,
# flag that the JSON parser should be tried.
if not (feed_parser.entries or feed_parser.feeddata or feed_parser.version):
use_json_parser = True
if use_json_parser:
result["version"] = None
feed_parser = JSONParser(baseuri, baselang, "utf-8")
try:
feed_parser.feed(stream_factory.get_file())
except Exception as e:
result["bozo"] = 1
result["bozo_exception"] = e
result["feed"] = feed_parser.feeddata
result["entries"] = feed_parser.entries
result["version"] = result["version"] or feed_parser.version
if isinstance(feed_parser, JSONParser):
result['namespaces'] = {}
else:
result['namespaces'] = feed_parser.namespaces_in_use
return result
result["namespaces"] = {}
else:
result["namespaces"] = feed_parser.namespaces_in_use

5
lib/feedparser/datetimes/__init__.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -27,11 +27,12 @@
from time import struct_time
from typing import Callable, List, Optional
from .asctime import _parse_date_asctime
from .greek import _parse_date_greek
from .hungarian import _parse_date_hungarian
from .iso8601 import _parse_date_iso8601
from .korean import _parse_date_onblog, _parse_date_nate
from .korean import _parse_date_nate, _parse_date_onblog
from .perforce import _parse_date_perforce
from .rfc822 import _parse_date_rfc822
from .w3dtf import _parse_date_w3dtf

43
lib/feedparser/datetimes/asctime.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -28,18 +28,18 @@
from .rfc822 import _parse_date_rfc822
_months = [
'jan',
'feb',
'mar',
'apr',
'may',
'jun',
'jul',
'aug',
'sep',
'oct',
'nov',
'dec',
"jan",
"feb",
"mar",
"apr",
"may",
"jun",
"jul",
"aug",
"sep",
"oct",
"nov",
"dec",
]
@ -59,13 +59,22 @@ def _parse_date_asctime(dt):
# Insert a GMT timezone, if needed.
if len(parts) == 5:
parts.insert(4, '+0000')
parts.insert(4, "+0000")
# Exit if there are not six parts.
if len(parts) != 6:
return None
# Reassemble the parts in an RFC822-compatible order and parse them.
return _parse_date_rfc822(' '.join([
parts[0], parts[2], parts[1], parts[5], parts[3], parts[4],
]))
return _parse_date_rfc822(
" ".join(
[
parts[0],
parts[2],
parts[1],
parts[5],
parts[3],
parts[4],
]
)
)

82
lib/feedparser/datetimes/greek.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -31,38 +31,40 @@ from .rfc822 import _parse_date_rfc822
# Unicode strings for Greek date strings
_greek_months = {
'\u0399\u03b1\u03bd': 'Jan', # c9e1ed in iso-8859-7
'\u03a6\u03b5\u03b2': 'Feb', # d6e5e2 in iso-8859-7
'\u039c\u03ac\u03ce': 'Mar', # ccdcfe in iso-8859-7
'\u039c\u03b1\u03ce': 'Mar', # cce1fe in iso-8859-7
'\u0391\u03c0\u03c1': 'Apr', # c1f0f1 in iso-8859-7
'\u039c\u03ac\u03b9': 'May', # ccdce9 in iso-8859-7
'\u039c\u03b1\u03ca': 'May', # cce1fa in iso-8859-7
'\u039c\u03b1\u03b9': 'May', # cce1e9 in iso-8859-7
'\u0399\u03bf\u03cd\u03bd': 'Jun', # c9effded in iso-8859-7
'\u0399\u03bf\u03bd': 'Jun', # c9efed in iso-8859-7
'\u0399\u03bf\u03cd\u03bb': 'Jul', # c9effdeb in iso-8859-7
'\u0399\u03bf\u03bb': 'Jul', # c9f9eb in iso-8859-7
'\u0391\u03cd\u03b3': 'Aug', # c1fde3 in iso-8859-7
'\u0391\u03c5\u03b3': 'Aug', # c1f5e3 in iso-8859-7
'\u03a3\u03b5\u03c0': 'Sep', # d3e5f0 in iso-8859-7
'\u039f\u03ba\u03c4': 'Oct', # cfeaf4 in iso-8859-7
'\u039d\u03bf\u03ad': 'Nov', # cdefdd in iso-8859-7
'\u039d\u03bf\u03b5': 'Nov', # cdefe5 in iso-8859-7
'\u0394\u03b5\u03ba': 'Dec', # c4e5ea in iso-8859-7
"\u0399\u03b1\u03bd": "Jan", # c9e1ed in iso-8859-7
"\u03a6\u03b5\u03b2": "Feb", # d6e5e2 in iso-8859-7
"\u039c\u03ac\u03ce": "Mar", # ccdcfe in iso-8859-7
"\u039c\u03b1\u03ce": "Mar", # cce1fe in iso-8859-7
"\u0391\u03c0\u03c1": "Apr", # c1f0f1 in iso-8859-7
"\u039c\u03ac\u03b9": "May", # ccdce9 in iso-8859-7
"\u039c\u03b1\u03ca": "May", # cce1fa in iso-8859-7
"\u039c\u03b1\u03b9": "May", # cce1e9 in iso-8859-7
"\u0399\u03bf\u03cd\u03bd": "Jun", # c9effded in iso-8859-7
"\u0399\u03bf\u03bd": "Jun", # c9efed in iso-8859-7
"\u0399\u03bf\u03cd\u03bb": "Jul", # c9effdeb in iso-8859-7
"\u0399\u03bf\u03bb": "Jul", # c9f9eb in iso-8859-7
"\u0391\u03cd\u03b3": "Aug", # c1fde3 in iso-8859-7
"\u0391\u03c5\u03b3": "Aug", # c1f5e3 in iso-8859-7
"\u03a3\u03b5\u03c0": "Sep", # d3e5f0 in iso-8859-7
"\u039f\u03ba\u03c4": "Oct", # cfeaf4 in iso-8859-7
"\u039d\u03bf\u03ad": "Nov", # cdefdd in iso-8859-7
"\u039d\u03bf\u03b5": "Nov", # cdefe5 in iso-8859-7
"\u0394\u03b5\u03ba": "Dec", # c4e5ea in iso-8859-7
}
_greek_wdays = {
'\u039a\u03c5\u03c1': 'Sun', # caf5f1 in iso-8859-7
'\u0394\u03b5\u03c5': 'Mon', # c4e5f5 in iso-8859-7
'\u03a4\u03c1\u03b9': 'Tue', # d4f1e9 in iso-8859-7
'\u03a4\u03b5\u03c4': 'Wed', # d4e5f4 in iso-8859-7
'\u03a0\u03b5\u03bc': 'Thu', # d0e5ec in iso-8859-7
'\u03a0\u03b1\u03c1': 'Fri', # d0e1f1 in iso-8859-7
'\u03a3\u03b1\u03b2': 'Sat', # d3e1e2 in iso-8859-7
"\u039a\u03c5\u03c1": "Sun", # caf5f1 in iso-8859-7
"\u0394\u03b5\u03c5": "Mon", # c4e5f5 in iso-8859-7
"\u03a4\u03c1\u03b9": "Tue", # d4f1e9 in iso-8859-7
"\u03a4\u03b5\u03c4": "Wed", # d4e5f4 in iso-8859-7
"\u03a0\u03b5\u03bc": "Thu", # d0e5ec in iso-8859-7
"\u03a0\u03b1\u03c1": "Fri", # d0e1f1 in iso-8859-7
"\u03a3\u03b1\u03b2": "Sat", # d3e1e2 in iso-8859-7
}
_greek_date_format_re = re.compile(r'([^,]+),\s+(\d{2})\s+([^\s]+)\s+(\d{4})\s+(\d{2}):(\d{2}):(\d{2})\s+([^\s]+)')
_greek_date_format_re = re.compile(
r"([^,]+),\s+(\d{2})\s+([^\s]+)\s+(\d{4})\s+(\d{2}):(\d{2}):(\d{2})\s+([^\s]+)"
)
def _parse_date_greek(date_string):
@ -72,15 +74,17 @@ def _parse_date_greek(date_string):
return
wday = _greek_wdays[m.group(1)]
month = _greek_months[m.group(3)]
rfc822date = '%(wday)s, %(day)s %(month)s %(year)s %(hour)s:%(minute)s:%(second)s %(zonediff)s' % \
{
'wday': wday,
'day': m.group(2),
'month': month,
'year': m.group(4),
'hour': m.group(5),
'minute': m.group(6),
'second': m.group(7),
'zonediff': m.group(8),
}
rfc822date = (
"%(wday)s, %(day)s %(month)s %(year)s %(hour)s:%(minute)s:%(second)s %(offset)s"
% {
"wday": wday,
"day": m.group(2),
"month": month,
"year": m.group(4),
"hour": m.group(5),
"minute": m.group(6),
"second": m.group(7),
"offset": m.group(8),
}
)
return _parse_date_rfc822(rfc822date)

44
lib/feedparser/datetimes/hungarian.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -31,21 +31,23 @@ from .w3dtf import _parse_date_w3dtf
# Unicode strings for Hungarian date strings
_hungarian_months = {
'janu\u00e1r': '01', # e1 in iso-8859-2
'febru\u00e1ri': '02', # e1 in iso-8859-2
'm\u00e1rcius': '03', # e1 in iso-8859-2
'\u00e1prilis': '04', # e1 in iso-8859-2
'm\u00e1ujus': '05', # e1 in iso-8859-2
'j\u00fanius': '06', # fa in iso-8859-2
'j\u00falius': '07', # fa in iso-8859-2
'augusztus': '08',
'szeptember': '09',
'okt\u00f3ber': '10', # f3 in iso-8859-2
'november': '11',
'december': '12',
"janu\u00e1r": "01", # e1 in iso-8859-2
"febru\u00e1ri": "02", # e1 in iso-8859-2
"m\u00e1rcius": "03", # e1 in iso-8859-2
"\u00e1prilis": "04", # e1 in iso-8859-2
"m\u00e1ujus": "05", # e1 in iso-8859-2
"j\u00fanius": "06", # fa in iso-8859-2
"j\u00falius": "07", # fa in iso-8859-2
"augusztus": "08",
"szeptember": "09",
"okt\u00f3ber": "10", # f3 in iso-8859-2
"november": "11",
"december": "12",
}
_hungarian_date_format_re = re.compile(r'(\d{4})-([^-]+)-(\d{,2})T(\d{,2}):(\d{2})([+-](\d{,2}:\d{2}))')
_hungarian_date_format_re = re.compile(
r"(\d{4})-([^-]+)-(\d{,2})T(\d{,2}):(\d{2})([+-](\d{,2}:\d{2}))"
)
def _parse_date_hungarian(date_string):
@ -56,17 +58,9 @@ def _parse_date_hungarian(date_string):
month = _hungarian_months[m.group(2)]
day = m.group(3)
if len(day) == 1:
day = '0' + day
day = "0" + day
hour = m.group(4)
if len(hour) == 1:
hour = '0' + hour
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s%(zonediff)s' % \
{
'year': m.group(1),
'month': month,
'day': day,
'hour': hour,
'minute': m.group(5),
'zonediff': m.group(6),
}
hour = "0" + hour
w3dtfdate = f"{m.group(1)}-{month}-{day}T{hour}:{m.group(5)}{m.group(6)}"
return _parse_date_w3dtf(w3dtfdate)

114
lib/feedparser/datetimes/iso8601.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -38,36 +38,36 @@ import time
# Please note the order in templates is significant because we need a
# greedy match.
_iso8601_tmpl = [
'YYYY-?MM-?DD',
'YYYY-0MM?-?DD',
'YYYY-MM',
'YYYY-?OOO',
'YY-?MM-?DD',
'YY-?OOO',
'YYYY',
'-YY-?MM',
'-OOO',
'-YY',
'--MM-?DD',
'--MM',
'---DD',
'CC',
'',
"YYYY-?MM-?DD",
"YYYY-0MM?-?DD",
"YYYY-MM",
"YYYY-?OOO",
"YY-?MM-?DD",
"YY-?OOO",
"YYYY",
"-YY-?MM",
"-OOO",
"-YY",
"--MM-?DD",
"--MM",
"---DD",
"CC",
"",
]
_iso8601_re = [
tmpl.replace(
'YYYY', r'(?P<year>\d{4})').replace(
'YY', r'(?P<year>\d\d)').replace(
'MM', r'(?P<month>[01]\d)').replace(
'DD', r'(?P<day>[0123]\d)').replace(
'OOO', r'(?P<ordinal>[0123]\d\d)').replace(
'CC', r'(?P<century>\d\d$)')
+ r'(T?(?P<hour>\d{2}):(?P<minute>\d{2})'
+ r'(:(?P<second>\d{2}))?'
+ r'(\.(?P<fracsecond>\d+))?'
+ r'(?P<tz>[+-](?P<tzhour>\d{2})(:(?P<tzmin>\d{2}))?|Z)?)?'
for tmpl in _iso8601_tmpl]
tmpl.replace("YYYY", r"(?P<year>\d{4})")
.replace("YY", r"(?P<year>\d\d)")
.replace("MM", r"(?P<month>[01]\d)")
.replace("DD", r"(?P<day>[0123]\d)")
.replace("OOO", r"(?P<ordinal>[0123]\d\d)")
.replace("CC", r"(?P<century>\d\d$)")
+ r"(T?(?P<hour>\d{2}):(?P<minute>\d{2})"
+ r"(:(?P<second>\d{2}))?"
+ r"(\.(?P<fracsecond>\d+))?"
+ r"(?P<tz>[+-](?P<tzhour>\d{2})(:(?P<tzmin>\d{2}))?|Z)?)?"
for tmpl in _iso8601_tmpl
]
_iso8601_matches = [re.compile(regex).match for regex in _iso8601_re]
@ -83,21 +83,21 @@ def _parse_date_iso8601(date_string):
if m.span() == (0, 0):
return
params = m.groupdict()
ordinal = params.get('ordinal', 0)
ordinal = params.get("ordinal", 0)
if ordinal:
ordinal = int(ordinal)
else:
ordinal = 0
year = params.get('year', '--')
if not year or year == '--':
year = params.get("year", "--")
if not year or year == "--":
year = time.gmtime()[0]
elif len(year) == 2:
# ISO 8601 assumes current century, i.e. 93 -> 2093, NOT 1993
year = 100 * int(time.gmtime()[0] / 100) + int(year)
else:
year = int(year)
month = params.get('month', '-')
if not month or month == '-':
month = params.get("month", "-")
if not month or month == "-":
# ordinals are NOT normalized by mktime, we simulate them
# by setting month=1, day=ordinal
if ordinal:
@ -105,13 +105,14 @@ def _parse_date_iso8601(date_string):
else:
month = time.gmtime()[1]
month = int(month)
day = params.get('day', 0)
day = params.get("day", 0)
if not day:
# see above
if ordinal:
day = ordinal
elif params.get('century', 0) or \
params.get('year', 0) or params.get('month', 0):
elif (
params.get("century", 0) or params.get("year", 0) or params.get("month", 0)
):
day = 1
else:
day = time.gmtime()[2]
@ -119,29 +120,38 @@ def _parse_date_iso8601(date_string):
day = int(day)
# special case of the century - is the first year of the 21st century
# 2000 or 2001 ? The debate goes on...
if 'century' in params:
year = (int(params['century']) - 1) * 100 + 1
if "century" in params:
year = (int(params["century"]) - 1) * 100 + 1
# in ISO 8601 most fields are optional
for field in ['hour', 'minute', 'second', 'tzhour', 'tzmin']:
for field in ["hour", "minute", "second", "tzhour", "tzmin"]:
if not params.get(field, None):
params[field] = 0
hour = int(params.get('hour', 0))
minute = int(params.get('minute', 0))
second = int(float(params.get('second', 0)))
hour = int(params.get("hour", 0))
minute = int(params.get("minute", 0))
second = int(float(params.get("second", 0)))
# weekday is normalized by mktime(), we can ignore it
weekday = 0
daylight_savings_flag = -1
tm = [year, month, day, hour, minute, second, weekday,
ordinal, daylight_savings_flag]
tm = [
year,
month,
day,
hour,
minute,
second,
weekday,
ordinal,
daylight_savings_flag,
]
# ISO 8601 time zone adjustments
tz = params.get('tz')
if tz and tz != 'Z':
if tz[0] == '-':
tm[3] += int(params.get('tzhour', 0))
tm[4] += int(params.get('tzmin', 0))
elif tz[0] == '+':
tm[3] -= int(params.get('tzhour', 0))
tm[4] -= int(params.get('tzmin', 0))
tz = params.get("tz")
if tz and tz != "Z":
if tz[0] == "-":
tm[3] += int(params.get("tzhour", 0))
tm[4] += int(params.get("tzmin", 0))
elif tz[0] == "+":
tm[3] -= int(params.get("tzhour", 0))
tm[4] -= int(params.get("tzmin", 0))
else:
return None
# Python's time.mktime() is a wrapper around the ANSI C mktime(3c)

59
lib/feedparser/datetimes/korean.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -30,20 +30,21 @@ import re
from .w3dtf import _parse_date_w3dtf
# 8-bit date handling routines written by ytrewq1.
_korean_year = '\ub144' # b3e2 in euc-kr
_korean_month = '\uc6d4' # bff9 in euc-kr
_korean_day = '\uc77c' # c0cf in euc-kr
_korean_am = '\uc624\uc804' # bfc0 c0fc in euc-kr
_korean_pm = '\uc624\ud6c4' # bfc0 c8c4 in euc-kr
_korean_year = "\ub144" # b3e2 in euc-kr
_korean_month = "\uc6d4" # bff9 in euc-kr
_korean_day = "\uc77c" # c0cf in euc-kr
_korean_am = "\uc624\uc804" # bfc0 c0fc in euc-kr
_korean_pm = "\uc624\ud6c4" # bfc0 c8c4 in euc-kr
_korean_onblog_date_re = re.compile(
r'(\d{4})%s\s+(\d{2})%s\s+(\d{2})%s\s+(\d{2}):(\d{2}):(\d{2})'
r"(\d{4})%s\s+(\d{2})%s\s+(\d{2})%s\s+(\d{2}):(\d{2}):(\d{2})"
% (_korean_year, _korean_month, _korean_day)
)
_korean_nate_date_re = re.compile(
r'(\d{4})-(\d{2})-(\d{2})\s+(%s|%s)\s+(\d{,2}):(\d{,2}):(\d{,2})'
% (_korean_am, _korean_pm))
r"(\d{4})-(\d{2})-(\d{2})\s+(%s|%s)\s+(\d{,2}):(\d{,2}):(\d{,2})"
% (_korean_am, _korean_pm)
)
def _parse_date_onblog(dateString):
@ -51,10 +52,18 @@ def _parse_date_onblog(dateString):
m = _korean_onblog_date_re.match(dateString)
if not m:
return
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \
{'year': m.group(1), 'month': m.group(2), 'day': m.group(3),
'hour': m.group(4), 'minute': m.group(5), 'second': m.group(6),
'zonediff': '+09:00'}
w3dtfdate = (
"%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s"
% {
"year": m.group(1),
"month": m.group(2),
"day": m.group(3),
"hour": m.group(4),
"minute": m.group(5),
"second": m.group(6),
"zonediff": "+09:00",
}
)
return _parse_date_w3dtf(w3dtfdate)
@ -69,15 +78,17 @@ def _parse_date_nate(dateString):
hour += 12
hour = str(hour)
if len(hour) == 1:
hour = '0' + hour
w3dtfdate = '%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s' % \
{
'year': m.group(1),
'month': m.group(2),
'day': m.group(3),
'hour': hour,
'minute': m.group(6),
'second': m.group(7),
'zonediff': '+09:00',
}
hour = "0" + hour
w3dtfdate = (
"%(year)s-%(month)s-%(day)sT%(hour)s:%(minute)s:%(second)s%(zonediff)s"
% {
"year": m.group(1),
"month": m.group(2),
"day": m.group(3),
"hour": hour,
"minute": m.group(6),
"second": m.group(7),
"zonediff": "+09:00",
}
)
return _parse_date_w3dtf(w3dtfdate)

25
lib/feedparser/datetimes/perforce.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -33,14 +33,31 @@ import time
def _parse_date_perforce(date_string):
"""parse a date in yyyy/mm/dd hh:mm:ss TTT format"""
# Fri, 2006/09/15 08:19:53 EDT
_my_date_pattern = re.compile(r'(\w{,3}), (\d{,4})/(\d{,2})/(\d{2}) (\d{,2}):(\d{2}):(\d{2}) (\w{,3})')
_my_date_pattern = re.compile(
r"(\w{,3}), (\d{,4})/(\d{,2})/(\d{2}) (\d{,2}):(\d{2}):(\d{2}) (\w{,3})"
)
m = _my_date_pattern.search(date_string)
if m is None:
return None
dow, year, month, day, hour, minute, second, tz = m.groups()
months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
new_date_string = "%s, %s %s %s %s:%s:%s %s" % (dow, day, months[int(month) - 1], year, hour, minute, second, tz)
months = [
"Jan",
"Feb",
"Mar",
"Apr",
"May",
"Jun",
"Jul",
"Aug",
"Sep",
"Oct",
"Nov",
"Dec",
]
new_date_string = (
f"{dow}, {day} {months[int(month) - 1]} {year} {hour}:{minute}:{second} {tz}"
)
tm = email.utils.parsedate_tz(new_date_string)
if tm:
return time.gmtime(email.utils.mktime_tz(tm))

71
lib/feedparser/datetimes/rfc822.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -28,20 +28,45 @@
import datetime
timezone_names = {
'ut': 0, 'gmt': 0, 'z': 0,
'adt': -3, 'ast': -4, 'at': -4,
'edt': -4, 'est': -5, 'et': -5,
'cdt': -5, 'cst': -6, 'ct': -6,
'mdt': -6, 'mst': -7, 'mt': -7,
'pdt': -7, 'pst': -8, 'pt': -8,
'a': -1, 'n': 1,
'm': -12, 'y': 12,
'met': 1, 'mest': 2,
"ut": 0,
"gmt": 0,
"z": 0,
"adt": -3,
"ast": -4,
"at": -4,
"edt": -4,
"est": -5,
"et": -5,
"cdt": -5,
"cst": -6,
"ct": -6,
"mdt": -6,
"mst": -7,
"mt": -7,
"pdt": -7,
"pst": -8,
"pt": -8,
"a": -1,
"n": 1,
"m": -12,
"y": 12,
"met": 1,
"mest": 2,
}
day_names = {'mon', 'tue', 'wed', 'thu', 'fri', 'sat', 'sun'}
day_names = {"mon", "tue", "wed", "thu", "fri", "sat", "sun"}
months = {
'jan': 1, 'feb': 2, 'mar': 3, 'apr': 4, 'may': 5, 'jun': 6,
'jul': 7, 'aug': 8, 'sep': 9, 'oct': 10, 'nov': 11, 'dec': 12,
"jan": 1,
"feb": 2,
"mar": 3,
"apr": 4,
"may": 5,
"jun": 6,
"jul": 7,
"aug": 8,
"sep": 9,
"oct": 10,
"nov": 11,
"dec": 12,
}
@ -63,7 +88,7 @@ def _parse_date_rfc822(date):
parts = date.lower().split()
if len(parts) < 5:
# Assume that the time and timezone are missing
parts.extend(('00:00:00', '0000'))
parts.extend(("00:00:00", "0000"))
# Remove the day name
if parts[0][:3] in day_names:
parts = parts[1:]
@ -101,26 +126,26 @@ def _parse_date_rfc822(date):
year += (1900, 2000)[year < 90]
# Handle the time (default to 00:00:00).
time_parts = parts[3].split(':')
time_parts.extend(('0',) * (3 - len(time_parts)))
time_parts = parts[3].split(":")
time_parts.extend(("0",) * (3 - len(time_parts)))
try:
(hour, minute, second) = [int(i) for i in time_parts]
(hour, minute, second) = (int(i) for i in time_parts)
except ValueError:
return None
# Handle the timezone information, if any (default to +0000).
# Strip 'Etc/' from the timezone.
if parts[4].startswith('etc/'):
if parts[4].startswith("etc/"):
parts[4] = parts[4][4:]
# Normalize timezones that start with 'gmt':
# GMT-05:00 => -0500
# GMT => GMT
if parts[4].startswith('gmt'):
parts[4] = ''.join(parts[4][3:].split(':')) or 'gmt'
if parts[4].startswith("gmt"):
parts[4] = "".join(parts[4][3:].split(":")) or "gmt"
# Handle timezones like '-0500', '+0500', and 'EST'
if parts[4] and parts[4][0] in ('-', '+'):
if parts[4] and parts[4][0] in ("-", "+"):
try:
if ':' in parts[4]:
if ":" in parts[4]:
timezone_hours = int(parts[4][1:3])
timezone_minutes = int(parts[4][4:])
else:
@ -128,7 +153,7 @@ def _parse_date_rfc822(date):
timezone_minutes = int(parts[4][3:])
except ValueError:
return None
if parts[4].startswith('-'):
if parts[4].startswith("-"):
timezone_hours *= -1
timezone_minutes *= -1
else:

58
lib/feedparser/datetimes/w3dtf.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -28,14 +28,28 @@
import datetime
timezonenames = {
'ut': 0, 'gmt': 0, 'z': 0,
'adt': -3, 'ast': -4, 'at': -4,
'edt': -4, 'est': -5, 'et': -5,
'cdt': -5, 'cst': -6, 'ct': -6,
'mdt': -6, 'mst': -7, 'mt': -7,
'pdt': -7, 'pst': -8, 'pt': -8,
'a': -1, 'n': 1,
'm': -12, 'y': 12,
"ut": 0,
"gmt": 0,
"z": 0,
"adt": -3,
"ast": -4,
"at": -4,
"edt": -4,
"est": -5,
"et": -5,
"cdt": -5,
"cst": -6,
"ct": -6,
"mdt": -6,
"mst": -7,
"mt": -7,
"pdt": -7,
"pst": -8,
"pt": -8,
"a": -1,
"n": 1,
"m": -12,
"y": 12,
}
# W3 date and time format parser
# http://www.w3.org/TR/NOTE-datetime
@ -47,57 +61,57 @@ timezonenames = {
def _parse_date_w3dtf(datestr):
if not datestr.strip():
return None
parts = datestr.lower().split('t')
parts = datestr.lower().split("t")
if len(parts) == 1:
# This may be a date only, or may be an MSSQL-style date
parts = parts[0].split()
if len(parts) == 1:
# Treat this as a date only
parts.append('00:00:00z')
parts.append("00:00:00z")
elif len(parts) > 2:
return None
date = parts[0].split('-', 2)
date = parts[0].split("-", 2)
if not date or len(date[0]) != 4:
return None
# Ensure that `date` has 3 elements. Using '1' sets the default
# month to January and the default day to the 1st of the month.
date.extend(['1'] * (3 - len(date)))
date.extend(["1"] * (3 - len(date)))
try:
year, month, day = [int(i) for i in date]
year, month, day = (int(i) for i in date)
except ValueError:
# `date` may have more than 3 elements or may contain
# non-integer strings.
return None
if parts[1].endswith('z'):
if parts[1].endswith("z"):
parts[1] = parts[1][:-1]
parts.append('z')
parts.append("z")
# Append the numeric timezone offset, if any, to parts.
# If this is an MSSQL-style date then parts[2] already contains
# the timezone information, so `append()` will not affect it.
# Add 1 to each value so that if `find()` returns -1 it will be
# treated as False.
loc = parts[1].find('-') + 1 or parts[1].find('+') + 1 or len(parts[1]) + 1
loc = parts[1].find("-") + 1 or parts[1].find("+") + 1 or len(parts[1]) + 1
loc = loc - 1
parts.append(parts[1][loc:])
parts[1] = parts[1][:loc]
time = parts[1].split(':', 2)
time = parts[1].split(":", 2)
# Ensure that time has 3 elements. Using '0' means that the
# minutes and seconds, if missing, will default to 0.
time.extend(['0'] * (3 - len(time)))
if parts[2][:1] in ('-', '+'):
time.extend(["0"] * (3 - len(time)))
if parts[2][:1] in ("-", "+"):
try:
tzhour = int(parts[2][1:3])
tzmin = int(parts[2][4:])
except ValueError:
return None
if parts[2].startswith('-'):
if parts[2].startswith("-"):
tzhour = tzhour * -1
tzmin = tzmin * -1
else:
tzhour = timezonenames.get(parts[2], 0)
tzmin = 0
try:
hour, minute, second = [int(float(i)) for i in time]
hour, minute, second = (int(float(i)) for i in time)
except ValueError:
return None
# Create the datetime object and timezone delta objects

558
lib/feedparser/encodings.py
View file

@ -1,5 +1,5 @@
# Character encoding routines
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -26,48 +26,53 @@
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
from __future__ import annotations
import codecs
import io
import re
import typing as t
import typing
try:
try:
import cchardet as chardet # type: ignore[import]
import cchardet as chardet # type: ignore[import]
except ImportError:
import chardet # type: ignore[no-redef]
import chardet # type: ignore[no-redef]
except ImportError:
lazy_chardet_encoding = None
else:
def lazy_chardet_encoding(data):
return chardet.detect(data)['encoding'] or ''
return chardet.detect(data)["encoding"] or ""
from .exceptions import (
CharacterEncodingOverride,
CharacterEncodingUnknown,
FeedparserError,
NonXMLContentType,
)
# Each marker represents some of the characters of the opening XML
# processing instruction ('<?xm') in the specified encoding.
EBCDIC_MARKER = b'\x4C\x6F\xA7\x94'
UTF16BE_MARKER = b'\x00\x3C\x00\x3F'
UTF16LE_MARKER = b'\x3C\x00\x3F\x00'
UTF32BE_MARKER = b'\x00\x00\x00\x3C'
UTF32LE_MARKER = b'\x3C\x00\x00\x00'
EBCDIC_MARKER = b"\x4C\x6F\xA7\x94"
UTF16BE_MARKER = b"\x00\x3C\x00\x3F"
UTF16LE_MARKER = b"\x3C\x00\x3F\x00"
UTF32BE_MARKER = b"\x00\x00\x00\x3C"
UTF32LE_MARKER = b"\x3C\x00\x00\x00"
ZERO_BYTES = '\x00\x00'
ZERO_BYTES = b"\x00\x00"
# Match the opening XML declaration.
# Example: <?xml version="1.0" encoding="utf-8"?>
RE_XML_DECLARATION = re.compile(r'^<\?xml[^>]*?>')
RE_XML_DECLARATION = re.compile(r"^<\?xml[^>]*?>")
# Capture the value of the XML processing instruction's encoding attribute.
# Example: <?xml version="1.0" encoding="utf-8"?>
RE_XML_PI_ENCODING = re.compile(br'^<\?.*encoding=[\'"](.*?)[\'"].*\?>')
RE_XML_PI_ENCODING = re.compile(rb'^<\?.*encoding=[\'"](.*?)[\'"].*\?>')
def parse_content_type(line: str) -> t.Tuple[str, str]:
def parse_content_type(line: str) -> tuple[str, str]:
"""Parse an HTTP Content-Type header.
The return value will be a tuple of strings:
@ -91,11 +96,10 @@ def parse_content_type(line: str) -> t.Tuple[str, str]:
return mime_type, charset_value
def convert_to_utf8(http_headers, data, result):
"""Detect and convert the character encoding to UTF-8.
http_headers is a dictionary
data is a raw string (not Unicode)"""
def convert_to_utf8(
http_headers: dict[str, str], data: bytes, result: dict[str, typing.Any]
) -> bytes:
"""Detect and convert the character encoding to UTF-8."""
# This is so much trickier than it sounds, it's not even funny.
# According to RFC 3023 ('XML Media Types'), if the HTTP Content-Type
@ -134,12 +138,10 @@ def convert_to_utf8(http_headers, data, result):
# Of course, none of this guarantees that we will be able to parse the
# feed in the declared character encoding (assuming it was declared
# correctly, which many are not). iconv_codec can help a lot;
# you should definitely install it if you can.
# http://cjkpython.i18n.org/
# correctly, which many are not).
bom_encoding = ''
xml_encoding = ''
bom_encoding = ""
xml_encoding = ""
# Look at the first few bytes of the document to guess what
# its encoding may be. We only need to decode enough of the
@ -149,50 +151,63 @@ def convert_to_utf8(http_headers, data, result):
# http://www.w3.org/TR/REC-xml/#sec-guessing-no-ext-info
# Check for BOMs first.
if data[:4] == codecs.BOM_UTF32_BE:
bom_encoding = 'utf-32be'
bom_encoding = "utf-32be"
data = data[4:]
elif data[:4] == codecs.BOM_UTF32_LE:
bom_encoding = 'utf-32le'
bom_encoding = "utf-32le"
data = data[4:]
elif data[:2] == codecs.BOM_UTF16_BE and data[2:4] != ZERO_BYTES:
bom_encoding = 'utf-16be'
bom_encoding = "utf-16be"
data = data[2:]
elif data[:2] == codecs.BOM_UTF16_LE and data[2:4] != ZERO_BYTES:
bom_encoding = 'utf-16le'
bom_encoding = "utf-16le"
data = data[2:]
elif data[:3] == codecs.BOM_UTF8:
bom_encoding = 'utf-8'
bom_encoding = "utf-8"
data = data[3:]
# Check for the characters '<?xm' in several encodings.
elif data[:4] == EBCDIC_MARKER:
bom_encoding = 'cp037'
bom_encoding = "cp037"
elif data[:4] == UTF16BE_MARKER:
bom_encoding = 'utf-16be'
bom_encoding = "utf-16be"
elif data[:4] == UTF16LE_MARKER:
bom_encoding = 'utf-16le'
bom_encoding = "utf-16le"
elif data[:4] == UTF32BE_MARKER:
bom_encoding = 'utf-32be'
bom_encoding = "utf-32be"
elif data[:4] == UTF32LE_MARKER:
bom_encoding = 'utf-32le'
bom_encoding = "utf-32le"
tempdata = data
try:
if bom_encoding:
tempdata = data.decode(bom_encoding).encode('utf-8')
tempdata = data.decode(bom_encoding).encode("utf-8")
except UnicodeDecodeError:
xml_encoding_match = None
else:
xml_encoding_match = RE_XML_PI_ENCODING.match(tempdata)
if xml_encoding_match:
xml_encoding = xml_encoding_match.groups()[0].decode('utf-8').lower()
xml_encoding = xml_encoding_match.groups()[0].decode("utf-8").lower()
# Normalize the xml_encoding if necessary.
if bom_encoding and (xml_encoding in (
'u16', 'utf-16', 'utf16', 'utf_16',
'u32', 'utf-32', 'utf32', 'utf_32',
'iso-10646-ucs-2', 'iso-10646-ucs-4',
'csucs4', 'csunicode', 'ucs-2', 'ucs-4'
)):
if bom_encoding and (
xml_encoding
in (
"u16",
"utf-16",
"utf16",
"utf_16",
"u32",
"utf-32",
"utf32",
"utf_32",
"iso-10646-ucs-2",
"iso-10646-ucs-4",
"csucs4",
"csunicode",
"ucs-2",
"ucs-4",
)
):
xml_encoding = bom_encoding
# Find the HTTP Content-Type and, hopefully, a character
@ -200,115 +215,436 @@ def convert_to_utf8(http_headers, data, result):
# to choose the "correct" encoding among the BOM encoding,
# XML declaration encoding, and HTTP encoding, following the
# heuristic defined in RFC 3023.
http_content_type = http_headers.get('content-type') or ''
http_content_type = http_headers.get("content-type") or ""
http_content_type, http_encoding = parse_content_type(http_content_type)
acceptable_content_type = 0
application_content_types = ('application/xml', 'application/xml-dtd',
'application/xml-external-parsed-entity')
text_content_types = ('text/xml', 'text/xml-external-parsed-entity')
json_content_types = ('application/feed+json', 'application/json')
application_content_types = (
"application/xml",
"application/xml-dtd",
"application/xml-external-parsed-entity",
)
text_content_types = ("text/xml", "text/xml-external-parsed-entity")
json_content_types = ("application/feed+json", "application/json")
json = False
if (
http_content_type in application_content_types
or (
http_content_type.startswith('application/')
and http_content_type.endswith('+xml')
)
if http_content_type in application_content_types or (
http_content_type.startswith("application/")
and http_content_type.endswith("+xml")
):
acceptable_content_type = 1
rfc3023_encoding = http_encoding or xml_encoding or 'utf-8'
elif (
http_content_type in text_content_types
or (
http_content_type.startswith('text/')
and http_content_type.endswith('+xml')
)
rfc3023_encoding = http_encoding or xml_encoding or "utf-8"
elif http_content_type in text_content_types or (
http_content_type.startswith("text/") and http_content_type.endswith("+xml")
):
acceptable_content_type = 1
rfc3023_encoding = http_encoding or 'us-ascii'
elif (
http_content_type in json_content_types
or (
not http_content_type
and data and data.lstrip()[0] == '{'
)
rfc3023_encoding = http_encoding or "us-ascii"
elif http_content_type in json_content_types or (
not http_content_type and data and data.lstrip().startswith(b"{")
):
http_content_type = json_content_types[0]
acceptable_content_type = 1
json = True
rfc3023_encoding = http_encoding or 'utf-8' # RFC 7159, 8.1.
elif http_content_type.startswith('text/'):
rfc3023_encoding = http_encoding or 'us-ascii'
elif http_headers and 'content-type' not in http_headers:
rfc3023_encoding = xml_encoding or 'iso-8859-1'
rfc3023_encoding = http_encoding or "utf-8" # RFC 7159, 8.1.
elif http_content_type.startswith("text/"):
rfc3023_encoding = http_encoding or "us-ascii"
elif http_headers and "content-type" not in http_headers:
rfc3023_encoding = xml_encoding or "iso-8859-1"
else:
rfc3023_encoding = xml_encoding or 'utf-8'
rfc3023_encoding = xml_encoding or "utf-8"
# gb18030 is a superset of gb2312, so always replace gb2312
# with gb18030 for greater compatibility.
if rfc3023_encoding.lower() == 'gb2312':
rfc3023_encoding = 'gb18030'
if xml_encoding.lower() == 'gb2312':
xml_encoding = 'gb18030'
if rfc3023_encoding.lower() == "gb2312":
rfc3023_encoding = "gb18030"
if xml_encoding.lower() == "gb2312":
xml_encoding = "gb18030"
# there are four encodings to keep track of:
# - http_encoding is the encoding declared in the Content-Type HTTP header
# - xml_encoding is the encoding declared in the <?xml declaration
# - bom_encoding is the encoding sniffed from the first 4 bytes of the XML data
# - rfc3023_encoding is the actual encoding, as per RFC 3023 and a variety of other conflicting specifications
error = None
# - rfc3023_encoding is the actual encoding, as per RFC 3023
# and a variety of other conflicting specifications
error: FeedparserError | None = None
if http_headers and (not acceptable_content_type):
if 'content-type' in http_headers:
msg = '%s is not an accepted media type' % http_headers['content-type']
if "content-type" in http_headers:
msg = "%s is not an accepted media type" % http_headers["content-type"]
else:
msg = 'no Content-type specified'
msg = "no Content-type specified"
error = NonXMLContentType(msg)
# determine character encoding
known_encoding = 0
known_encoding = False
tried_encodings = []
# try: HTTP encoding, declared XML encoding, encoding sniffed from BOM
for proposed_encoding in (rfc3023_encoding, xml_encoding, bom_encoding,
lazy_chardet_encoding, 'utf-8', 'windows-1252', 'iso-8859-2'):
if callable(proposed_encoding):
proposed_encoding = proposed_encoding(data)
for encoding_to_try in (
rfc3023_encoding,
xml_encoding,
bom_encoding,
lazy_chardet_encoding,
"utf-8",
"windows-1252",
"iso-8859-2",
):
if callable(encoding_to_try):
proposed_encoding = encoding_to_try(data)
else:
proposed_encoding = encoding_to_try
if not proposed_encoding:
continue
if proposed_encoding in tried_encodings:
continue
tried_encodings.append(proposed_encoding)
try:
data = data.decode(proposed_encoding)
text = data.decode(proposed_encoding)
except (UnicodeDecodeError, LookupError):
pass
else:
known_encoding = 1
if not json:
# Update the encoding in the opening XML processing instruction.
new_declaration = '''<?xml version='1.0' encoding='utf-8'?>'''
if RE_XML_DECLARATION.search(data):
data = RE_XML_DECLARATION.sub(new_declaration, data)
else:
data = new_declaration + '\n' + data
data = data.encode('utf-8')
break
continue
known_encoding = True
if not json:
# Update the encoding in the opening XML processing instruction.
new_declaration = """<?xml version='1.0' encoding='utf-8'?>"""
if RE_XML_DECLARATION.search(text):
text = RE_XML_DECLARATION.sub(new_declaration, text)
else:
text = new_declaration + "\n" + text
data = text.encode("utf-8")
break
# if still no luck, give up
if not known_encoding:
error = CharacterEncodingUnknown(
'document encoding unknown, I tried ' +
'%s, %s, utf-8, windows-1252, and iso-8859-2 but nothing worked' %
(rfc3023_encoding, xml_encoding))
rfc3023_encoding = ''
"document encoding unknown, I tried "
+ "%s, %s, utf-8, windows-1252, and iso-8859-2 but nothing worked"
% (rfc3023_encoding, xml_encoding)
)
rfc3023_encoding = ""
elif proposed_encoding != rfc3023_encoding:
error = CharacterEncodingOverride(
'document declared as %s, but parsed as %s' %
(rfc3023_encoding, proposed_encoding))
"document declared as %s, but parsed as %s"
% (rfc3023_encoding, proposed_encoding)
)
rfc3023_encoding = proposed_encoding
result['content-type'] = http_content_type # for selecting the parser
result['encoding'] = rfc3023_encoding
result["content-type"] = http_content_type # for selecting the parser
result["encoding"] = rfc3023_encoding
if error:
result['bozo'] = True
result['bozo_exception'] = error
result["bozo"] = True
result["bozo_exception"] = error
return data
# How much to read from a binary file in order to detect encoding.
# In inital tests, 4k was enough for ~160 mostly-English feeds;
# 64k seems like a safe margin.
CONVERT_FILE_PREFIX_LEN = 2**16
# How much to read from a text file, and use as an utf-8 bytes prefix.
# Note that no encoding detection is needed in this case.
CONVERT_FILE_STR_PREFIX_LEN = 2**13
CONVERT_FILE_TEST_CHUNK_LEN = 2**16
def convert_file_to_utf8(
http_headers, file, result, optimistic_encoding_detection=True
):
"""Like convert_to_utf8(), but for a stream.
Unlike convert_to_utf8(), do not read the entire file in memory;
instead, return a text stream that decodes it on the fly.
This should consume significantly less memory,
because it avoids (repeatedly) converting the entire file contents
from bytes to str and back.
To detect the encoding, only a prefix of the file contents is used.
In rare cases, the wrong encoding may be detected for this prefix;
use optimistic_encoding_detection=False to use the entire file contents
(equivalent to a plain convert_to_utf8() call).
Args:
http_headers (dict): The response headers.
file (IO[bytes] or IO[str]): A read()-able (binary) stream.
result (dict): The result dictionary.
optimistic_encoding_detection (bool):
If true, use only a prefix of the file content to detect encoding.
Returns:
StreamFactory: a stream factory, with the detected encoding set, if any
"""
# Currently, this wraps convert_to_utf8(), because the logic is simply
# too complicated to ensure it's re-implemented correctly for a stream.
# That said, it should be possible to change the implementation
# transparently (not sure it's worth it, though).
# If file is a text stream, we don't need to detect encoding;
# we still need a bytes prefix to run functions on for side effects:
# convert_to_utf8() to sniff / set result['content-type'], and
# replace_doctype() to extract safe_entities.
if isinstance(file.read(0), str):
prefix = file.read(CONVERT_FILE_STR_PREFIX_LEN).encode("utf-8")
prefix = convert_to_utf8(http_headers, prefix, result)
result["encoding"] = "utf-8"
return StreamFactory(prefix, file, "utf-8")
if optimistic_encoding_detection:
prefix = convert_file_prefix_to_utf8(http_headers, file, result)
factory = StreamFactory(prefix, file, result.get("encoding"))
# Before returning factory, ensure the entire file can be decoded;
# if it cannot, fall back to convert_to_utf8().
#
# Not doing this means feedparser.parse() may raise UnicodeDecodeError
# instead of setting bozo_exception to CharacterEncodingOverride,
# breaking the 6.x API.
try:
text_file = factory.get_text_file()
except MissingEncoding:
return factory
try:
# read in chunks to limit memory usage
while text_file.read(CONVERT_FILE_TEST_CHUNK_LEN):
pass
except UnicodeDecodeError:
# fall back to convert_to_utf8()
file = factory.get_binary_file()
else:
return factory
# this shouldn't increase memory usage if file is BytesIO,
# since BytesIO does copy-on-write; https://bugs.python.org/issue22003
data = convert_to_utf8(http_headers, file.read(), result)
# note that data *is* the prefix
return StreamFactory(data, io.BytesIO(b""), result.get("encoding"))
def convert_file_prefix_to_utf8(
http_headers,
file: typing.IO[bytes],
result,
*,
prefix_len: int = CONVERT_FILE_PREFIX_LEN,
read_to_ascii_len: int = 2**8,
) -> bytes:
"""Like convert_to_utf8(), but only use the prefix of a binary file.
Set result like convert_to_utf8() would.
Return the updated prefix, as bytes.
"""
# This is complicated by convert_to_utf8() detecting the wrong encoding
# if we have only part of the bytes that make a code-point:
#
# '😀'.encode('utf-8') -> utf-8
# '😀'.encode('utf-8')[:-1] -> windows-1252 + bozo
prefix = file.read(prefix_len - 1)
# reading up to after an ASCII byte increases
# the likelihood of being on a code point boundary
prefix += read_to_after_ascii_byte(file, read_to_ascii_len)
# call convert_to_utf8() up to 4 times,
# to make sure we eventually land on a code point boundary
candidates = []
for attempt in range(4):
byte = file.read(1)
# we're at the end of the file, and the loop already ran once
if not byte and attempt != 0:
break
prefix += byte
fake_result: typing.Any = {}
converted_prefix = convert_to_utf8(http_headers, prefix, fake_result)
# an encoding was detected successfully, keep it
if not fake_result.get("bozo"):
break
candidates.append((file.tell(), converted_prefix, fake_result))
# no encoding was detected successfully, pick the "best" one
else:
def key(candidate):
*_, result = candidate
exc = result.get("bozo_exception")
exc_score = 0
if isinstance(exc, NonXMLContentType):
exc_score = 20
elif isinstance(exc, CharacterEncodingOverride):
exc_score = 10
return (
exc_score,
# prefer utf- encodings to anything else
result.get("encoding").startswith("utf-"),
)
candidates.sort(key=key)
offset, converted_prefix, fake_result = candidates[-1]
file.seek(offset)
result.update(fake_result)
return converted_prefix
def read_to_after_ascii_byte(file: typing.IO[bytes], max_len: int) -> bytes:
offset = file.tell()
buffer = b""
for _ in range(max_len):
byte = file.read(1)
# end of file, nothing to do
if not byte:
break
buffer += byte
# we stop after a ASCII character
if byte < b"\x80":
break
# couldn't find an ASCII character, reset the file to the original offset
else:
file.seek(offset)
return b""
return buffer
class MissingEncoding(io.UnsupportedOperation):
pass
class StreamFactory:
"""Decode on the fly a binary stream that *may* have a known encoding.
If the underlying stream is seekable, it is possible to call
the get_{text,binary}_file() methods more than once.
"""
def __init__(self, prefix: bytes, file, encoding=None):
self.prefix = prefix
self.file = ResetFileWrapper(file)
self.encoding = encoding
self.should_reset = False
def get_text_file(self, fallback_encoding=None, errors="strict"):
encoding = self.encoding or fallback_encoding
if encoding is None:
raise MissingEncoding("cannot create text stream without encoding")
if isinstance(self.file.read(0), str):
file = PrefixFileWrapper(self.prefix.decode(encoding), self.file)
else:
file = PrefixFileWrapper(
self.prefix.decode("utf-8", errors),
codecs.getreader(encoding)(self.file, errors),
)
self.reset()
return file
def get_binary_file(self):
if isinstance(self.file.read(0), str):
raise io.UnsupportedOperation(
"underlying stream is text, not binary"
) from None
file = PrefixFileWrapper(self.prefix, self.file)
self.reset()
return file
def get_file(self):
try:
return self.get_text_file()
except MissingEncoding:
return self.get_binary_file()
def reset(self):
if self.should_reset:
self.file.reset()
self.should_reset = True
class ResetFileWrapper:
"""Given a seekable file, allow reading its content again
(from the current position) by calling reset().
"""
def __init__(self, file):
self.file = file
try:
self.file_initial_offset = file.tell()
except OSError:
self.file_initial_offset = None
def read(self, size=-1):
return self.file.read(size)
def reset(self):
# raises io.UnsupportedOperation if the underlying stream is not seekable
self.file.seek(self.file_initial_offset)
class PrefixFileWrapper:
"""Stitch a (possibly modified) prefix and a file into a new file object.
>>> file = io.StringIO('abcdef')
>>> file.read(2)
'ab'
>>> wrapped = PrefixFileWrapper(file.read(2).upper(), file)
>>> wrapped.read()
'CDef'
"""
def __init__(self, prefix, file):
self.prefix = prefix
self.file = file
self.offset = 0
def read(self, size=-1):
buffer = self.file.read(0)
if self.offset < len(self.prefix):
if size < 0:
chunk = self.prefix
else:
chunk = self.prefix[self.offset : self.offset + size]
size -= len(chunk)
buffer += chunk
self.offset += len(chunk)
while True:
chunk = self.file.read(size)
if not chunk:
break
buffer += chunk
self.offset += len(chunk)
if size <= 0:
break
size -= len(chunk)
return buffer
def close(self):
# do not touch the underlying stream
pass

12
lib/feedparser/exceptions.py
View file

@ -1,5 +1,5 @@
# Exceptions used throughout feedparser
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -27,11 +27,11 @@
# POSSIBILITY OF SUCH DAMAGE.
__all__ = [
'FeedparserError',
'CharacterEncodingOverride',
'CharacterEncodingUnknown',
'NonXMLContentType',
'UndeclaredNamespace',
"FeedparserError",
"CharacterEncodingOverride",
"CharacterEncodingUnknown",
"NonXMLContentType",
"UndeclaredNamespace",
]

177
lib/feedparser/html.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -28,36 +28,49 @@
import html.entities
import re
import sgmllib3k as sgmllib
# These items must all be imported into this module due to .__code__ replacements.
from .sgml import ( # noqa: F401
attrfind,
charref,
endbracket,
entityref,
incomplete,
interesting,
sgmllib,
shorttag,
shorttagopen,
starttagopen,
tagfind,
)
_cp1252 = {
128: '\u20ac', # euro sign
130: '\u201a', # single low-9 quotation mark
131: '\u0192', # latin small letter f with hook
132: '\u201e', # double low-9 quotation mark
133: '\u2026', # horizontal ellipsis
134: '\u2020', # dagger
135: '\u2021', # double dagger
136: '\u02c6', # modifier letter circumflex accent
137: '\u2030', # per mille sign
138: '\u0160', # latin capital letter s with caron
139: '\u2039', # single left-pointing angle quotation mark
140: '\u0152', # latin capital ligature oe
142: '\u017d', # latin capital letter z with caron
145: '\u2018', # left single quotation mark
146: '\u2019', # right single quotation mark
147: '\u201c', # left double quotation mark
148: '\u201d', # right double quotation mark
149: '\u2022', # bullet
150: '\u2013', # en dash
151: '\u2014', # em dash
152: '\u02dc', # small tilde
153: '\u2122', # trade mark sign
154: '\u0161', # latin small letter s with caron
155: '\u203a', # single right-pointing angle quotation mark
156: '\u0153', # latin small ligature oe
158: '\u017e', # latin small letter z with caron
159: '\u0178', # latin capital letter y with diaeresis
128: "\u20ac", # euro sign
130: "\u201a", # single low-9 quotation mark
131: "\u0192", # latin small letter f with hook
132: "\u201e", # double low-9 quotation mark
133: "\u2026", # horizontal ellipsis
134: "\u2020", # dagger
135: "\u2021", # double dagger
136: "\u02c6", # modifier letter circumflex accent
137: "\u2030", # per mille sign
138: "\u0160", # latin capital letter s with caron
139: "\u2039", # single left-pointing angle quotation mark
140: "\u0152", # latin capital ligature oe
142: "\u017d", # latin capital letter z with caron
145: "\u2018", # left single quotation mark
146: "\u2019", # right single quotation mark
147: "\u201c", # left double quotation mark
148: "\u201d", # right double quotation mark
149: "\u2022", # bullet
150: "\u2013", # en dash
151: "\u2014", # em dash
152: "\u02dc", # small tilde
153: "\u2122", # trade mark sign
154: "\u0161", # latin small letter s with caron
155: "\u203a", # single right-pointing angle quotation mark
156: "\u0153", # latin small ligature oe
158: "\u017e", # latin small letter z with caron
159: "\u0178", # latin capital letter y with diaeresis
}
@ -65,28 +78,28 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
special = re.compile("""[<>'"]""")
bare_ampersand = re.compile(r"&(?!#\d+;|#x[0-9a-fA-F]+;|\w+;)")
elements_no_end_tag = {
'area',
'base',
'basefont',
'br',
'col',
'command',
'embed',
'frame',
'hr',
'img',
'input',
'isindex',
'keygen',
'link',
'meta',
'param',
'source',
'track',
'wbr',
"area",
"base",
"basefont",
"br",
"col",
"command",
"embed",
"frame",
"hr",
"img",
"input",
"isindex",
"keygen",
"link",
"meta",
"param",
"source",
"track",
"wbr",
}
def __init__(self, encoding=None, _type='application/xhtml+xml'):
def __init__(self, encoding=None, _type="application/xhtml+xml"):
if encoding:
self.encoding = encoding
self._type = _type
@ -105,9 +118,9 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
tag = match.group(1)
if tag in self.elements_no_end_tag:
return '<' + tag + ' />'
return "<" + tag + " />"
else:
return '<' + tag + '></' + tag + '>'
return "<" + tag + "></" + tag + ">"
# By declaring these methods and overriding their compiled code
# with the code from sgmllib, the original code will execute in
@ -128,8 +141,8 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
def parse_starttag(self, i):
j = self.__parse_starttag(i)
if self._type == 'application/xhtml+xml':
if j > 2 and self.rawdata[j-2:j] == '/>':
if self._type == "application/xhtml+xml":
if j > 2 and self.rawdata[j - 2 : j] == "/>":
self.unknown_endtag(self.lasttag)
return j
@ -139,10 +152,10 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
:rtype: None
"""
data = re.sub(r'<!((?!DOCTYPE|--|\[))', r'&lt;!\1', data, re.IGNORECASE)
data = re.sub(r'<([^<>\s]+?)\s*/>', self._shorttag_replace, data)
data = data.replace('&#39;', "'")
data = data.replace('&#34;', '"')
data = re.sub(r"<!((?!DOCTYPE|--|\[))", r"&lt;!\1", data, re.IGNORECASE)
data = re.sub(r"<([^<>\s]+?)\s*/>", self._shorttag_replace, data)
data = data.replace("&#39;", "'")
data = data.replace("&#34;", '"')
super().feed(data)
super().close()
@ -160,8 +173,7 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# *attrs* into a dictionary, then convert it back to a list.
attrs_d = {k.lower(): v for k, v in attrs}
attrs = [
(k, k in ('rel', 'type') and v.lower() or v)
for k, v in attrs_d.items()
(k, k in ("rel", "type") and v.lower() or v) for k, v in attrs_d.items()
]
attrs.sort()
return attrs
@ -177,22 +189,19 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# attrs is a list of (attr, value) tuples
# e.g. for <pre class='screen'>, tag='pre', attrs=[('class', 'screen')]
uattrs = []
strattrs = ''
strattrs = ""
if attrs:
for key, value in attrs:
value = value.replace('>', '&gt;')
value = value.replace('<', '&lt;')
value = value.replace('"', '&quot;')
value = value.replace(">", "&gt;")
value = value.replace("<", "&lt;")
value = value.replace('"', "&quot;")
value = self.bare_ampersand.sub("&amp;", value)
uattrs.append((key, value))
strattrs = ''.join(
' %s="%s"' % (key, value)
for key, value in uattrs
)
strattrs = "".join(f' {key}="{value}"' for key, value in uattrs)
if tag in self.elements_no_end_tag:
self.pieces.append('<%s%s />' % (tag, strattrs))
self.pieces.append(f"<{tag}{strattrs} />")
else:
self.pieces.append('<%s%s>' % (tag, strattrs))
self.pieces.append(f"<{tag}{strattrs}>")
def unknown_endtag(self, tag):
"""
@ -214,15 +223,15 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# Called for each character reference, e.g. '&#160;' will extract '160'
# Reconstruct the original character reference.
ref = ref.lower()
if ref.startswith('x'):
if ref.startswith("x"):
value = int(ref[1:], 16)
else:
value = int(ref)
if value in _cp1252:
self.pieces.append('&#%s;' % hex(ord(_cp1252[value]))[1:])
self.pieces.append("&#%s;" % hex(ord(_cp1252[value]))[1:])
else:
self.pieces.append('&#%s;' % ref)
self.pieces.append("&#%s;" % ref)
def handle_entityref(self, ref):
"""
@ -232,10 +241,10 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# Called for each entity reference, e.g. '&copy;' will extract 'copy'
# Reconstruct the original entity reference.
if ref in html.entities.name2codepoint or ref == 'apos':
self.pieces.append('&%s;' % ref)
if ref in html.entities.name2codepoint or ref == "apos":
self.pieces.append("&%s;" % ref)
else:
self.pieces.append('&amp;%s' % ref)
self.pieces.append("&amp;%s" % ref)
def handle_data(self, text):
"""
@ -256,7 +265,7 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# Called for HTML comments, e.g. <!-- insert Javascript code here -->
# Reconstruct the original comment.
self.pieces.append('<!--%s-->' % text)
self.pieces.append("<!--%s-->" % text)
def handle_pi(self, text):
"""
@ -266,7 +275,7 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# Called for each processing instruction, e.g. <?instruction>
# Reconstruct original processing instruction.
self.pieces.append('<?%s>' % text)
self.pieces.append("<?%s>" % text)
def handle_decl(self, text):
"""
@ -278,9 +287,9 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
# <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
# "http://www.w3.org/TR/html4/loose.dtd">
# Reconstruct original DOCTYPE
self.pieces.append('<!%s>' % text)
self.pieces.append("<!%s>" % text)
_new_declname_match = re.compile(r'[a-zA-Z][-_.a-zA-Z0-9:]*\s*').match
_new_declname_match = re.compile(r"[a-zA-Z][-_.a-zA-Z0-9:]*\s*").match
def _scan_name(self, i, declstartpos):
"""
@ -311,7 +320,7 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
:rtype: str
"""
return '&#%s;' % name
return "&#%s;" % name
def convert_entityref(self, name):
"""
@ -319,7 +328,7 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
:rtype: str
"""
return '&%s;' % name
return "&%s;" % name
def output(self):
"""Return processed HTML as a single string.
@ -327,7 +336,7 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
:rtype: str
"""
return ''.join(self.pieces)
return "".join(self.pieces)
def parse_declaration(self, i):
"""
@ -339,5 +348,5 @@ class BaseHTMLProcessor(sgmllib.SGMLParser):
return sgmllib.SGMLParser.parse_declaration(self, i)
except (AssertionError, sgmllib.SGMLParseError):
# Escape the doctype declaration and continue parsing.
self.handle_data('&lt;')
return i+1
self.handle_data("&lt;")
return i + 1

233
lib/feedparser/http.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -25,203 +25,54 @@
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
import base64
import datetime
import gzip
import io
import re
import struct
import urllib.parse
import urllib.request
import zlib
from __future__ import annotations
import typing
import requests
from .datetimes import _parse_date
from .urls import convert_to_idn
# HTTP "Accept" header to send to servers when downloading feeds.
ACCEPT_HEADER: str = (
"application/atom+xml"
",application/rdf+xml"
",application/rss+xml"
",application/x-netcdf"
",application/xml"
";q=0.9,text/xml"
";q=0.2,*/*"
";q=0.1"
)
# HTTP "Accept" header to send to servers when downloading feeds. If you don't
# want to send an Accept header, set this to None.
ACCEPT_HEADER = "application/atom+xml,application/rdf+xml,application/rss+xml,application/x-netcdf,application/xml;q=0.9,text/xml;q=0.2,*/*;q=0.1"
def get(url: str, result: dict[str, typing.Any]) -> bytes:
from . import USER_AGENT
agent = USER_AGENT
class URLHandler(urllib.request.HTTPDigestAuthHandler, urllib.request.HTTPRedirectHandler, urllib.request.HTTPDefaultErrorHandler):
def http_error_default(self, req, fp, code, msg, headers):
# The default implementation just raises HTTPError.
# Forget that.
fp.status = code
return fp
try:
response = requests.get(
url,
headers={"User-Agent": agent, "Accept": ACCEPT_HEADER},
timeout=10,
)
except requests.RequestException as exception:
result["bozo"] = True
result["bozo_exception"] = exception
return b""
def http_error_301(self, req, fp, code, msg, hdrs):
result = urllib.request.HTTPRedirectHandler.http_error_301(self, req, fp, code, msg, hdrs)
if not result:
return fp
result.status = code
result.newurl = result.geturl()
return result
# The default implementations in urllib.request.HTTPRedirectHandler
# are identical, so hardcoding a http_error_301 call above
# won't affect anything
http_error_300 = http_error_301
http_error_302 = http_error_301
http_error_303 = http_error_301
http_error_307 = http_error_301
def http_error_401(self, req, fp, code, msg, headers):
# Check if
# - server requires digest auth, AND
# - we tried (unsuccessfully) with basic auth, AND
# If all conditions hold, parse authentication information
# out of the Authorization header we sent the first time
# (for the username and password) and the WWW-Authenticate
# header the server sent back (for the realm) and retry
# the request with the appropriate digest auth headers instead.
# This evil genius hack has been brought to you by Aaron Swartz.
host = urllib.parse.urlparse(req.get_full_url())[1]
if 'Authorization' not in req.headers or 'WWW-Authenticate' not in headers:
return self.http_error_default(req, fp, code, msg, headers)
auth = base64.decodebytes(req.headers['Authorization'].split(' ')[1].encode()).decode()
user, passw = auth.split(':')
realm = re.findall('realm="([^"]*)"', headers['WWW-Authenticate'])[0]
self.add_password(realm, host, user, passw)
retry = self.http_error_auth_reqed('www-authenticate', host, req, headers)
self.reset_retry_count()
return retry
def _build_urllib2_request(url, agent, accept_header, etag, modified, referrer, auth, request_headers):
request = urllib.request.Request(url)
request.add_header('User-Agent', agent)
if etag:
request.add_header('If-None-Match', etag)
if isinstance(modified, str):
modified = _parse_date(modified)
elif isinstance(modified, datetime.datetime):
modified = modified.utctimetuple()
if modified:
# format into an RFC 1123-compliant timestamp. We can't use
# time.strftime() since the %a and %b directives can be affected
# by the current locale, but RFC 2616 states that dates must be
# in English.
short_weekdays = ['Mon', 'Tue', 'Wed', 'Thu', 'Fri', 'Sat', 'Sun']
months = ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun', 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']
request.add_header('If-Modified-Since', '%s, %02d %s %04d %02d:%02d:%02d GMT' % (short_weekdays[modified[6]], modified[2], months[modified[1] - 1], modified[0], modified[3], modified[4], modified[5]))
if referrer:
request.add_header('Referer', referrer)
request.add_header('Accept-encoding', 'gzip, deflate')
if auth:
request.add_header('Authorization', 'Basic %s' % auth)
if accept_header:
request.add_header('Accept', accept_header)
# use this for whatever -- cookies, special headers, etc
# [('Cookie','Something'),('x-special-header','Another Value')]
for header_name, header_value in request_headers.items():
request.add_header(header_name, header_value)
request.add_header('A-IM', 'feed') # RFC 3229 support
return request
def get(url, etag=None, modified=None, agent=None, referrer=None, handlers=None, request_headers=None, result=None):
if handlers is None:
handlers = []
elif not isinstance(handlers, list):
handlers = [handlers]
if request_headers is None:
request_headers = {}
# Deal with the feed URI scheme
if url.startswith('feed:http'):
url = url[5:]
elif url.startswith('feed:'):
url = 'http:' + url[5:]
if not agent:
from . import USER_AGENT
agent = USER_AGENT
# Test for inline user:password credentials for HTTP basic auth
auth = None
if not url.startswith('ftp:'):
url_pieces = urllib.parse.urlparse(url)
if url_pieces.username:
new_pieces = list(url_pieces)
new_pieces[1] = url_pieces.hostname
if url_pieces.port:
new_pieces[1] = f'{url_pieces.hostname}:{url_pieces.port}'
url = urllib.parse.urlunparse(new_pieces)
auth = base64.standard_b64encode(f'{url_pieces.username}:{url_pieces.password}'.encode()).decode()
# iri support
if not isinstance(url, bytes):
url = convert_to_idn(url)
# Prevent UnicodeEncodeErrors caused by Unicode characters in the path.
bits = []
for c in url:
try:
c.encode('ascii')
except UnicodeEncodeError:
bits.append(urllib.parse.quote(c))
else:
bits.append(c)
url = ''.join(bits)
# try to open with urllib2 (to use optional headers)
request = _build_urllib2_request(url, agent, ACCEPT_HEADER, etag, modified, referrer, auth, request_headers)
opener = urllib.request.build_opener(*tuple(handlers + [URLHandler()]))
opener.addheaders = [] # RMK - must clear so we only send our custom User-Agent
f = opener.open(request)
data = f.read()
f.close()
# lowercase all of the HTTP headers for comparisons per RFC 2616
result['headers'] = {k.lower(): v for k, v in f.headers.items()}
# if feed is gzip-compressed, decompress it
if data and 'gzip' in result['headers'].get('content-encoding', ''):
try:
data = gzip.GzipFile(fileobj=io.BytesIO(data)).read()
except (EOFError, IOError, struct.error) as e:
# IOError can occur if the gzip header is bad.
# struct.error can occur if the data is damaged.
result['bozo'] = True
result['bozo_exception'] = e
if isinstance(e, struct.error):
# A gzip header was found but the data is corrupt.
# Ideally, we should re-request the feed without the
# 'Accept-encoding: gzip' header, but we don't.
data = None
elif data and 'deflate' in result['headers'].get('content-encoding', ''):
try:
data = zlib.decompress(data)
except zlib.error:
try:
# The data may have no headers and no checksum.
data = zlib.decompress(data, -15)
except zlib.error as e:
result['bozo'] = True
result['bozo_exception'] = e
# Lowercase the HTTP header keys for comparisons per RFC 2616.
result["headers"] = {k.lower(): v for k, v in response.headers.items()}
# save HTTP headers
if 'etag' in result['headers']:
etag = result['headers'].get('etag', '')
if isinstance(etag, bytes):
etag = etag.decode('utf-8', 'ignore')
if etag:
result['etag'] = etag
if 'last-modified' in result['headers']:
modified = result['headers'].get('last-modified', '')
if "etag" in result["headers"]:
result["etag"] = result["headers"]["etag"]
if "last-modified" in result["headers"]:
modified = result["headers"]["last-modified"]
if modified:
result['modified'] = modified
result['modified_parsed'] = _parse_date(modified)
if isinstance(f.url, bytes):
result['href'] = f.url.decode('utf-8', 'ignore')
else:
result['href'] = f.url
result['status'] = getattr(f, 'status', None) or 200
# Stop processing if the server sent HTTP 304 Not Modified.
if getattr(f, 'code', 0) == 304:
result['version'] = ''
result['debug_message'] = 'The feed has not changed since you last checked, ' + \
'so the server sent no data. This is a feature, not a bug!'
return data
result["modified"] = modified
result["modified_parsed"] = _parse_date(modified)
result["href"] = response.url
result["status"] = response.status_code
return response.content

598
lib/feedparser/mixin.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -30,137 +30,144 @@ import binascii
import copy
import html.entities
import re
from typing import Dict
import xml.sax.saxutils
from typing import Dict
from .html import _cp1252
from .namespaces import _base, cc, dc, georss, itunes, mediarss, psc
from .sanitizer import sanitize_html, HTMLSanitizer
from .util import FeedParserDict
from .sanitizer import HTMLSanitizer, sanitize_html
from .urls import _urljoin, make_safe_absolute_uri, resolve_relative_uris
from .util import FeedParserDict
email_pattern = re.compile(
r"(([a-zA-Z0-9_.+-]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)"
r"|(([a-zA-Z0-9-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(]?))"
r"(\?subject=\S+)?"
)
class XMLParserMixin(
_base.Namespace,
cc.Namespace,
dc.Namespace,
georss.Namespace,
itunes.Namespace,
mediarss.Namespace,
psc.Namespace,
_base.Namespace,
cc.Namespace,
dc.Namespace,
georss.Namespace,
itunes.Namespace,
mediarss.Namespace,
psc.Namespace,
):
namespaces = {
'': '',
'http://backend.userland.com/rss': '',
'http://blogs.law.harvard.edu/tech/rss': '',
'http://purl.org/rss/1.0/': '',
'http://my.netscape.com/rdf/simple/0.9/': '',
'http://example.com/newformat#': '',
'http://example.com/necho': '',
'http://purl.org/echo/': '',
'uri/of/echo/namespace#': '',
'http://purl.org/pie/': '',
'http://purl.org/atom/ns#': '',
'http://www.w3.org/2005/Atom': '',
'http://purl.org/rss/1.0/modules/rss091#': '',
'http://webns.net/mvcb/': 'admin',
'http://purl.org/rss/1.0/modules/aggregation/': 'ag',
'http://purl.org/rss/1.0/modules/annotate/': 'annotate',
'http://media.tangent.org/rss/1.0/': 'audio',
'http://backend.userland.com/blogChannelModule': 'blogChannel',
'http://creativecommons.org/ns#license': 'cc',
'http://web.resource.org/cc/': 'cc',
'http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html': 'creativeCommons',
'http://backend.userland.com/creativeCommonsRssModule': 'creativeCommons',
'http://purl.org/rss/1.0/modules/company': 'co',
'http://purl.org/rss/1.0/modules/content/': 'content',
'http://my.theinfo.org/changed/1.0/rss/': 'cp',
'http://purl.org/dc/elements/1.1/': 'dc',
'http://purl.org/dc/terms/': 'dcterms',
'http://purl.org/rss/1.0/modules/email/': 'email',
'http://purl.org/rss/1.0/modules/event/': 'ev',
'http://rssnamespace.org/feedburner/ext/1.0': 'feedburner',
'http://freshmeat.net/rss/fm/': 'fm',
'http://xmlns.com/foaf/0.1/': 'foaf',
'http://www.w3.org/2003/01/geo/wgs84_pos#': 'geo',
'http://www.georss.org/georss': 'georss',
'http://www.opengis.net/gml': 'gml',
'http://postneo.com/icbm/': 'icbm',
'http://purl.org/rss/1.0/modules/image/': 'image',
'http://www.itunes.com/DTDs/PodCast-1.0.dtd': 'itunes',
'http://example.com/DTDs/PodCast-1.0.dtd': 'itunes',
'http://purl.org/rss/1.0/modules/link/': 'l',
'http://search.yahoo.com/mrss': 'media',
"": "",
"http://backend.userland.com/rss": "",
"http://blogs.law.harvard.edu/tech/rss": "",
"http://purl.org/rss/1.0/": "",
"http://my.netscape.com/rdf/simple/0.9/": "",
"http://example.com/newformat#": "",
"http://example.com/necho": "",
"http://purl.org/echo/": "",
"uri/of/echo/namespace#": "",
"http://purl.org/pie/": "",
"http://purl.org/atom/ns#": "",
"http://www.w3.org/2005/Atom": "",
"http://purl.org/rss/1.0/modules/rss091#": "",
"http://webns.net/mvcb/": "admin",
"http://purl.org/rss/1.0/modules/aggregation/": "ag",
"http://purl.org/rss/1.0/modules/annotate/": "annotate",
"http://media.tangent.org/rss/1.0/": "audio",
"http://backend.userland.com/blogChannelModule": "blogChannel",
"http://creativecommons.org/ns#license": "cc",
"http://web.resource.org/cc/": "cc",
"http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html": (
"creativeCommons"
),
"http://backend.userland.com/creativeCommonsRssModule": "creativeCommons",
"http://purl.org/rss/1.0/modules/company": "co",
"http://purl.org/rss/1.0/modules/content/": "content",
"http://my.theinfo.org/changed/1.0/rss/": "cp",
"http://purl.org/dc/elements/1.1/": "dc",
"http://purl.org/dc/terms/": "dcterms",
"http://purl.org/rss/1.0/modules/email/": "email",
"http://purl.org/rss/1.0/modules/event/": "ev",
"http://rssnamespace.org/feedburner/ext/1.0": "feedburner",
"http://freshmeat.net/rss/fm/": "fm",
"http://xmlns.com/foaf/0.1/": "foaf",
"http://www.w3.org/2003/01/geo/wgs84_pos#": "geo",
"http://www.georss.org/georss": "georss",
"http://www.opengis.net/gml": "gml",
"http://postneo.com/icbm/": "icbm",
"http://purl.org/rss/1.0/modules/image/": "image",
"http://www.itunes.com/DTDs/PodCast-1.0.dtd": "itunes",
"http://example.com/DTDs/PodCast-1.0.dtd": "itunes",
"http://purl.org/rss/1.0/modules/link/": "l",
"http://search.yahoo.com/mrss": "media",
# Version 1.1.2 of the Media RSS spec added the trailing slash on the namespace
'http://search.yahoo.com/mrss/': 'media',
'http://madskills.com/public/xml/rss/module/pingback/': 'pingback',
'http://prismstandard.org/namespaces/1.2/basic/': 'prism',
'http://www.w3.org/1999/02/22-rdf-syntax-ns#': 'rdf',
'http://www.w3.org/2000/01/rdf-schema#': 'rdfs',
'http://purl.org/rss/1.0/modules/reference/': 'ref',
'http://purl.org/rss/1.0/modules/richequiv/': 'reqv',
'http://purl.org/rss/1.0/modules/search/': 'search',
'http://purl.org/rss/1.0/modules/slash/': 'slash',
'http://schemas.xmlsoap.org/soap/envelope/': 'soap',
'http://purl.org/rss/1.0/modules/servicestatus/': 'ss',
'http://hacks.benhammersley.com/rss/streaming/': 'str',
'http://purl.org/rss/1.0/modules/subscription/': 'sub',
'http://purl.org/rss/1.0/modules/syndication/': 'sy',
'http://schemas.pocketsoap.com/rss/myDescModule/': 'szf',
'http://purl.org/rss/1.0/modules/taxonomy/': 'taxo',
'http://purl.org/rss/1.0/modules/threading/': 'thr',
'http://purl.org/rss/1.0/modules/textinput/': 'ti',
'http://madskills.com/public/xml/rss/module/trackback/': 'trackback',
'http://wellformedweb.org/commentAPI/': 'wfw',
'http://purl.org/rss/1.0/modules/wiki/': 'wiki',
'http://www.w3.org/1999/xhtml': 'xhtml',
'http://www.w3.org/1999/xlink': 'xlink',
'http://www.w3.org/XML/1998/namespace': 'xml',
'http://podlove.org/simple-chapters': 'psc',
"http://search.yahoo.com/mrss/": "media",
"http://madskills.com/public/xml/rss/module/pingback/": "pingback",
"http://prismstandard.org/namespaces/1.2/basic/": "prism",
"http://www.w3.org/1999/02/22-rdf-syntax-ns#": "rdf",
"http://www.w3.org/2000/01/rdf-schema#": "rdfs",
"http://purl.org/rss/1.0/modules/reference/": "ref",
"http://purl.org/rss/1.0/modules/richequiv/": "reqv",
"http://purl.org/rss/1.0/modules/search/": "search",
"http://purl.org/rss/1.0/modules/slash/": "slash",
"http://schemas.xmlsoap.org/soap/envelope/": "soap",
"http://purl.org/rss/1.0/modules/servicestatus/": "ss",
"http://hacks.benhammersley.com/rss/streaming/": "str",
"http://purl.org/rss/1.0/modules/subscription/": "sub",
"http://purl.org/rss/1.0/modules/syndication/": "sy",
"http://schemas.pocketsoap.com/rss/myDescModule/": "szf",
"http://purl.org/rss/1.0/modules/taxonomy/": "taxo",
"http://purl.org/rss/1.0/modules/threading/": "thr",
"http://purl.org/rss/1.0/modules/textinput/": "ti",
"http://madskills.com/public/xml/rss/module/trackback/": "trackback",
"http://wellformedweb.org/commentAPI/": "wfw",
"http://purl.org/rss/1.0/modules/wiki/": "wiki",
"http://www.w3.org/1999/xhtml": "xhtml",
"http://www.w3.org/1999/xlink": "xlink",
"http://www.w3.org/XML/1998/namespace": "xml",
"http://podlove.org/simple-chapters": "psc",
}
_matchnamespaces: Dict[str, str] = {}
can_be_relative_uri = {
'comments',
'docs',
'href',
'icon',
'id',
'link',
'logo',
'url',
'wfw_comment',
'wfw_commentrss',
"comments",
"docs",
"href",
"icon",
"id",
"link",
"logo",
"url",
"wfw_comment",
"wfw_commentrss",
}
can_contain_relative_uris = {
'content',
'copyright',
'description',
'info',
'rights',
'subtitle',
'summary',
'tagline',
'title',
"content",
"copyright",
"description",
"info",
"rights",
"subtitle",
"summary",
"tagline",
"title",
}
can_contain_dangerous_markup = {
'content',
'copyright',
'description',
'info',
'rights',
'subtitle',
'summary',
'tagline',
'title',
"content",
"copyright",
"description",
"info",
"rights",
"subtitle",
"summary",
"tagline",
"title",
}
html_types = {
'application/xhtml+xml',
'text/html',
"application/xhtml+xml",
"text/html",
}
def __init__(self):
@ -169,7 +176,7 @@ class XMLParserMixin(
self._matchnamespaces[k.lower()] = v
self.feeddata = FeedParserDict() # feed-level data
self.entries = [] # list of entry-level data
self.version = '' # feed type/version, see SUPPORTED_VERSIONS
self.version = "" # feed type/version, see SUPPORTED_VERSIONS
self.namespaces_in_use = {} # dictionary of namespaces defined by the feed
self.resolve_relative_uris = False
self.sanitize_html = False
@ -198,7 +205,7 @@ class XMLParserMixin(
self.depth = 0
self.hasContent = 0
if self.lang:
self.feeddata['language'] = self.lang.replace('_', '-')
self.feeddata["language"] = self.lang.replace("_", "-")
# A map of the following form:
# {
@ -208,7 +215,7 @@ class XMLParserMixin(
# },
# }
self.property_depth_map = {}
super(XMLParserMixin, self).__init__()
super().__init__()
def _normalize_attributes(self, kv):
raise NotImplementedError
@ -222,72 +229,80 @@ class XMLParserMixin(
# track xml:base and xml:lang
attrs_d = dict(attrs)
baseuri = attrs_d.get('xml:base', attrs_d.get('base')) or self.baseuri
baseuri = attrs_d.get("xml:base", attrs_d.get("base")) or self.baseuri
if isinstance(baseuri, bytes):
baseuri = baseuri.decode(self.encoding, 'ignore')
baseuri = baseuri.decode(self.encoding, "ignore")
# ensure that self.baseuri is always an absolute URI that
# uses a whitelisted URI scheme (e.g. not `javscript:`)
if self.baseuri:
self.baseuri = make_safe_absolute_uri(self.baseuri, baseuri) or self.baseuri
else:
self.baseuri = _urljoin(self.baseuri, baseuri)
lang = attrs_d.get('xml:lang', attrs_d.get('lang'))
if lang == '':
lang = attrs_d.get("xml:lang", attrs_d.get("lang"))
if lang == "":
# xml:lang could be explicitly set to '', we need to capture that
lang = None
elif lang is None:
# if no xml:lang is specified, use parent lang
lang = self.lang
if lang:
if tag in ('feed', 'rss', 'rdf:RDF'):
self.feeddata['language'] = lang.replace('_', '-')
if tag in ("feed", "rss", "rdf:RDF"):
self.feeddata["language"] = lang.replace("_", "-")
self.lang = lang
self.basestack.append(self.baseuri)
self.langstack.append(lang)
# track namespaces
for prefix, uri in attrs:
if prefix.startswith('xmlns:'):
if prefix.startswith("xmlns:"):
self.track_namespace(prefix[6:], uri)
elif prefix == 'xmlns':
elif prefix == "xmlns":
self.track_namespace(None, uri)
# track inline content
if self.incontent and not self.contentparams.get('type', 'xml').endswith('xml'):
if tag in ('xhtml:div', 'div'):
if self.incontent and not self.contentparams.get("type", "xml").endswith("xml"):
if tag in ("xhtml:div", "div"):
return # typepad does this 10/2007
# element declared itself as escaped markup, but it isn't really
self.contentparams['type'] = 'application/xhtml+xml'
if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml':
if tag.find(':') != -1:
prefix, tag = tag.split(':', 1)
namespace = self.namespaces_in_use.get(prefix, '')
if tag == 'math' and namespace == 'http://www.w3.org/1998/Math/MathML':
attrs.append(('xmlns', namespace))
if tag == 'svg' and namespace == 'http://www.w3.org/2000/svg':
attrs.append(('xmlns', namespace))
if tag == 'svg':
self.contentparams["type"] = "application/xhtml+xml"
if self.incontent and self.contentparams.get("type") == "application/xhtml+xml":
if tag.find(":") != -1:
prefix, tag = tag.split(":", 1)
namespace = self.namespaces_in_use.get(prefix, "")
if tag == "math" and namespace == "http://www.w3.org/1998/Math/MathML":
attrs.append(("xmlns", namespace))
if tag == "svg" and namespace == "http://www.w3.org/2000/svg":
attrs.append(("xmlns", namespace))
if tag == "svg":
self.svgOK += 1
return self.handle_data('<%s%s>' % (tag, self.strattrs(attrs)), escape=0)
return self.handle_data(f"<{tag}{self.strattrs(attrs)}>", escape=0)
# match namespaces
if tag.find(':') != -1:
prefix, suffix = tag.split(':', 1)
if tag.find(":") != -1:
prefix, suffix = tag.split(":", 1)
else:
prefix, suffix = '', tag
prefix, suffix = "", tag
prefix = self.namespacemap.get(prefix, prefix)
if prefix:
prefix = prefix + '_'
prefix = prefix + "_"
# Special hack for better tracking of empty textinput/image elements in
# illformed feeds.
if (not prefix) and tag not in ('title', 'link', 'description', 'name'):
if (not prefix) and tag not in ("title", "link", "description", "name"):
self.intextinput = 0
if (not prefix) and tag not in ('title', 'link', 'description', 'url', 'href', 'width', 'height'):
if (not prefix) and tag not in (
"title",
"link",
"description",
"url",
"href",
"width",
"height",
):
self.inimage = 0
# call special handler (if defined) or default handler
methodname = '_start_' + prefix + suffix
methodname = "_start_" + prefix + suffix
try:
method = getattr(self, methodname)
return method(attrs_d)
@ -305,18 +320,18 @@ class XMLParserMixin(
def unknown_endtag(self, tag):
# match namespaces
if tag.find(':') != -1:
prefix, suffix = tag.split(':', 1)
if tag.find(":") != -1:
prefix, suffix = tag.split(":", 1)
else:
prefix, suffix = '', tag
prefix, suffix = "", tag
prefix = self.namespacemap.get(prefix, prefix)
if prefix:
prefix = prefix + '_'
if suffix == 'svg' and self.svgOK:
prefix = prefix + "_"
if suffix == "svg" and self.svgOK:
self.svgOK -= 1
# call special handler (if defined) or default handler
methodname = '_end_' + prefix + suffix
methodname = "_end_" + prefix + suffix
try:
if self.svgOK:
raise AttributeError()
@ -326,14 +341,14 @@ class XMLParserMixin(
self.pop(prefix + suffix)
# track inline content
if self.incontent and not self.contentparams.get('type', 'xml').endswith('xml'):
if self.incontent and not self.contentparams.get("type", "xml").endswith("xml"):
# element declared itself as escaped markup, but it isn't really
if tag in ('xhtml:div', 'div'):
if tag in ("xhtml:div", "div"):
return # typepad does this 10/2007
self.contentparams['type'] = 'application/xhtml+xml'
if self.incontent and self.contentparams.get('type') == 'application/xhtml+xml':
tag = tag.split(':')[-1]
self.handle_data('</%s>' % tag, escape=0)
self.contentparams["type"] = "application/xhtml+xml"
if self.incontent and self.contentparams.get("type") == "application/xhtml+xml":
tag = tag.split(":")[-1]
self.handle_data("</%s>" % tag, escape=0)
# track xml:base and xml:lang going out of scope
if self.basestack:
@ -352,33 +367,33 @@ class XMLParserMixin(
if not self.elementstack:
return
ref = ref.lower()
if ref in ('34', '38', '39', '60', '62', 'x22', 'x26', 'x27', 'x3c', 'x3e'):
text = '&#%s;' % ref
if ref in ("34", "38", "39", "60", "62", "x22", "x26", "x27", "x3c", "x3e"):
text = "&#%s;" % ref
else:
if ref[0] == 'x':
if ref[0] == "x":
c = int(ref[1:], 16)
else:
c = int(ref)
text = chr(c).encode('utf-8')
text = chr(c).encode("utf-8")
self.elementstack[-1][2].append(text)
def handle_entityref(self, ref):
# Called for each entity reference, e.g. for '&copy;', ref is 'copy'
if not self.elementstack:
return
if ref in ('lt', 'gt', 'quot', 'amp', 'apos'):
text = '&%s;' % ref
if ref in ("lt", "gt", "quot", "amp", "apos"):
text = "&%s;" % ref
elif ref in self.entities:
text = self.entities[ref]
if text.startswith('&#') and text.endswith(';'):
if text.startswith("&#") and text.endswith(";"):
return self.handle_entityref(text)
else:
try:
html.entities.name2codepoint[ref]
except KeyError:
text = '&%s;' % ref
text = "&%s;" % ref
else:
text = chr(html.entities.name2codepoint[ref]).encode('utf-8')
text = chr(html.entities.name2codepoint[ref]).encode("utf-8")
self.elementstack[-1][2].append(text)
def handle_data(self, text, escape=1):
@ -386,7 +401,7 @@ class XMLParserMixin(
# not containing any character or entity references
if not self.elementstack:
return
if escape and self.contentparams.get('type') == 'application/xhtml+xml':
if escape and self.contentparams.get("type") == "application/xhtml+xml":
text = xml.sax.saxutils.escape(text)
self.elementstack[-1][2].append(text)
@ -403,18 +418,18 @@ class XMLParserMixin(
def parse_declaration(self, i):
# Override internal declaration handler to handle CDATA blocks.
if self.rawdata[i:i+9] == '<![CDATA[':
k = self.rawdata.find(']]>', i)
if self.rawdata[i : i + 9] == "<![CDATA[":
k = self.rawdata.find("]]>", i)
if k == -1:
# CDATA block began but didn't finish
k = len(self.rawdata)
return k
self.handle_data(xml.sax.saxutils.escape(self.rawdata[i+9:k]), 0)
return k+3
self.handle_data(xml.sax.saxutils.escape(self.rawdata[i + 9 : k]), 0)
return k + 3
else:
k = self.rawdata.find('>', i)
k = self.rawdata.find(">", i)
if k >= 0:
return k+1
return k + 1
else:
# We have an incomplete CDATA block.
return k
@ -422,35 +437,35 @@ class XMLParserMixin(
@staticmethod
def map_content_type(content_type):
content_type = content_type.lower()
if content_type == 'text' or content_type == 'plain':
content_type = 'text/plain'
elif content_type == 'html':
content_type = 'text/html'
elif content_type == 'xhtml':
content_type = 'application/xhtml+xml'
if content_type == "text" or content_type == "plain":
content_type = "text/plain"
elif content_type == "html":
content_type = "text/html"
elif content_type == "xhtml":
content_type = "application/xhtml+xml"
return content_type
def track_namespace(self, prefix, uri):
loweruri = uri.lower()
if not self.version:
if (prefix, loweruri) == (None, 'http://my.netscape.com/rdf/simple/0.9/'):
self.version = 'rss090'
elif loweruri == 'http://purl.org/rss/1.0/':
self.version = 'rss10'
elif loweruri == 'http://www.w3.org/2005/atom':
self.version = 'atom10'
if loweruri.find('backend.userland.com/rss') != -1:
if (prefix, loweruri) == (None, "http://my.netscape.com/rdf/simple/0.9/"):
self.version = "rss090"
elif loweruri == "http://purl.org/rss/1.0/":
self.version = "rss10"
elif loweruri == "http://www.w3.org/2005/atom":
self.version = "atom10"
if loweruri.find("backend.userland.com/rss") != -1:
# match any backend.userland.com namespace
uri = 'http://backend.userland.com/rss'
uri = "http://backend.userland.com/rss"
loweruri = uri
if loweruri in self._matchnamespaces:
self.namespacemap[prefix] = self._matchnamespaces[loweruri]
self.namespaces_in_use[self._matchnamespaces[loweruri]] = uri
else:
self.namespaces_in_use[prefix or ''] = uri
self.namespaces_in_use[prefix or ""] = uri
def resolve_uri(self, uri):
return _urljoin(self.baseuri or '', uri)
return _urljoin(self.baseuri or "", uri)
@staticmethod
def decode_entities(element, data):
@ -458,8 +473,8 @@ class XMLParserMixin(
@staticmethod
def strattrs(attrs):
return ''.join(
' %s="%s"' % (t[0], xml.sax.saxutils.escape(t[1], {'"': '&quot;'}))
return "".join(
' {}="{}"'.format(t[0], xml.sax.saxutils.escape(t[1], {'"': "&quot;"}))
for t in attrs
)
@ -475,11 +490,14 @@ class XMLParserMixin(
element, expecting_text, pieces = self.elementstack.pop()
# Ensure each piece is a str for Python 3
for (i, v) in enumerate(pieces):
for i, v in enumerate(pieces):
if isinstance(v, bytes):
pieces[i] = v.decode('utf-8')
pieces[i] = v.decode("utf-8")
if self.version == 'atom10' and self.contentparams.get('type', 'text') == 'application/xhtml+xml':
if (
self.version == "atom10"
and self.contentparams.get("type", "text") == "application/xhtml+xml"
):
# remove enclosing child element, but only if it is a <div> and
# only if all the remaining content is nested underneath it.
# This means that the divs would be retained in the following:
@ -488,76 +506,95 @@ class XMLParserMixin(
del pieces[-1]
while pieces and len(pieces) > 1 and not pieces[0].strip():
del pieces[0]
if pieces and (pieces[0] == '<div>' or pieces[0].startswith('<div ')) and pieces[-1] == '</div>':
if (
pieces
and (pieces[0] == "<div>" or pieces[0].startswith("<div "))
and pieces[-1] == "</div>"
):
depth = 0
for piece in pieces[:-1]:
if piece.startswith('</'):
if piece.startswith("</"):
depth -= 1
if depth == 0:
break
elif piece.startswith('<') and not piece.endswith('/>'):
elif piece.startswith("<") and not piece.endswith("/>"):
depth += 1
else:
pieces = pieces[1:-1]
output = ''.join(pieces)
output = "".join(pieces)
if strip_whitespace:
output = output.strip()
if not expecting_text:
return output
# decode base64 content
if base64 and self.contentparams.get('base64', 0):
if base64 and self.contentparams.get("base64", 0):
try:
output = base64.decodebytes(output.encode('utf8')).decode('utf8')
output = base64.decodebytes(output.encode("utf8")).decode("utf8")
except (binascii.Error, binascii.Incomplete, UnicodeDecodeError):
pass
# resolve relative URIs
if (element in self.can_be_relative_uri) and output:
# do not resolve guid elements with isPermalink="false"
if not element == 'id' or self.guidislink:
if not element == "id" or self.guidislink:
output = self.resolve_uri(output)
# decode entities within embedded markup
if not self.contentparams.get('base64', 0):
if not self.contentparams.get("base64", 0):
output = self.decode_entities(element, output)
# some feed formats require consumers to guess
# whether the content is html or plain text
if not self.version.startswith('atom') and self.contentparams.get('type') == 'text/plain':
if (
not self.version.startswith("atom")
and self.contentparams.get("type") == "text/plain"
):
if self.looks_like_html(output):
self.contentparams['type'] = 'text/html'
self.contentparams["type"] = "text/html"
# remove temporary cruft from contentparams
try:
del self.contentparams['mode']
del self.contentparams["mode"]
except KeyError:
pass
try:
del self.contentparams['base64']
del self.contentparams["base64"]
except KeyError:
pass
is_htmlish = self.map_content_type(self.contentparams.get('type', 'text/html')) in self.html_types
is_htmlish = (
self.map_content_type(self.contentparams.get("type", "text/html"))
in self.html_types
)
# resolve relative URIs within embedded markup
if is_htmlish and self.resolve_relative_uris:
if element in self.can_contain_relative_uris:
output = resolve_relative_uris(output, self.baseuri, self.encoding, self.contentparams.get('type', 'text/html'))
output = resolve_relative_uris(
output,
self.baseuri,
self.encoding,
self.contentparams.get("type", "text/html"),
)
# sanitize embedded markup
if is_htmlish and self.sanitize_html:
if element in self.can_contain_dangerous_markup:
output = sanitize_html(output, self.encoding, self.contentparams.get('type', 'text/html'))
output = sanitize_html(
output, self.encoding, self.contentparams.get("type", "text/html")
)
if self.encoding and isinstance(output, bytes):
output = output.decode(self.encoding, 'ignore')
output = output.decode(self.encoding, "ignore")
# address common error where people take data that is already
# utf-8, presume that it is iso-8859-1, and re-encode it.
if self.encoding in ('utf-8', 'utf-8_INVALID_PYTHON_3') and not isinstance(output, bytes):
if self.encoding in ("utf-8", "utf-8_INVALID_PYTHON_3") and not isinstance(
output, bytes
):
try:
output = output.encode('iso-8859-1').decode('utf-8')
output = output.encode("iso-8859-1").decode("utf-8")
except (UnicodeEncodeError, UnicodeDecodeError):
pass
@ -567,65 +604,74 @@ class XMLParserMixin(
# categories/tags/keywords/whatever are handled in _end_category or
# _end_tags or _end_itunes_keywords
if element in ('category', 'tags', 'itunes_keywords'):
if element in ("category", "tags", "itunes_keywords"):
return output
if element == 'title' and -1 < self.title_depth <= self.depth:
if element == "title" and -1 < self.title_depth <= self.depth:
return output
# store output in appropriate place(s)
if self.inentry and not self.insource:
if element == 'content':
if element == "content":
self.entries[-1].setdefault(element, [])
contentparams = copy.deepcopy(self.contentparams)
contentparams['value'] = output
contentparams["value"] = output
self.entries[-1][element].append(contentparams)
elif element == 'link':
elif element == "link":
if not self.inimage:
# query variables in urls in link elements are improperly
# converted from `?a=1&b=2` to `?a=1&b;=2` as if they're
# unhandled character references. fix this special case.
output = output.replace('&amp;', '&')
output = output.replace("&amp;", "&")
output = re.sub("&([A-Za-z0-9_]+);", r"&\g<1>", output)
self.entries[-1][element] = output
if output:
self.entries[-1]['links'][-1]['href'] = output
self.entries[-1]["links"][-1]["href"] = output
else:
if element == 'description':
element = 'summary'
old_value_depth = self.property_depth_map.setdefault(self.entries[-1], {}).get(element)
if element == "description":
element = "summary"
old_value_depth = self.property_depth_map.setdefault(
self.entries[-1], {}
).get(element)
if old_value_depth is None or self.depth <= old_value_depth:
self.property_depth_map[self.entries[-1]][element] = self.depth
self.entries[-1][element] = output
if self.incontent:
contentparams = copy.deepcopy(self.contentparams)
contentparams['value'] = output
self.entries[-1][element + '_detail'] = contentparams
elif self.infeed or self.insource: # and (not self.intextinput) and (not self.inimage):
contentparams["value"] = output
self.entries[-1][element + "_detail"] = contentparams
elif (
self.infeed or self.insource
): # and (not self.intextinput) and (not self.inimage):
context = self._get_context()
if element == 'description':
element = 'subtitle'
if element == "description":
element = "subtitle"
context[element] = output
if element == 'link':
if element == "link":
# fix query variables; see above for the explanation
output = re.sub("&([A-Za-z0-9_]+);", r"&\g<1>", output)
context[element] = output
context['links'][-1]['href'] = output
context["links"][-1]["href"] = output
elif self.incontent:
contentparams = copy.deepcopy(self.contentparams)
contentparams['value'] = output
context[element + '_detail'] = contentparams
contentparams["value"] = output
context[element + "_detail"] = contentparams
return output
def push_content(self, tag, attrs_d, default_content_type, expecting_text):
self.incontent += 1
if self.lang:
self.lang = self.lang.replace('_', '-')
self.contentparams = FeedParserDict({
'type': self.map_content_type(attrs_d.get('type', default_content_type)),
'language': self.lang,
'base': self.baseuri})
self.contentparams['base64'] = self._is_base64(attrs_d, self.contentparams)
self.lang = self.lang.replace("_", "-")
self.contentparams = FeedParserDict(
{
"type": self.map_content_type(
attrs_d.get("type", default_content_type)
),
"language": self.lang,
"base": self.baseuri,
}
)
self.contentparams["base64"] = self._is_base64(attrs_d, self.contentparams)
self.push(tag, expecting_text)
def pop_content(self, tag):
@ -646,55 +692,61 @@ class XMLParserMixin(
"""
# must have a close tag or an entity reference to qualify
if not (re.search(r'</(\w+)>', s) or re.search(r'&#?\w+;', s)):
if not (re.search(r"</(\w+)>", s) or re.search(r"&#?\w+;", s)):
return False
# all tags must be in a restricted subset of valid HTML tags
if any((t for t in re.findall(r'</?(\w+)', s) if t.lower() not in HTMLSanitizer.acceptable_elements)):
if any(
t
for t in re.findall(r"</?(\w+)", s)
if t.lower() not in HTMLSanitizer.acceptable_elements
):
return False
# all entities must have been defined as valid HTML entities
if any((e for e in re.findall(r'&(\w+);', s) if e not in html.entities.entitydefs)):
if any(
e for e in re.findall(r"&(\w+);", s) if e not in html.entities.entitydefs
):
return False
return True
def _map_to_standard_prefix(self, name):
colonpos = name.find(':')
colonpos = name.find(":")
if colonpos != -1:
prefix = name[:colonpos]
suffix = name[colonpos+1:]
suffix = name[colonpos + 1 :]
prefix = self.namespacemap.get(prefix, prefix)
name = prefix + ':' + suffix
name = prefix + ":" + suffix
return name
def _get_attribute(self, attrs_d, name):
return attrs_d.get(self._map_to_standard_prefix(name))
def _is_base64(self, attrs_d, contentparams):
if attrs_d.get('mode', '') == 'base64':
if attrs_d.get("mode", "") == "base64":
return 1
if self.contentparams['type'].startswith('text/'):
if self.contentparams["type"].startswith("text/"):
return 0
if self.contentparams['type'].endswith('+xml'):
if self.contentparams["type"].endswith("+xml"):
return 0
if self.contentparams['type'].endswith('/xml'):
if self.contentparams["type"].endswith("/xml"):
return 0
return 1
@staticmethod
def _enforce_href(attrs_d):
href = attrs_d.get('url', attrs_d.get('uri', attrs_d.get('href', None)))
href = attrs_d.get("url", attrs_d.get("uri", attrs_d.get("href", None)))
if href:
try:
del attrs_d['url']
del attrs_d["url"]
except KeyError:
pass
try:
del attrs_d['uri']
del attrs_d["uri"]
except KeyError:
pass
attrs_d['href'] = href
attrs_d["href"] = href
return attrs_d
def _save(self, key, value, overwrite=False):
@ -707,37 +759,37 @@ class XMLParserMixin(
def _get_context(self):
if self.insource:
context = self.sourcedata
elif self.inimage and 'image' in self.feeddata:
context = self.feeddata['image']
elif self.inimage and "image" in self.feeddata:
context = self.feeddata["image"]
elif self.intextinput:
context = self.feeddata['textinput']
context = self.feeddata["textinput"]
elif self.inentry:
context = self.entries[-1]
else:
context = self.feeddata
return context
def _save_author(self, key, value, prefix='author'):
def _save_author(self, key, value, prefix="author"):
context = self._get_context()
context.setdefault(prefix + '_detail', FeedParserDict())
context[prefix + '_detail'][key] = value
context.setdefault(prefix + "_detail", FeedParserDict())
context[prefix + "_detail"][key] = value
self._sync_author_detail()
context.setdefault('authors', [FeedParserDict()])
context['authors'][-1][key] = value
context.setdefault("authors", [FeedParserDict()])
context["authors"][-1][key] = value
def _save_contributor(self, key, value):
context = self._get_context()
context.setdefault('contributors', [FeedParserDict()])
context['contributors'][-1][key] = value
context.setdefault("contributors", [FeedParserDict()])
context["contributors"][-1][key] = value
def _sync_author_detail(self, key='author'):
def _sync_author_detail(self, key="author"):
context = self._get_context()
detail = context.get('%ss' % key, [FeedParserDict()])[-1]
detail = context.get("%ss" % key, [FeedParserDict()])[-1]
if detail:
name = detail.get('name')
email = detail.get('email')
name = detail.get("name")
email = detail.get("email")
if name and email:
context[key] = '%s (%s)' % (name, email)
context[key] = f"{name} ({email})"
elif name:
context[key] = name
elif email:
@ -746,31 +798,31 @@ class XMLParserMixin(
author, email = context.get(key), None
if not author:
return
emailmatch = re.search(r"(([a-zA-Z0-9_.+-]+)@((\[[0-9]{1,3}\.[0-9]{1,3}\.[0-9]{1,3}\.)|(([a-zA-Z0-9-]+\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(]?))(\?subject=\S+)?", author)
emailmatch = email_pattern.search(author)
if emailmatch:
email = emailmatch.group(0)
# probably a better way to do the following, but it passes
# all the tests
author = author.replace(email, '')
author = author.replace('()', '')
author = author.replace('<>', '')
author = author.replace('&lt;&gt;', '')
author = author.replace(email, "")
author = author.replace("()", "")
author = author.replace("<>", "")
author = author.replace("&lt;&gt;", "")
author = author.strip()
if author and (author[0] == '('):
if author and (author[0] == "("):
author = author[1:]
if author and (author[-1] == ')'):
if author and (author[-1] == ")"):
author = author[:-1]
author = author.strip()
if author or email:
context.setdefault('%s_detail' % key, detail)
context.setdefault("%s_detail" % key, detail)
if author:
detail['name'] = author
detail["name"] = author
if email:
detail['email'] = email
detail["email"] = email
def _add_tag(self, term, scheme, label):
context = self._get_context()
tags = context.setdefault('tags', [])
tags = context.setdefault("tags", [])
if (not term) and (not scheme) and (not label):
return
value = FeedParserDict(term=term, scheme=scheme, label=label)
@ -781,8 +833,8 @@ class XMLParserMixin(
# This is a completely-made up element. Its semantics are determined
# only by a single feed that precipitated bug report 392 on Google Code.
# In short, this is junk code.
self.push('tags', 1)
self.push("tags", 1)
def _end_tags(self):
for term in self.pop('tags').split(','):
for term in self.pop("tags").split(","):
self._add_tag(term.strip(), None, None)

366
lib/feedparser/namespaces/_base.py
View file

@ -1,5 +1,5 @@
# Support for the Atom, RSS, RDF, and CDF feed formats
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -33,7 +33,7 @@ from ..urls import make_safe_absolute_uri
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
"""Support for the Atom, RSS, RDF, and CDF feed formats.
The feed formats all share common elements, some of which have conflicting
@ -42,452 +42,490 @@ class Namespace(object):
"""
supported_namespaces = {
'': '',
'http://backend.userland.com/rss': '',
'http://blogs.law.harvard.edu/tech/rss': '',
'http://purl.org/rss/1.0/': '',
'http://my.netscape.com/rdf/simple/0.9/': '',
'http://example.com/newformat#': '',
'http://example.com/necho': '',
'http://purl.org/echo/': '',
'uri/of/echo/namespace#': '',
'http://purl.org/pie/': '',
'http://purl.org/atom/ns#': '',
'http://www.w3.org/2005/Atom': '',
'http://purl.org/rss/1.0/modules/rss091#': '',
"": "",
"http://backend.userland.com/rss": "",
"http://blogs.law.harvard.edu/tech/rss": "",
"http://purl.org/rss/1.0/": "",
"http://my.netscape.com/rdf/simple/0.9/": "",
"http://example.com/newformat#": "",
"http://example.com/necho": "",
"http://purl.org/echo/": "",
"uri/of/echo/namespace#": "",
"http://purl.org/pie/": "",
"http://purl.org/atom/ns#": "",
"http://www.w3.org/2005/Atom": "",
"http://purl.org/rss/1.0/modules/rss091#": "",
}
def _start_rss(self, attrs_d):
versionmap = {
'0.91': 'rss091u',
'0.92': 'rss092',
'0.93': 'rss093',
'0.94': 'rss094',
"0.91": "rss091u",
"0.92": "rss092",
"0.93": "rss093",
"0.94": "rss094",
}
# If we're here then this is an RSS feed.
# If we don't have a version or have a version that starts with something
# other than RSS then there's been a mistake. Correct it.
if not self.version or not self.version.startswith('rss'):
attr_version = attrs_d.get('version', '')
if not self.version or not self.version.startswith("rss"):
attr_version = attrs_d.get("version", "")
version = versionmap.get(attr_version)
if version:
self.version = version
elif attr_version.startswith('2.'):
self.version = 'rss20'
elif attr_version.startswith("2."):
self.version = "rss20"
else:
self.version = 'rss'
self.version = "rss"
def _start_channel(self, attrs_d):
self.infeed = 1
self._cdf_common(attrs_d)
def _cdf_common(self, attrs_d):
if 'lastmod' in attrs_d:
if "lastmod" in attrs_d:
self._start_modified({})
self.elementstack[-1][-1] = attrs_d['lastmod']
self.elementstack[-1][-1] = attrs_d["lastmod"]
self._end_modified()
if 'href' in attrs_d:
if "href" in attrs_d:
self._start_link({})
self.elementstack[-1][-1] = attrs_d['href']
self.elementstack[-1][-1] = attrs_d["href"]
self._end_link()
def _start_feed(self, attrs_d):
self.infeed = 1
versionmap = {'0.1': 'atom01',
'0.2': 'atom02',
'0.3': 'atom03'}
versionmap = {"0.1": "atom01", "0.2": "atom02", "0.3": "atom03"}
if not self.version:
attr_version = attrs_d.get('version')
attr_version = attrs_d.get("version")
version = versionmap.get(attr_version)
if version:
self.version = version
else:
self.version = 'atom'
self.version = "atom"
def _end_channel(self):
self.infeed = 0
_end_feed = _end_channel
def _start_image(self, attrs_d):
context = self._get_context()
if not self.inentry:
context.setdefault('image', FeedParserDict())
context.setdefault("image", FeedParserDict())
self.inimage = 1
self.title_depth = -1
self.push('image', 0)
self.push("image", 0)
def _end_image(self):
self.pop('image')
self.pop("image")
self.inimage = 0
def _start_textinput(self, attrs_d):
context = self._get_context()
context.setdefault('textinput', FeedParserDict())
context.setdefault("textinput", FeedParserDict())
self.intextinput = 1
self.title_depth = -1
self.push('textinput', 0)
self.push("textinput", 0)
_start_textInput = _start_textinput
def _end_textinput(self):
self.pop('textinput')
self.pop("textinput")
self.intextinput = 0
_end_textInput = _end_textinput
def _start_author(self, attrs_d):
self.inauthor = 1
self.push('author', 1)
self.push("author", 1)
# Append a new FeedParserDict when expecting an author
context = self._get_context()
context.setdefault('authors', [])
context['authors'].append(FeedParserDict())
context.setdefault("authors", [])
context["authors"].append(FeedParserDict())
_start_managingeditor = _start_author
def _end_author(self):
self.pop('author')
self.pop("author")
self.inauthor = 0
self._sync_author_detail()
_end_managingeditor = _end_author
def _start_contributor(self, attrs_d):
self.incontributor = 1
context = self._get_context()
context.setdefault('contributors', [])
context['contributors'].append(FeedParserDict())
self.push('contributor', 0)
context.setdefault("contributors", [])
context["contributors"].append(FeedParserDict())
self.push("contributor", 0)
def _end_contributor(self):
self.pop('contributor')
self.pop("contributor")
self.incontributor = 0
def _start_name(self, attrs_d):
self.push('name', 0)
self.push("name", 0)
def _end_name(self):
value = self.pop('name')
value = self.pop("name")
if self.inpublisher:
self._save_author('name', value, 'publisher')
self._save_author("name", value, "publisher")
elif self.inauthor:
self._save_author('name', value)
self._save_author("name", value)
elif self.incontributor:
self._save_contributor('name', value)
self._save_contributor("name", value)
elif self.intextinput:
context = self._get_context()
context['name'] = value
context["name"] = value
def _start_width(self, attrs_d):
self.push('width', 0)
self.push("width", 0)
def _end_width(self):
value = self.pop('width')
value = self.pop("width")
try:
value = int(value)
except ValueError:
value = 0
if self.inimage:
context = self._get_context()
context['width'] = value
context["width"] = value
def _start_height(self, attrs_d):
self.push('height', 0)
self.push("height", 0)
def _end_height(self):
value = self.pop('height')
value = self.pop("height")
try:
value = int(value)
except ValueError:
value = 0
if self.inimage:
context = self._get_context()
context['height'] = value
context["height"] = value
def _start_url(self, attrs_d):
self.push('href', 1)
self.push("href", 1)
_start_homepage = _start_url
_start_uri = _start_url
def _end_url(self):
value = self.pop('href')
value = self.pop("href")
if self.inauthor:
self._save_author('href', value)
self._save_author("href", value)
elif self.incontributor:
self._save_contributor('href', value)
self._save_contributor("href", value)
_end_homepage = _end_url
_end_uri = _end_url
def _start_email(self, attrs_d):
self.push('email', 0)
self.push("email", 0)
def _end_email(self):
value = self.pop('email')
value = self.pop("email")
if self.inpublisher:
self._save_author('email', value, 'publisher')
self._save_author("email", value, "publisher")
elif self.inauthor:
self._save_author('email', value)
self._save_author("email", value)
elif self.incontributor:
self._save_contributor('email', value)
self._save_contributor("email", value)
def _start_subtitle(self, attrs_d):
self.push_content('subtitle', attrs_d, 'text/plain', 1)
self.push_content("subtitle", attrs_d, "text/plain", 1)
_start_tagline = _start_subtitle
def _end_subtitle(self):
self.pop_content('subtitle')
self.pop_content("subtitle")
_end_tagline = _end_subtitle
def _start_rights(self, attrs_d):
self.push_content('rights', attrs_d, 'text/plain', 1)
self.push_content("rights", attrs_d, "text/plain", 1)
_start_copyright = _start_rights
def _end_rights(self):
self.pop_content('rights')
self.pop_content("rights")
_end_copyright = _end_rights
def _start_item(self, attrs_d):
self.entries.append(FeedParserDict())
self.push('item', 0)
self.push("item", 0)
self.inentry = 1
self.guidislink = 0
self.title_depth = -1
id = self._get_attribute(attrs_d, 'rdf:about')
id = self._get_attribute(attrs_d, "rdf:about")
if id:
context = self._get_context()
context['id'] = id
context["id"] = id
self._cdf_common(attrs_d)
_start_entry = _start_item
def _end_item(self):
self.pop('item')
self.pop("item")
self.inentry = 0
self.hasContent = 0
_end_entry = _end_item
def _start_language(self, attrs_d):
self.push('language', 1)
self.push("language", 1)
def _end_language(self):
self.lang = self.pop('language')
self.lang = self.pop("language")
def _start_webmaster(self, attrs_d):
self.push('publisher', 1)
self.push("publisher", 1)
def _end_webmaster(self):
self.pop('publisher')
self._sync_author_detail('publisher')
self.pop("publisher")
self._sync_author_detail("publisher")
def _start_published(self, attrs_d):
self.push('published', 1)
self.push("published", 1)
_start_issued = _start_published
_start_pubdate = _start_published
def _end_published(self):
value = self.pop('published')
self._save('published_parsed', _parse_date(value), overwrite=True)
value = self.pop("published")
self._save("published_parsed", _parse_date(value), overwrite=True)
_end_issued = _end_published
_end_pubdate = _end_published
def _start_updated(self, attrs_d):
self.push('updated', 1)
self.push("updated", 1)
_start_modified = _start_updated
_start_lastbuilddate = _start_updated
def _end_updated(self):
value = self.pop('updated')
value = self.pop("updated")
parsed_value = _parse_date(value)
self._save('updated_parsed', parsed_value, overwrite=True)
self._save("updated_parsed", parsed_value, overwrite=True)
_end_modified = _end_updated
_end_lastbuilddate = _end_updated
def _start_created(self, attrs_d):
self.push('created', 1)
self.push("created", 1)
def _end_created(self):
value = self.pop('created')
self._save('created_parsed', _parse_date(value), overwrite=True)
value = self.pop("created")
self._save("created_parsed", _parse_date(value), overwrite=True)
def _start_expirationdate(self, attrs_d):
self.push('expired', 1)
self.push("expired", 1)
def _end_expirationdate(self):
self._save('expired_parsed', _parse_date(self.pop('expired')), overwrite=True)
self._save("expired_parsed", _parse_date(self.pop("expired")), overwrite=True)
def _start_category(self, attrs_d):
term = attrs_d.get('term')
scheme = attrs_d.get('scheme', attrs_d.get('domain'))
label = attrs_d.get('label')
term = attrs_d.get("term")
scheme = attrs_d.get("scheme", attrs_d.get("domain"))
label = attrs_d.get("label")
self._add_tag(term, scheme, label)
self.push('category', 1)
self.push("category", 1)
_start_keywords = _start_category
def _end_category(self):
value = self.pop('category')
value = self.pop("category")
if not value:
return
context = self._get_context()
tags = context['tags']
if value and len(tags) and not tags[-1]['term']:
tags[-1]['term'] = value
tags = context["tags"]
if value and len(tags) and not tags[-1]["term"]:
tags[-1]["term"] = value
else:
self._add_tag(value, None, None)
_end_keywords = _end_category
def _start_cloud(self, attrs_d):
self._get_context()['cloud'] = FeedParserDict(attrs_d)
self._get_context()["cloud"] = FeedParserDict(attrs_d)
def _start_link(self, attrs_d):
attrs_d.setdefault('rel', 'alternate')
if attrs_d['rel'] == 'self':
attrs_d.setdefault('type', 'application/atom+xml')
attrs_d.setdefault("rel", "alternate")
if attrs_d["rel"] == "self":
attrs_d.setdefault("type", "application/atom+xml")
else:
attrs_d.setdefault('type', 'text/html')
attrs_d.setdefault("type", "text/html")
context = self._get_context()
attrs_d = self._enforce_href(attrs_d)
if 'href' in attrs_d:
attrs_d['href'] = self.resolve_uri(attrs_d['href'])
if "href" in attrs_d:
attrs_d["href"] = self.resolve_uri(attrs_d["href"])
expecting_text = self.infeed or self.inentry or self.insource
context.setdefault('links', [])
context.setdefault("links", [])
if not (self.inentry and self.inimage):
context['links'].append(FeedParserDict(attrs_d))
if 'href' in attrs_d:
context["links"].append(FeedParserDict(attrs_d))
if "href" in attrs_d:
if (
attrs_d.get('rel') == 'alternate'
and self.map_content_type(attrs_d.get('type')) in self.html_types
attrs_d.get("rel") == "alternate"
and self.map_content_type(attrs_d.get("type")) in self.html_types
):
context['link'] = attrs_d['href']
context["link"] = attrs_d["href"]
else:
self.push('link', expecting_text)
self.push("link", expecting_text)
def _end_link(self):
self.pop('link')
self.pop("link")
def _start_guid(self, attrs_d):
self.guidislink = (attrs_d.get('ispermalink', 'true') == 'true')
self.push('id', 1)
self.guidislink = attrs_d.get("ispermalink", "true") == "true"
self.push("id", 1)
_start_id = _start_guid
def _end_guid(self):
value = self.pop('id')
self._save('guidislink', self.guidislink and 'link' not in self._get_context())
value = self.pop("id")
self._save("guidislink", self.guidislink and "link" not in self._get_context())
if self.guidislink:
# guid acts as link, but only if 'ispermalink' is not present or is 'true',
# and only if the item doesn't already have a link element
self._save('link', value)
self._save("link", value)
_end_id = _end_guid
def _start_title(self, attrs_d):
if self.svgOK:
return self.unknown_starttag('title', list(attrs_d.items()))
self.push_content('title', attrs_d, 'text/plain', self.infeed or self.inentry or self.insource)
return self.unknown_starttag("title", list(attrs_d.items()))
self.push_content(
"title", attrs_d, "text/plain", self.infeed or self.inentry or self.insource
)
def _end_title(self):
if self.svgOK:
return
value = self.pop_content('title')
value = self.pop_content("title")
if not value:
return
self.title_depth = self.depth
def _start_description(self, attrs_d):
context = self._get_context()
if 'summary' in context and not self.hasContent:
self._summaryKey = 'content'
if "summary" in context and not self.hasContent:
self._summaryKey = "content"
self._start_content(attrs_d)
else:
self.push_content('description', attrs_d, 'text/html', self.infeed or self.inentry or self.insource)
self.push_content(
"description",
attrs_d,
"text/html",
self.infeed or self.inentry or self.insource,
)
def _start_abstract(self, attrs_d):
self.push_content('description', attrs_d, 'text/plain', self.infeed or self.inentry or self.insource)
self.push_content(
"description",
attrs_d,
"text/plain",
self.infeed or self.inentry or self.insource,
)
def _end_description(self):
if self._summaryKey == 'content':
if self._summaryKey == "content":
self._end_content()
else:
self.pop_content('description')
self.pop_content("description")
self._summaryKey = None
_end_abstract = _end_description
def _start_info(self, attrs_d):
self.push_content('info', attrs_d, 'text/plain', 1)
self.push_content("info", attrs_d, "text/plain", 1)
_start_feedburner_browserfriendly = _start_info
def _end_info(self):
self.pop_content('info')
self.pop_content("info")
_end_feedburner_browserfriendly = _end_info
def _start_generator(self, attrs_d):
if attrs_d:
attrs_d = self._enforce_href(attrs_d)
if 'href' in attrs_d:
attrs_d['href'] = self.resolve_uri(attrs_d['href'])
self._get_context()['generator_detail'] = FeedParserDict(attrs_d)
self.push('generator', 1)
if "href" in attrs_d:
attrs_d["href"] = self.resolve_uri(attrs_d["href"])
self._get_context()["generator_detail"] = FeedParserDict(attrs_d)
self.push("generator", 1)
def _end_generator(self):
value = self.pop('generator')
value = self.pop("generator")
context = self._get_context()
if 'generator_detail' in context:
context['generator_detail']['name'] = value
if "generator_detail" in context:
context["generator_detail"]["name"] = value
def _start_summary(self, attrs_d):
context = self._get_context()
if 'summary' in context and not self.hasContent:
self._summaryKey = 'content'
if "summary" in context and not self.hasContent:
self._summaryKey = "content"
self._start_content(attrs_d)
else:
self._summaryKey = 'summary'
self.push_content(self._summaryKey, attrs_d, 'text/plain', 1)
self._summaryKey = "summary"
self.push_content(self._summaryKey, attrs_d, "text/plain", 1)
def _end_summary(self):
if self._summaryKey == 'content':
if self._summaryKey == "content":
self._end_content()
else:
self.pop_content(self._summaryKey or 'summary')
self.pop_content(self._summaryKey or "summary")
self._summaryKey = None
def _start_enclosure(self, attrs_d):
attrs_d = self._enforce_href(attrs_d)
context = self._get_context()
attrs_d['rel'] = 'enclosure'
context.setdefault('links', []).append(FeedParserDict(attrs_d))
attrs_d["rel"] = "enclosure"
context.setdefault("links", []).append(FeedParserDict(attrs_d))
def _start_source(self, attrs_d):
if 'url' in attrs_d:
if "url" in attrs_d:
# This means that we're processing a source element from an RSS 2.0 feed
self.sourcedata['href'] = attrs_d['url']
self.push('source', 1)
self.sourcedata["href"] = attrs_d["url"]
self.push("source", 1)
self.insource = 1
self.title_depth = -1
def _end_source(self):
self.insource = 0
value = self.pop('source')
value = self.pop("source")
if value:
self.sourcedata['title'] = value
self._get_context()['source'] = copy.deepcopy(self.sourcedata)
self.sourcedata["title"] = value
self._get_context()["source"] = copy.deepcopy(self.sourcedata)
self.sourcedata.clear()
def _start_content(self, attrs_d):
self.hasContent = 1
self.push_content('content', attrs_d, 'text/plain', 1)
src = attrs_d.get('src')
self.push_content("content", attrs_d, "text/plain", 1)
src = attrs_d.get("src")
if src:
self.contentparams['src'] = src
self.push('content', 1)
self.contentparams["src"] = src
self.push("content", 1)
def _start_body(self, attrs_d):
self.push_content('content', attrs_d, 'application/xhtml+xml', 1)
self.push_content("content", attrs_d, "application/xhtml+xml", 1)
_start_xhtml_body = _start_body
def _start_content_encoded(self, attrs_d):
self.hasContent = 1
self.push_content('content', attrs_d, 'text/html', 1)
self.push_content("content", attrs_d, "text/html", 1)
_start_fullitem = _start_content_encoded
def _end_content(self):
copyToSummary = self.map_content_type(self.contentparams.get('type')) in ({'text/plain'} | self.html_types)
value = self.pop_content('content')
copyToSummary = self.map_content_type(self.contentparams.get("type")) in (
{"text/plain"} | self.html_types
)
value = self.pop_content("content")
if copyToSummary:
self._save('summary', value)
self._save("summary", value)
_end_body = _end_content
_end_xhtml_body = _end_content
@ -495,12 +533,12 @@ class Namespace(object):
_end_fullitem = _end_content
def _start_newlocation(self, attrs_d):
self.push('newlocation', 1)
self.push("newlocation", 1)
def _end_newlocation(self):
url = self.pop('newlocation')
url = self.pop("newlocation")
context = self._get_context()
# don't set newlocation if the context isn't right
if context is not self.feeddata:
return
context['newlocation'] = make_safe_absolute_uri(self.baseuri, url.strip())
context["newlocation"] = make_safe_absolute_uri(self.baseuri, url.strip())

20
lib/feedparser/namespaces/admin.py
View file

@ -1,5 +1,5 @@
# Support for the administrative elements extension
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -29,25 +29,25 @@
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
# RDF Site Summary 1.0 Modules: Administrative
# http://web.resource.org/rss/1.0/modules/admin/
supported_namespaces = {
'http://webns.net/mvcb/': 'admin',
"http://webns.net/mvcb/": "admin",
}
def _start_admin_generatoragent(self, attrs_d):
self.push('generator', 1)
value = self._get_attribute(attrs_d, 'rdf:resource')
self.push("generator", 1)
value = self._get_attribute(attrs_d, "rdf:resource")
if value:
self.elementstack[-1][2].append(value)
self.pop('generator')
self._get_context()['generator_detail'] = FeedParserDict({'href': value})
self.pop("generator")
self._get_context()["generator_detail"] = FeedParserDict({"href": value})
def _start_admin_errorreportsto(self, attrs_d):
self.push('errorreportsto', 1)
value = self._get_attribute(attrs_d, 'rdf:resource')
self.push("errorreportsto", 1)
value = self._get_attribute(attrs_d, "rdf:resource")
if value:
self.elementstack[-1][2].append(value)
self.pop('errorreportsto')
self.pop("errorreportsto")

39
lib/feedparser/namespaces/cc.py
View file

@ -1,5 +1,5 @@
# Support for the Creative Commons licensing extensions
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -29,41 +29,42 @@
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
supported_namespaces = {
# RDF-based namespace
'http://creativecommons.org/ns#license': 'cc',
"http://creativecommons.org/ns#license": "cc",
# Old RDF-based namespace
'http://web.resource.org/cc/': 'cc',
"http://web.resource.org/cc/": "cc",
# RSS-based namespace
'http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html': 'creativecommons',
"http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html": (
"creativecommons"
),
# Old RSS-based namespace
'http://backend.userland.com/creativeCommonsRssModule': 'creativecommons',
"http://backend.userland.com/creativeCommonsRssModule": "creativecommons",
}
def _start_cc_license(self, attrs_d):
context = self._get_context()
value = self._get_attribute(attrs_d, 'rdf:resource')
value = self._get_attribute(attrs_d, "rdf:resource")
attrs_d = FeedParserDict()
attrs_d['rel'] = 'license'
attrs_d["rel"] = "license"
if value:
attrs_d['href'] = value
context.setdefault('links', []).append(attrs_d)
attrs_d["href"] = value
context.setdefault("links", []).append(attrs_d)
def _start_creativecommons_license(self, attrs_d):
self.push('license', 1)
self.push("license", 1)
_start_creativeCommons_license = _start_creativecommons_license
def _end_creativecommons_license(self):
value = self.pop('license')
value = self.pop("license")
context = self._get_context()
attrs_d = FeedParserDict()
attrs_d['rel'] = 'license'
attrs_d["rel"] = "license"
if value:
attrs_d['href'] = value
context.setdefault('links', []).append(attrs_d)
del context['license']
attrs_d["href"] = value
context.setdefault("links", []).append(attrs_d)
del context["license"]
_end_creativeCommons_license = _end_creativecommons_license

38
lib/feedparser/namespaces/dc.py
View file

@ -1,5 +1,5 @@
# Support for the Dublin Core metadata extensions
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -30,10 +30,10 @@ from ..datetimes import _parse_date
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
supported_namespaces = {
'http://purl.org/dc/elements/1.1/': 'dc',
'http://purl.org/dc/terms/': 'dcterms',
"http://purl.org/dc/elements/1.1/": "dc",
"http://purl.org/dc/terms/": "dcterms",
}
def _end_dc_author(self):
@ -109,25 +109,29 @@ class Namespace(object):
self._start_updated(attrs_d)
def _start_dcterms_valid(self, attrs_d):
self.push('validity', 1)
self.push("validity", 1)
def _end_dcterms_valid(self):
for validity_detail in self.pop('validity').split(';'):
if '=' in validity_detail:
key, value = validity_detail.split('=', 1)
if key == 'start':
self._save('validity_start', value, overwrite=True)
self._save('validity_start_parsed', _parse_date(value), overwrite=True)
elif key == 'end':
self._save('validity_end', value, overwrite=True)
self._save('validity_end_parsed', _parse_date(value), overwrite=True)
for validity_detail in self.pop("validity").split(";"):
if "=" in validity_detail:
key, value = validity_detail.split("=", 1)
if key == "start":
self._save("validity_start", value, overwrite=True)
self._save(
"validity_start_parsed", _parse_date(value), overwrite=True
)
elif key == "end":
self._save("validity_end", value, overwrite=True)
self._save(
"validity_end_parsed", _parse_date(value), overwrite=True
)
def _start_dc_contributor(self, attrs_d):
self.incontributor = 1
context = self._get_context()
context.setdefault('contributors', [])
context['contributors'].append(FeedParserDict())
self.push('name', 0)
context.setdefault("contributors", [])
context["contributors"].append(FeedParserDict())
self.push("name", 0)
def _end_dc_contributor(self):
self._end_name()

577
lib/feedparser/namespaces/georss.py
View file

@ -1,5 +1,5 @@
# Support for the GeoRSS format
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -26,27 +26,24 @@
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
# Required for Python 3.6 compatibility.
from __future__ import generator_stop
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
supported_namespaces = {
'http://www.w3.org/2003/01/geo/wgs84_pos#': 'geo',
'http://www.georss.org/georss': 'georss',
'http://www.opengis.net/gml': 'gml',
"http://www.w3.org/2003/01/geo/wgs84_pos#": "geo",
"http://www.georss.org/georss": "georss",
"http://www.opengis.net/gml": "gml",
}
def __init__(self):
self.ingeometry = 0
super(Namespace, self).__init__()
super().__init__()
def _start_georssgeom(self, attrs_d):
self.push('geometry', 0)
self.push("geometry", 0)
context = self._get_context()
context['where'] = FeedParserDict()
context["where"] = FeedParserDict()
_start_georss_point = _start_georssgeom
_start_georss_line = _start_georssgeom
@ -55,76 +52,77 @@ class Namespace(object):
def _save_where(self, geometry):
context = self._get_context()
context['where'].update(geometry)
context["where"].update(geometry)
def _end_georss_point(self):
geometry = _parse_georss_point(self.pop('geometry'))
geometry = _parse_georss_point(self.pop("geometry"))
if geometry:
self._save_where(geometry)
def _end_georss_line(self):
geometry = _parse_georss_line(self.pop('geometry'))
geometry = _parse_georss_line(self.pop("geometry"))
if geometry:
self._save_where(geometry)
def _end_georss_polygon(self):
this = self.pop('geometry')
this = self.pop("geometry")
geometry = _parse_georss_polygon(this)
if geometry:
self._save_where(geometry)
def _end_georss_box(self):
geometry = _parse_georss_box(self.pop('geometry'))
geometry = _parse_georss_box(self.pop("geometry"))
if geometry:
self._save_where(geometry)
def _start_where(self, attrs_d):
self.push('where', 0)
self.push("where", 0)
context = self._get_context()
context['where'] = FeedParserDict()
context["where"] = FeedParserDict()
_start_georss_where = _start_where
def _parse_srs_attrs(self, attrs_d):
srs_name = attrs_d.get('srsname')
srs_name = attrs_d.get("srsname")
try:
srs_dimension = int(attrs_d.get('srsdimension', '2'))
srs_dimension = int(attrs_d.get("srsdimension", "2"))
except ValueError:
srs_dimension = 2
context = self._get_context()
if 'where' not in context:
context['where'] = {}
context['where']['srsName'] = srs_name
context['where']['srsDimension'] = srs_dimension
if "where" not in context:
context["where"] = {}
context["where"]["srsName"] = srs_name
context["where"]["srsDimension"] = srs_dimension
def _start_gml_point(self, attrs_d):
self._parse_srs_attrs(attrs_d)
self.ingeometry = 1
self.push('geometry', 0)
self.push("geometry", 0)
def _start_gml_linestring(self, attrs_d):
self._parse_srs_attrs(attrs_d)
self.ingeometry = 'linestring'
self.push('geometry', 0)
self.ingeometry = "linestring"
self.push("geometry", 0)
def _start_gml_polygon(self, attrs_d):
self._parse_srs_attrs(attrs_d)
self.push('geometry', 0)
self.push("geometry", 0)
def _start_gml_exterior(self, attrs_d):
self.push('geometry', 0)
self.push("geometry", 0)
def _start_gml_linearring(self, attrs_d):
self.ingeometry = 'polygon'
self.push('geometry', 0)
self.ingeometry = "polygon"
self.push("geometry", 0)
def _start_gml_pos(self, attrs_d):
self.push('pos', 0)
self.push("pos", 0)
def _end_gml_pos(self):
this = self.pop('pos')
this = self.pop("pos")
context = self._get_context()
srs_name = context['where'].get('srsName')
srs_dimension = context['where'].get('srsDimension', 2)
srs_name = context["where"].get("srsName")
srs_dimension = context["where"].get("srsDimension", 2)
swap = True
if srs_name and "EPSG" in srs_name:
epsg = int(srs_name.split(":")[-1])
@ -134,25 +132,25 @@ class Namespace(object):
self._save_where(geometry)
def _start_gml_poslist(self, attrs_d):
self.push('pos', 0)
self.push("pos", 0)
def _end_gml_poslist(self):
this = self.pop('pos')
this = self.pop("pos")
context = self._get_context()
srs_name = context['where'].get('srsName')
srs_dimension = context['where'].get('srsDimension', 2)
srs_name = context["where"].get("srsName")
srs_dimension = context["where"].get("srsDimension", 2)
swap = True
if srs_name and "EPSG" in srs_name:
epsg = int(srs_name.split(":")[-1])
swap = bool(epsg in _geogCS)
geometry = _parse_poslist(
this, self.ingeometry, swap=swap, dims=srs_dimension)
geometry = _parse_poslist(this, self.ingeometry, swap=swap, dims=srs_dimension)
if geometry:
self._save_where(geometry)
def _end_geom(self):
self.ingeometry = 0
self.pop('geometry')
self.pop("geometry")
_end_gml_point = _end_geom
_end_gml_linestring = _end_geom
_end_gml_linearring = _end_geom
@ -160,19 +158,21 @@ class Namespace(object):
_end_gml_polygon = _end_geom
def _end_where(self):
self.pop('where')
self.pop("where")
_end_georss_where = _end_where
# GeoRSS geometry parsers. Each return a dict with 'type' and 'coordinates'
# items, or None in the case of a parsing error.
def _parse_poslist(value, geom_type, swap=True, dims=2):
if geom_type == 'linestring':
if geom_type == "linestring":
return _parse_georss_line(value, swap, dims)
elif geom_type == 'polygon':
elif geom_type == "polygon":
ring = _parse_georss_line(value, swap, dims)
return {'type': 'Polygon', 'coordinates': (ring['coordinates'],)}
return {"type": "Polygon", "coordinates": (ring["coordinates"],)}
else:
return None
@ -180,10 +180,10 @@ def _parse_poslist(value, geom_type, swap=True, dims=2):
def _gen_georss_coords(value, swap=True, dims=2):
# A generator of (lon, lat) pairs from a string of encoded GeoRSS
# coordinates. Converts to floats and swaps order.
latlons = (float(ll) for ll in value.replace(',', ' ').split())
latlons = (float(ll) for ll in value.replace(",", " ").split())
while True:
try:
t = [next(latlons), next(latlons)][::swap and -1 or 1]
t = [next(latlons), next(latlons)][:: swap and -1 or 1]
if dims == 3:
t.append(next(latlons))
yield tuple(t)
@ -196,7 +196,7 @@ def _parse_georss_point(value, swap=True, dims=2):
# whitespace. We'll also handle comma separators.
try:
coords = list(_gen_georss_coords(value, swap, dims))
return {'type': 'Point', 'coordinates': coords[0]}
return {"type": "Point", "coordinates": coords[0]}
except (IndexError, ValueError):
return None
@ -207,7 +207,7 @@ def _parse_georss_line(value, swap=True, dims=2):
# whitespace. There must be at least two pairs.
try:
coords = list(_gen_georss_coords(value, swap, dims))
return {'type': 'LineString', 'coordinates': coords}
return {"type": "LineString", "coordinates": coords}
except (IndexError, ValueError):
return None
@ -223,7 +223,7 @@ def _parse_georss_polygon(value, swap=True, dims=2):
return None
if len(ring) < 4:
return None
return {'type': 'Polygon', 'coordinates': (ring,)}
return {"type": "Polygon", "coordinates": (ring,)}
def _parse_georss_box(value, swap=True, dims=2):
@ -233,7 +233,7 @@ def _parse_georss_box(value, swap=True, dims=2):
# first pair is the lower corner, the second is the upper corner.
try:
coords = list(_gen_georss_coords(value, swap, dims))
return {'type': 'Box', 'coordinates': tuple(coords)}
return {"type": "Box", "coordinates": tuple(coords)}
except (IndexError, ValueError):
return None
@ -241,38 +241,443 @@ def _parse_georss_box(value, swap=True, dims=2):
# The list of EPSG codes for geographic (latitude/longitude) coordinate
# systems to support decoding of GeoRSS GML profiles.
_geogCS = [
3819, 3821, 3824, 3889, 3906, 4001, 4002, 4003, 4004, 4005, 4006, 4007, 4008,
4009, 4010, 4011, 4012, 4013, 4014, 4015, 4016, 4018, 4019, 4020, 4021, 4022,
4023, 4024, 4025, 4027, 4028, 4029, 4030, 4031, 4032, 4033, 4034, 4035, 4036,
4041, 4042, 4043, 4044, 4045, 4046, 4047, 4052, 4053, 4054, 4055, 4075, 4081,
4120, 4121, 4122, 4123, 4124, 4125, 4126, 4127, 4128, 4129, 4130, 4131, 4132,
4133, 4134, 4135, 4136, 4137, 4138, 4139, 4140, 4141, 4142, 4143, 4144, 4145,
4146, 4147, 4148, 4149, 4150, 4151, 4152, 4153, 4154, 4155, 4156, 4157, 4158,
4159, 4160, 4161, 4162, 4163, 4164, 4165, 4166, 4167, 4168, 4169, 4170, 4171,
4172, 4173, 4174, 4175, 4176, 4178, 4179, 4180, 4181, 4182, 4183, 4184, 4185,
4188, 4189, 4190, 4191, 4192, 4193, 4194, 4195, 4196, 4197, 4198, 4199, 4200,
4201, 4202, 4203, 4204, 4205, 4206, 4207, 4208, 4209, 4210, 4211, 4212, 4213,
4214, 4215, 4216, 4218, 4219, 4220, 4221, 4222, 4223, 4224, 4225, 4226, 4227,
4228, 4229, 4230, 4231, 4232, 4233, 4234, 4235, 4236, 4237, 4238, 4239, 4240,
4241, 4242, 4243, 4244, 4245, 4246, 4247, 4248, 4249, 4250, 4251, 4252, 4253,
4254, 4255, 4256, 4257, 4258, 4259, 4260, 4261, 4262, 4263, 4264, 4265, 4266,
4267, 4268, 4269, 4270, 4271, 4272, 4273, 4274, 4275, 4276, 4277, 4278, 4279,
4280, 4281, 4282, 4283, 4284, 4285, 4286, 4287, 4288, 4289, 4291, 4292, 4293,
4294, 4295, 4296, 4297, 4298, 4299, 4300, 4301, 4302, 4303, 4304, 4306, 4307,
4308, 4309, 4310, 4311, 4312, 4313, 4314, 4315, 4316, 4317, 4318, 4319, 4322,
4324, 4326, 4463, 4470, 4475, 4483, 4490, 4555, 4558, 4600, 4601, 4602, 4603,
4604, 4605, 4606, 4607, 4608, 4609, 4610, 4611, 4612, 4613, 4614, 4615, 4616,
4617, 4618, 4619, 4620, 4621, 4622, 4623, 4624, 4625, 4626, 4627, 4628, 4629,
4630, 4631, 4632, 4633, 4634, 4635, 4636, 4637, 4638, 4639, 4640, 4641, 4642,
4643, 4644, 4645, 4646, 4657, 4658, 4659, 4660, 4661, 4662, 4663, 4664, 4665,
4666, 4667, 4668, 4669, 4670, 4671, 4672, 4673, 4674, 4675, 4676, 4677, 4678,
4679, 4680, 4681, 4682, 4683, 4684, 4685, 4686, 4687, 4688, 4689, 4690, 4691,
4692, 4693, 4694, 4695, 4696, 4697, 4698, 4699, 4700, 4701, 4702, 4703, 4704,
4705, 4706, 4707, 4708, 4709, 4710, 4711, 4712, 4713, 4714, 4715, 4716, 4717,
4718, 4719, 4720, 4721, 4722, 4723, 4724, 4725, 4726, 4727, 4728, 4729, 4730,
4731, 4732, 4733, 4734, 4735, 4736, 4737, 4738, 4739, 4740, 4741, 4742, 4743,
4744, 4745, 4746, 4747, 4748, 4749, 4750, 4751, 4752, 4753, 4754, 4755, 4756,
4757, 4758, 4759, 4760, 4761, 4762, 4763, 4764, 4765, 4801, 4802, 4803, 4804,
4805, 4806, 4807, 4808, 4809, 4810, 4811, 4813, 4814, 4815, 4816, 4817, 4818,
4819, 4820, 4821, 4823, 4824, 4901, 4902, 4903, 4904, 4979,
3819,
3821,
3824,
3889,
3906,
4001,
4002,
4003,
4004,
4005,
4006,
4007,
4008,
4009,
4010,
4011,
4012,
4013,
4014,
4015,
4016,
4018,
4019,
4020,
4021,
4022,
4023,
4024,
4025,
4027,
4028,
4029,
4030,
4031,
4032,
4033,
4034,
4035,
4036,
4041,
4042,
4043,
4044,
4045,
4046,
4047,
4052,
4053,
4054,
4055,
4075,
4081,
4120,
4121,
4122,
4123,
4124,
4125,
4126,
4127,
4128,
4129,
4130,
4131,
4132,
4133,
4134,
4135,
4136,
4137,
4138,
4139,
4140,
4141,
4142,
4143,
4144,
4145,
4146,
4147,
4148,
4149,
4150,
4151,
4152,
4153,
4154,
4155,
4156,
4157,
4158,
4159,
4160,
4161,
4162,
4163,
4164,
4165,
4166,
4167,
4168,
4169,
4170,
4171,
4172,
4173,
4174,
4175,
4176,
4178,
4179,
4180,
4181,
4182,
4183,
4184,
4185,
4188,
4189,
4190,
4191,
4192,
4193,
4194,
4195,
4196,
4197,
4198,
4199,
4200,
4201,
4202,
4203,
4204,
4205,
4206,
4207,
4208,
4209,
4210,
4211,
4212,
4213,
4214,
4215,
4216,
4218,
4219,
4220,
4221,
4222,
4223,
4224,
4225,
4226,
4227,
4228,
4229,
4230,
4231,
4232,
4233,
4234,
4235,
4236,
4237,
4238,
4239,
4240,
4241,
4242,
4243,
4244,
4245,
4246,
4247,
4248,
4249,
4250,
4251,
4252,
4253,
4254,
4255,
4256,
4257,
4258,
4259,
4260,
4261,
4262,
4263,
4264,
4265,
4266,
4267,
4268,
4269,
4270,
4271,
4272,
4273,
4274,
4275,
4276,
4277,
4278,
4279,
4280,
4281,
4282,
4283,
4284,
4285,
4286,
4287,
4288,
4289,
4291,
4292,
4293,
4294,
4295,
4296,
4297,
4298,
4299,
4300,
4301,
4302,
4303,
4304,
4306,
4307,
4308,
4309,
4310,
4311,
4312,
4313,
4314,
4315,
4316,
4317,
4318,
4319,
4322,
4324,
4326,
4463,
4470,
4475,
4483,
4490,
4555,
4558,
4600,
4601,
4602,
4603,
4604,
4605,
4606,
4607,
4608,
4609,
4610,
4611,
4612,
4613,
4614,
4615,
4616,
4617,
4618,
4619,
4620,
4621,
4622,
4623,
4624,
4625,
4626,
4627,
4628,
4629,
4630,
4631,
4632,
4633,
4634,
4635,
4636,
4637,
4638,
4639,
4640,
4641,
4642,
4643,
4644,
4645,
4646,
4657,
4658,
4659,
4660,
4661,
4662,
4663,
4664,
4665,
4666,
4667,
4668,
4669,
4670,
4671,
4672,
4673,
4674,
4675,
4676,
4677,
4678,
4679,
4680,
4681,
4682,
4683,
4684,
4685,
4686,
4687,
4688,
4689,
4690,
4691,
4692,
4693,
4694,
4695,
4696,
4697,
4698,
4699,
4700,
4701,
4702,
4703,
4704,
4705,
4706,
4707,
4708,
4709,
4710,
4711,
4712,
4713,
4714,
4715,
4716,
4717,
4718,
4719,
4720,
4721,
4722,
4723,
4724,
4725,
4726,
4727,
4728,
4729,
4730,
4731,
4732,
4733,
4734,
4735,
4736,
4737,
4738,
4739,
4740,
4741,
4742,
4743,
4744,
4745,
4746,
4747,
4748,
4749,
4750,
4751,
4752,
4753,
4754,
4755,
4756,
4757,
4758,
4759,
4760,
4761,
4762,
4763,
4764,
4765,
4801,
4802,
4803,
4804,
4805,
4806,
4807,
4808,
4809,
4810,
4811,
4813,
4814,
4815,
4816,
4817,
4818,
4819,
4820,
4821,
4823,
4824,
4901,
4902,
4903,
4904,
4979,
]

46
lib/feedparser/namespaces/itunes.py
View file

@ -1,5 +1,5 @@
# Support for the iTunes format
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -29,13 +29,12 @@
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
supported_namespaces = {
# Canonical namespace
'http://www.itunes.com/DTDs/PodCast-1.0.dtd': 'itunes',
"http://www.itunes.com/DTDs/PodCast-1.0.dtd": "itunes",
# Extra namespace
'http://example.com/DTDs/PodCast-1.0.dtd': 'itunes',
"http://example.com/DTDs/PodCast-1.0.dtd": "itunes",
}
def _start_itunes_author(self, attrs_d):
@ -73,37 +72,42 @@ class Namespace(object):
def _start_itunes_owner(self, attrs_d):
self.inpublisher = 1
self.push('publisher', 0)
self.push("publisher", 0)
def _end_itunes_owner(self):
self.pop('publisher')
self.pop("publisher")
self.inpublisher = 0
self._sync_author_detail('publisher')
self._sync_author_detail("publisher")
def _end_itunes_keywords(self):
for term in self.pop('itunes_keywords').split(','):
for term in self.pop("itunes_keywords").split(","):
if term.strip():
self._add_tag(term.strip(), 'http://www.itunes.com/', None)
self._add_tag(term.strip(), "http://www.itunes.com/", None)
def _start_itunes_category(self, attrs_d):
self._add_tag(attrs_d.get('text'), 'http://www.itunes.com/', None)
self.push('category', 1)
self._add_tag(attrs_d.get("text"), "http://www.itunes.com/", None)
self.push("category", 1)
def _start_itunes_image(self, attrs_d):
self.push('itunes_image', 0)
if attrs_d.get('href'):
self._get_context()['image'] = FeedParserDict({'href': attrs_d.get('href')})
elif attrs_d.get('url'):
self._get_context()['image'] = FeedParserDict({'href': attrs_d.get('url')})
self.push("itunes_image", 0)
if attrs_d.get("href"):
self._get_context()["image"] = FeedParserDict({"href": attrs_d.get("href")})
elif attrs_d.get("url"):
self._get_context()["image"] = FeedParserDict({"href": attrs_d.get("url")})
_start_itunes_link = _start_itunes_image
def _end_itunes_block(self):
value = self.pop('itunes_block', 0)
self._get_context()['itunes_block'] = (value == 'yes' or value == 'Yes') and 1 or 0
value = self.pop("itunes_block", 0)
self._get_context()["itunes_block"] = (
(value == "yes" or value == "Yes") and 1 or 0
)
def _end_itunes_explicit(self):
value = self.pop('itunes_explicit', 0)
value = self.pop("itunes_explicit", 0)
# Convert 'yes' -> True, 'clean' to False, and any other value to None
# False and None both evaluate as False, so the difference can be ignored
# by applications that only need to know if the content is explicit.
self._get_context()['itunes_explicit'] = (None, False, True)[(value == 'yes' and 2) or value == 'clean' or 0]
self._get_context()["itunes_explicit"] = (None, False, True)[
(value == "yes" and 2) or value == "clean" or 0
]

73
lib/feedparser/namespaces/mediarss.py
View file

@ -1,5 +1,5 @@
# Support for the Media RSS format
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -29,24 +29,23 @@
from ..util import FeedParserDict
class Namespace(object):
class Namespace:
supported_namespaces = {
# Canonical namespace
'http://search.yahoo.com/mrss/': 'media',
"http://search.yahoo.com/mrss/": "media",
# Old namespace (no trailing slash)
'http://search.yahoo.com/mrss': 'media',
"http://search.yahoo.com/mrss": "media",
}
def _start_media_category(self, attrs_d):
attrs_d.setdefault('scheme', 'http://search.yahoo.com/mrss/category_schema')
attrs_d.setdefault("scheme", "http://search.yahoo.com/mrss/category_schema")
self._start_category(attrs_d)
def _end_media_category(self):
self._end_category()
def _end_media_keywords(self):
for term in self.pop('media_keywords').split(','):
for term in self.pop("media_keywords").split(","):
if term.strip():
self._add_tag(term.strip(), None, None)
@ -64,26 +63,26 @@ class Namespace(object):
def _start_media_rating(self, attrs_d):
context = self._get_context()
context.setdefault('media_rating', attrs_d)
self.push('rating', 1)
context.setdefault("media_rating", attrs_d)
self.push("rating", 1)
def _end_media_rating(self):
rating = self.pop('rating')
rating = self.pop("rating")
if rating is not None and rating.strip():
context = self._get_context()
context['media_rating']['content'] = rating
context["media_rating"]["content"] = rating
def _start_media_credit(self, attrs_d):
context = self._get_context()
context.setdefault('media_credit', [])
context['media_credit'].append(attrs_d)
self.push('credit', 1)
context.setdefault("media_credit", [])
context["media_credit"].append(attrs_d)
self.push("credit", 1)
def _end_media_credit(self):
credit = self.pop('credit')
credit = self.pop("credit")
if credit is not None and credit.strip():
context = self._get_context()
context['media_credit'][-1]['content'] = credit
context["media_credit"][-1]["content"] = credit
def _start_media_description(self, attrs_d):
self._start_description(attrs_d)
@ -93,49 +92,51 @@ class Namespace(object):
def _start_media_restriction(self, attrs_d):
context = self._get_context()
context.setdefault('media_restriction', attrs_d)
self.push('restriction', 1)
context.setdefault("media_restriction", attrs_d)
self.push("restriction", 1)
def _end_media_restriction(self):
restriction = self.pop('restriction')
restriction = self.pop("restriction")
if restriction is not None and restriction.strip():
context = self._get_context()
context['media_restriction']['content'] = [cc.strip().lower() for cc in restriction.split(' ')]
context["media_restriction"]["content"] = [
cc.strip().lower() for cc in restriction.split(" ")
]
def _start_media_license(self, attrs_d):
context = self._get_context()
context.setdefault('media_license', attrs_d)
self.push('license', 1)
context.setdefault("media_license", attrs_d)
self.push("license", 1)
def _end_media_license(self):
license_ = self.pop('license')
license_ = self.pop("license")
if license_ is not None and license_.strip():
context = self._get_context()
context['media_license']['content'] = license_
context["media_license"]["content"] = license_
def _start_media_content(self, attrs_d):
context = self._get_context()
context.setdefault('media_content', [])
context['media_content'].append(attrs_d)
context.setdefault("media_content", [])
context["media_content"].append(attrs_d)
def _start_media_thumbnail(self, attrs_d):
context = self._get_context()
context.setdefault('media_thumbnail', [])
self.push('url', 1) # new
context['media_thumbnail'].append(attrs_d)
context.setdefault("media_thumbnail", [])
self.push("url", 1) # new
context["media_thumbnail"].append(attrs_d)
def _end_media_thumbnail(self):
url = self.pop('url')
url = self.pop("url")
context = self._get_context()
if url is not None and url.strip():
if 'url' not in context['media_thumbnail'][-1]:
context['media_thumbnail'][-1]['url'] = url
if "url" not in context["media_thumbnail"][-1]:
context["media_thumbnail"][-1]["url"] = url
def _start_media_player(self, attrs_d):
self.push('media_player', 0)
self._get_context()['media_player'] = FeedParserDict(attrs_d)
self.push("media_player", 0)
self._get_context()["media_player"] = FeedParserDict(attrs_d)
def _end_media_player(self):
value = self.pop('media_player')
value = self.pop("media_player")
context = self._get_context()
context['media_player']['content'] = value
context["media_player"]["content"] = value

26
lib/feedparser/namespaces/psc.py
View file

@ -1,5 +1,5 @@
# Support for the Podlove Simple Chapters format
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -32,36 +32,36 @@ import re
from .. import util
class Namespace(object):
class Namespace:
supported_namespaces = {
'http://podlove.org/simple-chapters': 'psc',
"http://podlove.org/simple-chapters": "psc",
}
def __init__(self):
# chapters will only be captured while psc_chapters_flag is True.
self.psc_chapters_flag = False
super(Namespace, self).__init__()
super().__init__()
def _start_psc_chapters(self, attrs_d):
context = self._get_context()
if 'psc_chapters' not in context:
if "psc_chapters" not in context:
self.psc_chapters_flag = True
attrs_d['chapters'] = []
context['psc_chapters'] = util.FeedParserDict(attrs_d)
attrs_d["chapters"] = []
context["psc_chapters"] = util.FeedParserDict(attrs_d)
def _end_psc_chapters(self):
self.psc_chapters_flag = False
def _start_psc_chapter(self, attrs_d):
if self.psc_chapters_flag:
start = self._get_attribute(attrs_d, 'start')
attrs_d['start_parsed'] = _parse_psc_chapter_start(start)
start = self._get_attribute(attrs_d, "start")
attrs_d["start_parsed"] = _parse_psc_chapter_start(start)
context = self._get_context()['psc_chapters']
context['chapters'].append(util.FeedParserDict(attrs_d))
context = self._get_context()["psc_chapters"]
context["chapters"].append(util.FeedParserDict(attrs_d))
format_ = re.compile(r'^((\d{2}):)?(\d{2}):(\d{2})(\.(\d{3}))?$')
format_ = re.compile(r"^((\d{2}):)?(\d{2}):(\d{2})(\.(\d{3}))?$")
def _parse_psc_chapter_start(start):
@ -71,4 +71,4 @@ def _parse_psc_chapter_start(start):
_, h, m, s, _, ms = m.groups()
h, m, s, ms = (int(h or 0), int(m), int(s), int(ms or 0))
return datetime.timedelta(0, h*60*60 + m*60 + s, ms*1000)
return datetime.timedelta(0, h * 60 * 60 + m * 60 + s, ms * 1000)

102
lib/feedparser/parsers/json.py
View file

@ -34,37 +34,37 @@ from ..util import FeedParserDict
class JSONParser:
VERSIONS = {
'https://jsonfeed.org/version/1': 'json1',
'https://jsonfeed.org/version/1.1': 'json11',
"https://jsonfeed.org/version/1": "json1",
"https://jsonfeed.org/version/1.1": "json11",
}
FEED_FIELDS = (
('title', 'title'),
('icon', 'image'),
('home_page_url', 'link'),
('description', 'description'),
("title", "title"),
("icon", "image"),
("home_page_url", "link"),
("description", "description"),
)
ITEM_FIELDS = (
('title', 'title'),
('id', 'guid'),
('url', 'link'),
('summary', 'summary'),
('external_url', 'source'),
("title", "title"),
("id", "guid"),
("url", "link"),
("summary", "summary"),
("external_url", "source"),
)
def __init__(self, baseuri=None, baselang=None, encoding=None):
self.baseuri = baseuri or ''
self.baseuri = baseuri or ""
self.lang = baselang or None
self.encoding = encoding or 'utf-8' # character encoding
self.encoding = encoding or "utf-8" # character encoding
self.version = None
self.feeddata = FeedParserDict()
self.namespacesInUse = []
self.entries = []
def feed(self, data):
data = json.loads(data)
def feed(self, file):
data = json.load(file)
v = data.get('version', '')
v = data.get("version", "")
try:
self.version = self.VERSIONS[v]
except KeyError:
@ -73,11 +73,11 @@ class JSONParser:
for src, dst in self.FEED_FIELDS:
if src in data:
self.feeddata[dst] = data[src]
if 'author' in data:
self.parse_author(data['author'], self.feeddata)
if "author" in data:
self.parse_author(data["author"], self.feeddata)
# TODO: hubs; expired has no RSS equivalent
self.entries = [self.parse_entry(e) for e in data['items']]
self.entries = [self.parse_entry(e) for e in data["items"]]
def parse_entry(self, e):
entry = FeedParserDict()
@ -85,49 +85,51 @@ class JSONParser:
if src in e:
entry[dst] = e[src]
if 'content_text' in e:
entry['content'] = c = FeedParserDict()
c['value'] = e['content_text']
c['type'] = 'text'
elif 'content_html' in e:
entry['content'] = c = FeedParserDict()
c['value'] = sanitize_html(e['content_html'], self.encoding, 'application/json')
c['type'] = 'html'
if "content_text" in e:
entry["content"] = c = FeedParserDict()
c["value"] = e["content_text"]
c["type"] = "text"
elif "content_html" in e:
entry["content"] = c = FeedParserDict()
c["value"] = sanitize_html(
e["content_html"], self.encoding, "application/json"
)
c["type"] = "html"
if 'date_published' in e:
entry['published'] = e['date_published']
entry['published_parsed'] = _parse_date(e['date_published'])
if 'date_updated' in e:
entry['updated'] = e['date_modified']
entry['updated_parsed'] = _parse_date(e['date_modified'])
if "date_published" in e:
entry["published"] = e["date_published"]
entry["published_parsed"] = _parse_date(e["date_published"])
if "date_updated" in e:
entry["updated"] = e["date_modified"]
entry["updated_parsed"] = _parse_date(e["date_modified"])
if 'tags' in e:
entry['category'] = e['tags']
if "tags" in e:
entry["category"] = e["tags"]
if 'author' in e:
self.parse_author(e['author'], entry)
if "author" in e:
self.parse_author(e["author"], entry)
if 'attachments' in e:
entry['enclosures'] = [self.parse_attachment(a) for a in e['attachments']]
if "attachments" in e:
entry["enclosures"] = [self.parse_attachment(a) for a in e["attachments"]]
return entry
@staticmethod
def parse_author(parent, dest):
dest['author_detail'] = detail = FeedParserDict()
if 'name' in parent:
dest['author'] = detail['name'] = parent['name']
if 'url' in parent:
if parent['url'].startswith('mailto:'):
detail['email'] = parent['url'][7:]
dest["author_detail"] = detail = FeedParserDict()
if "name" in parent:
dest["author"] = detail["name"] = parent["name"]
if "url" in parent:
if parent["url"].startswith("mailto:"):
detail["email"] = parent["url"][7:]
else:
detail['href'] = parent['url']
detail["href"] = parent["url"]
@staticmethod
def parse_attachment(attachment):
enc = FeedParserDict()
enc['href'] = attachment['url']
enc['type'] = attachment['mime_type']
if 'size_in_bytes' in attachment:
enc['length'] = attachment['size_in_bytes']
enc["href"] = attachment["url"]
enc["type"] = attachment["mime_type"]
if "size_in_bytes" in attachment:
enc["length"] = attachment["size_in_bytes"]
return enc

56
lib/feedparser/parsers/loose.py
View file

@ -1,5 +1,5 @@
# The loose feed parser that interfaces with an SGML parsing library
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -26,52 +26,50 @@
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
class LooseXMLParser:
contentparams = None
def __init__(self, baseuri=None, baselang=None, encoding=None, entities=None):
self.baseuri = baseuri or ''
self.baseuri = baseuri or ""
self.lang = baselang or None
self.encoding = encoding or 'utf-8' # character encoding
self.encoding = encoding or "utf-8" # character encoding
self.entities = entities or {}
super().__init__()
@staticmethod
def _normalize_attributes(kv):
k = kv[0].lower()
v = k in ('rel', 'type') and kv[1].lower() or kv[1]
v = k in ("rel", "type") and kv[1].lower() or kv[1]
# the sgml parser doesn't handle entities in attributes, nor
# does it pass the attribute values through as unicode, while
# strict xml parsers do -- account for this difference
v = v.replace('&amp;', '&')
v = v.replace("&amp;", "&")
return k, v
def decode_entities(self, element, data):
data = data.replace('&#60;', '&lt;')
data = data.replace('&#x3c;', '&lt;')
data = data.replace('&#x3C;', '&lt;')
data = data.replace('&#62;', '&gt;')
data = data.replace('&#x3e;', '&gt;')
data = data.replace('&#x3E;', '&gt;')
data = data.replace('&#38;', '&amp;')
data = data.replace('&#x26;', '&amp;')
data = data.replace('&#34;', '&quot;')
data = data.replace('&#x22;', '&quot;')
data = data.replace('&#39;', '&apos;')
data = data.replace('&#x27;', '&apos;')
if not self.contentparams.get('type', 'xml').endswith('xml'):
data = data.replace('&lt;', '<')
data = data.replace('&gt;', '>')
data = data.replace('&amp;', '&')
data = data.replace('&quot;', '"')
data = data.replace('&apos;', "'")
data = data.replace('&#x2f;', '/')
data = data.replace('&#x2F;', '/')
data = data.replace("&#60;", "&lt;")
data = data.replace("&#x3c;", "&lt;")
data = data.replace("&#x3C;", "&lt;")
data = data.replace("&#62;", "&gt;")
data = data.replace("&#x3e;", "&gt;")
data = data.replace("&#x3E;", "&gt;")
data = data.replace("&#38;", "&amp;")
data = data.replace("&#x26;", "&amp;")
data = data.replace("&#34;", "&quot;")
data = data.replace("&#x22;", "&quot;")
data = data.replace("&#39;", "&apos;")
data = data.replace("&#x27;", "&apos;")
if not self.contentparams.get("type", "xml").endswith("xml"):
data = data.replace("&lt;", "<")
data = data.replace("&gt;", ">")
data = data.replace("&amp;", "&")
data = data.replace("&quot;", '"')
data = data.replace("&apos;", "'")
data = data.replace("&#x2f;", "/")
data = data.replace("&#x2F;", "/")
return data
@staticmethod
def strattrs(attrs):
return ''.join(
' %s="%s"' % (n, v.replace('"', '&quot;'))
for n, v in attrs
)
return "".join(' {}="{}"'.format(n, v.replace('"', "&quot;")) for n, v in attrs)

62
lib/feedparser/parsers/strict.py
View file

@ -1,5 +1,5 @@
# The strict feed parser that interfaces with an XML parsing library
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -34,15 +34,15 @@ class StrictXMLParser:
self.bozo = 0
self.exc = None
self.decls = {}
self.baseuri = baseuri or ''
self.baseuri = baseuri or ""
self.lang = baselang
self.encoding = encoding
super(StrictXMLParser, self).__init__()
super().__init__()
@staticmethod
def _normalize_attributes(kv):
k = kv[0].lower()
v = k in ('rel', 'type') and kv[1].lower() or kv[1]
v = k in ("rel", "type") and kv[1].lower() or kv[1]
return k, v
def startPrefixMapping(self, prefix, uri):
@ -51,23 +51,29 @@ class StrictXMLParser:
# Jython uses '' instead of None; standardize on None
prefix = prefix or None
self.track_namespace(prefix, uri)
if prefix and uri == 'http://www.w3.org/1999/xlink':
self.decls['xmlns:' + prefix] = uri
if prefix and uri == "http://www.w3.org/1999/xlink":
self.decls["xmlns:" + prefix] = uri
def startElementNS(self, name, qname, attrs):
namespace, localname = name
lowernamespace = str(namespace or '').lower()
if lowernamespace.find('backend.userland.com/rss') != -1:
lowernamespace = str(namespace or "").lower()
if lowernamespace.find("backend.userland.com/rss") != -1:
# match any backend.userland.com namespace
namespace = 'http://backend.userland.com/rss'
namespace = "http://backend.userland.com/rss"
lowernamespace = namespace
if qname and qname.find(':') > 0:
givenprefix = qname.split(':')[0]
if qname and qname.find(":") > 0:
givenprefix = qname.split(":")[0]
else:
givenprefix = None
prefix = self._matchnamespaces.get(lowernamespace, givenprefix)
if givenprefix and (prefix is None or (prefix == '' and lowernamespace == '')) and givenprefix not in self.namespaces_in_use:
raise UndeclaredNamespace("'%s' is not associated with a namespace" % givenprefix)
if (
givenprefix
and (prefix is None or (prefix == "" and lowernamespace == ""))
and givenprefix not in self.namespaces_in_use
):
raise UndeclaredNamespace(
"'%s' is not associated with a namespace" % givenprefix
)
localname = str(localname).lower()
# qname implementation is horribly broken in Python 2.1 (it
@ -78,24 +84,24 @@ class StrictXMLParser:
# at all). Thanks to MatejC for helping me test this and
# tirelessly telling me that it didn't work yet.
attrsD, self.decls = self.decls, {}
if localname == 'math' and namespace == 'http://www.w3.org/1998/Math/MathML':
attrsD['xmlns'] = namespace
if localname == 'svg' and namespace == 'http://www.w3.org/2000/svg':
attrsD['xmlns'] = namespace
if localname == "math" and namespace == "http://www.w3.org/1998/Math/MathML":
attrsD["xmlns"] = namespace
if localname == "svg" and namespace == "http://www.w3.org/2000/svg":
attrsD["xmlns"] = namespace
if prefix:
localname = prefix.lower() + ':' + localname
localname = prefix.lower() + ":" + localname
elif namespace and not qname: # Expat
for name, value in self.namespaces_in_use.items():
if name and value == namespace:
localname = name + ':' + localname
localname = name + ":" + localname
break
for (namespace, attrlocalname), attrvalue in attrs.items():
lowernamespace = (namespace or '').lower()
prefix = self._matchnamespaces.get(lowernamespace, '')
lowernamespace = (namespace or "").lower()
prefix = self._matchnamespaces.get(lowernamespace, "")
if prefix:
attrlocalname = prefix + ':' + attrlocalname
attrlocalname = prefix + ":" + attrlocalname
attrsD[str(attrlocalname).lower()] = attrvalue
for qname in attrs.getQNames():
attrsD[str(qname).lower()] = attrs.getValueByQName(qname)
@ -107,18 +113,18 @@ class StrictXMLParser:
def endElementNS(self, name, qname):
namespace, localname = name
lowernamespace = str(namespace or '').lower()
if qname and qname.find(':') > 0:
givenprefix = qname.split(':')[0]
lowernamespace = str(namespace or "").lower()
if qname and qname.find(":") > 0:
givenprefix = qname.split(":")[0]
else:
givenprefix = ''
givenprefix = ""
prefix = self._matchnamespaces.get(lowernamespace, givenprefix)
if prefix:
localname = prefix + ':' + localname
localname = prefix + ":" + localname
elif namespace and not qname: # Expat
for name, value in self.namespaces_in_use.items():
if name and value == namespace:
localname = name + ':' + localname
localname = name + ":" + localname
break
localname = str(localname).lower()
self.unknown_endtag(localname)

0
lib/feedparser/py.typed Normal file
View file

1462
lib/feedparser/sanitizer.py
View file

File diff suppressed because it is too large Load diff

46
lib/feedparser/sgml.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -27,20 +27,20 @@
import re
import sgmllib # type: ignore[import]
import sgmllib3k as sgmllib
__all__ = [
'sgmllib',
'charref',
'tagfind',
'attrfind',
'entityref',
'incomplete',
'interesting',
'shorttag',
'shorttagopen',
'starttagopen',
'endbracket',
"sgmllib",
"charref",
"tagfind",
"attrfind",
"entityref",
"incomplete",
"interesting",
"shorttag",
"shorttagopen",
"starttagopen",
"endbracket",
]
# sgmllib defines a number of module-level regular expressions that are
@ -49,20 +49,20 @@ __all__ = [
# names, and the compiled code objects of several sgmllib.SGMLParser
# methods are copied into _BaseHTMLProcessor so that they execute in
# feedparser's scope instead of sgmllib's scope.
charref = re.compile(r'&#(\d+|[xX][0-9a-fA-F]+);')
tagfind = re.compile(r'[a-zA-Z][-_.:a-zA-Z0-9]*')
charref = re.compile(r"&#(\d+|[xX][0-9a-fA-F]+);")
tagfind = re.compile(r"[a-zA-Z][-_.:a-zA-Z0-9]*")
attrfind = re.compile(
r"""\s*([a-zA-Z_][-:.a-zA-Z_0-9]*)[$]?(\s*=\s*"""
r"""('[^']*'|"[^"]*"|[][\-a-zA-Z0-9./,:;+*%?!&$()_#=~'"@]*))?"""
)
# Unfortunately, these must be copied over to prevent NameError exceptions
entityref = sgmllib.entityref
incomplete = sgmllib.incomplete
interesting = sgmllib.interesting
shorttag = sgmllib.shorttag
shorttagopen = sgmllib.shorttagopen
starttagopen = sgmllib.starttagopen
entityref = sgmllib.SGMLParser.entityref
incomplete = sgmllib.SGMLParser.incomplete
interesting = sgmllib.SGMLParser.interesting
shorttag = sgmllib.SGMLParser.shorttag
shorttagopen = sgmllib.SGMLParser.shorttagopen
starttagopen = sgmllib.SGMLParser.starttagopen
class _EndBracketRegEx:
@ -70,12 +70,12 @@ class _EndBracketRegEx:
# Overriding the built-in sgmllib.endbracket regex allows the
# parser to find angle brackets embedded in element attributes.
self.endbracket = re.compile(
r'('
r"("
r"""[^'"<>]"""
r"""|"[^"]*"(?=>|/|\s|\w+=)"""
r"""|'[^']*'(?=>|/|\s|\w+=))*(?=[<>])"""
r"""|.*?(?=[<>]"""
r')'
r")"
)
def search(self, target, index=0):

163
lib/feedparser/urls.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -37,103 +37,116 @@ from .html import BaseHTMLProcessor
# https://secure.wikimedia.org/wikipedia/en/wiki/URI_scheme
# Many more will likely need to be added!
ACCEPTABLE_URI_SCHEMES = (
'file', 'ftp', 'gopher', 'h323', 'hdl', 'http', 'https', 'imap', 'magnet',
'mailto', 'mms', 'news', 'nntp', 'prospero', 'rsync', 'rtsp', 'rtspu',
'sftp', 'shttp', 'sip', 'sips', 'snews', 'svn', 'svn+ssh', 'telnet',
'wais',
"file",
"ftp",
"gopher",
"h323",
"hdl",
"http",
"https",
"imap",
"magnet",
"mailto",
"mms",
"news",
"nntp",
"prospero",
"rsync",
"rtsp",
"rtspu",
"sftp",
"shttp",
"sip",
"sips",
"snews",
"svn",
"svn+ssh",
"telnet",
"wais",
# Additional common-but-unofficial schemes
'aim', 'callto', 'cvs', 'facetime', 'feed', 'git', 'gtalk', 'irc', 'ircs',
'irc6', 'itms', 'mms', 'msnim', 'skype', 'ssh', 'smb', 'svn', 'ymsg',
"aim",
"callto",
"cvs",
"facetime",
"feed",
"git",
"gtalk",
"irc",
"ircs",
"irc6",
"itms",
"mms",
"msnim",
"skype",
"ssh",
"smb",
"svn",
"ymsg",
)
_urifixer = re.compile('^([A-Za-z][A-Za-z0-9+-.]*://)(/*)(.*?)')
_urifixer = re.compile("^([A-Za-z][A-Za-z0-9+-.]*://)(/*)(.*?)")
def _urljoin(base, uri):
uri = _urifixer.sub(r'\1\3', uri)
uri = _urifixer.sub(r"\1\3", uri)
try:
uri = urllib.parse.urljoin(base, uri)
except ValueError:
uri = ''
uri = ""
return uri
def convert_to_idn(url):
"""Convert a URL to IDN notation"""
# this function should only be called with a unicode string
# strategy: if the host cannot be encoded in ascii, then
# it'll be necessary to encode it in idn form
parts = list(urllib.parse.urlsplit(url))
try:
parts[1].encode('ascii')
except UnicodeEncodeError:
# the url needs to be converted to idn notation
host = parts[1].rsplit(':', 1)
newhost = []
port = ''
if len(host) == 2:
port = host.pop()
for h in host[0].split('.'):
newhost.append(h.encode('idna').decode('utf-8'))
parts[1] = '.'.join(newhost)
if port:
parts[1] += ':' + port
return urllib.parse.urlunsplit(parts)
else:
return url
def make_safe_absolute_uri(base, rel=None):
# bail if ACCEPTABLE_URI_SCHEMES is empty
if not ACCEPTABLE_URI_SCHEMES:
return _urljoin(base, rel or '')
return _urljoin(base, rel or "")
if not base:
return rel or ''
return rel or ""
if not rel:
try:
scheme = urllib.parse.urlparse(base)[0]
except ValueError:
return ''
return ""
if not scheme or scheme in ACCEPTABLE_URI_SCHEMES:
return base
return ''
return ""
uri = _urljoin(base, rel)
if uri.strip().split(':', 1)[0] not in ACCEPTABLE_URI_SCHEMES:
return ''
if uri.strip().split(":", 1)[0] not in ACCEPTABLE_URI_SCHEMES:
return ""
return uri
class RelativeURIResolver(BaseHTMLProcessor):
relative_uris = {
('a', 'href'),
('applet', 'codebase'),
('area', 'href'),
('audio', 'src'),
('blockquote', 'cite'),
('body', 'background'),
('del', 'cite'),
('form', 'action'),
('frame', 'longdesc'),
('frame', 'src'),
('iframe', 'longdesc'),
('iframe', 'src'),
('head', 'profile'),
('img', 'longdesc'),
('img', 'src'),
('img', 'usemap'),
('input', 'src'),
('input', 'usemap'),
('ins', 'cite'),
('link', 'href'),
('object', 'classid'),
('object', 'codebase'),
('object', 'data'),
('object', 'usemap'),
('q', 'cite'),
('script', 'src'),
('source', 'src'),
('video', 'poster'),
('video', 'src'),
("a", "href"),
("applet", "codebase"),
("area", "href"),
("audio", "src"),
("blockquote", "cite"),
("body", "background"),
("del", "cite"),
("form", "action"),
("frame", "longdesc"),
("frame", "src"),
("iframe", "longdesc"),
("iframe", "src"),
("head", "profile"),
("img", "longdesc"),
("img", "src"),
("img", "usemap"),
("input", "src"),
("input", "usemap"),
("ins", "cite"),
("link", "href"),
("object", "classid"),
("object", "codebase"),
("object", "data"),
("object", "usemap"),
("q", "cite"),
("script", "src"),
("source", "src"),
("video", "poster"),
("video", "src"),
}
def __init__(self, baseuri, encoding, _type):
@ -145,8 +158,14 @@ class RelativeURIResolver(BaseHTMLProcessor):
def unknown_starttag(self, tag, attrs):
attrs = self.normalize_attrs(attrs)
attrs = [(key, ((tag, key) in self.relative_uris) and self.resolve_uri(value) or value) for key, value in attrs]
super(RelativeURIResolver, self).unknown_starttag(tag, attrs)
attrs = [
(
key,
((tag, key) in self.relative_uris) and self.resolve_uri(value) or value,
)
for key, value in attrs
]
super().unknown_starttag(tag, attrs)
def resolve_relative_uris(html_source, base_uri, encoding, type_):

80
lib/feedparser/util.py
View file

@ -1,4 +1,4 @@
# Copyright 2010-2022 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2010-2023 Kurt McKee <contactme@kurtmckee.org>
# Copyright 2002-2008 Mark Pilgrim
# All rights reserved.
#
@ -30,22 +30,22 @@ import warnings
class FeedParserDict(dict):
keymap = {
'channel': 'feed',
'items': 'entries',
'guid': 'id',
'date': 'updated',
'date_parsed': 'updated_parsed',
'description': ['summary', 'subtitle'],
'description_detail': ['summary_detail', 'subtitle_detail'],
'url': ['href'],
'modified': 'updated',
'modified_parsed': 'updated_parsed',
'issued': 'published',
'issued_parsed': 'published_parsed',
'copyright': 'rights',
'copyright_detail': 'rights_detail',
'tagline': 'subtitle',
'tagline_detail': 'subtitle_detail',
"channel": "feed",
"items": "entries",
"guid": "id",
"date": "updated",
"date_parsed": "updated_parsed",
"description": ["summary", "subtitle"],
"description_detail": ["summary_detail", "subtitle_detail"],
"url": ["href"],
"modified": "updated",
"modified_parsed": "updated_parsed",
"issued": "published",
"issued_parsed": "published_parsed",
"copyright": "rights",
"copyright_detail": "rights_detail",
"tagline": "subtitle",
"tagline_detail": "subtitle_detail",
}
def __getitem__(self, key, _stacklevel=2):
@ -53,28 +53,29 @@ class FeedParserDict(dict):
:return: A :class:`FeedParserDict`.
"""
if key == 'category':
if key == "category":
try:
return dict.__getitem__(self, 'tags')[0]['term']
return dict.__getitem__(self, "tags")[0]["term"]
except IndexError:
raise KeyError("object doesn't have key 'category'")
elif key == 'enclosures':
elif key == "enclosures":
return [
FeedParserDict([(name, value) for (name, value) in link.items() if name != 'rel'])
for link in dict.__getitem__(self, 'links')
if link['rel'] == 'enclosure'
FeedParserDict(
[(name, value) for (name, value) in link.items() if name != "rel"]
)
for link in dict.__getitem__(self, "links")
if link["rel"] == "enclosure"
]
elif key == 'license':
for link in dict.__getitem__(self, 'links'):
if link['rel'] == 'license' and 'href' in link:
return link['href']
elif key == 'updated':
elif key == "license":
for link in dict.__getitem__(self, "links"):
if link["rel"] == "license" and "href" in link:
return link["href"]
elif key == "updated":
# Temporarily help developers out by keeping the old
# broken behavior that was reported in issue 310.
# This fix was proposed in issue 328.
if (
not dict.__contains__(self, 'updated')
and dict.__contains__(self, 'published')
if not dict.__contains__(self, "updated") and dict.__contains__(
self, "published"
):
warnings.warn(
"To avoid breaking existing software while "
@ -85,12 +86,11 @@ class FeedParserDict(dict):
DeprecationWarning,
stacklevel=_stacklevel,
)
return dict.__getitem__(self, 'published')
return dict.__getitem__(self, 'updated')
elif key == 'updated_parsed':
if (
not dict.__contains__(self, 'updated_parsed')
and dict.__contains__(self, 'published_parsed')
return dict.__getitem__(self, "published")
return dict.__getitem__(self, "updated")
elif key == "updated_parsed":
if not dict.__contains__(self, "updated_parsed") and dict.__contains__(
self, "published_parsed"
):
warnings.warn(
"To avoid breaking existing software while "
@ -101,8 +101,8 @@ class FeedParserDict(dict):
DeprecationWarning,
stacklevel=_stacklevel,
)
return dict.__getitem__(self, 'published_parsed')
return dict.__getitem__(self, 'updated_parsed')
return dict.__getitem__(self, "published_parsed")
return dict.__getitem__(self, "updated_parsed")
else:
realkey = self.keymap.get(key, key)
if isinstance(realkey, list):
@ -114,7 +114,7 @@ class FeedParserDict(dict):
return dict.__getitem__(self, key)
def __contains__(self, key):
if key in ('updated', 'updated_parsed'):
if key in ("updated", "updated_parsed"):
# Temporarily help developers out by keeping the old
# broken behavior that was reported in issue 310.
# This fix was proposed in issue 328.