Merge v6.0.13 back to main

CHANGELOG.rst

@@ -20,6 +20,24 @@ Please see the fragment files in the `changelog.d directory`_.
 
 ..  scriv-insert-here
 
+6.0.13 - 2026-07-28
+===================
+
+Python support
+--------------
+
+*   Drop support for Python 3.9 and lower.
+
+Changed
+-------
+
+*   Migrate from ``sgmllib3k`` to ``feedparser-sgmllib``.
+
+Fixed
+-----
+
+*   Resolve package build warnings.
+
 6.0.12 - 2025-09-10
 ===================
 

changelog.d/20230219_143414_kurtmckee_update_python_support.rst

@@ -2,4 +2,3 @@ Python support
 --------------
 
 *   Support Python 3.11.
-*   Drop support for Python 3.6.

changelog.d/20240806_090500_kurtmckee_updates.rst

@@ -1,8 +1,3 @@
-Python support
---------------
-
-*   Drop Python 3.8 support.
-
 Project development
 -------------------
 

changelog.d/20250910_091527_kurtmckee_drop_py3_9_support.rst

@@ -1,8 +1,7 @@
 Python support
 --------------
 
-*   Drop Python 3.9 support.
-*   Begin testing Python 3.14 beta releases.
+*   Support Python 3.14.
 
 
 Project development

docs/introduction.rst

@@ -12,10 +12,8 @@ Notation)` feeds.  It also parses several
 popular extension modules, including Dublin Core and Apple's :program:`iTunes`
 extensions.
 
-To use :program:`Universal Feed Parser`, you will need :program:`Python` 3.8 or
-later. :program:`Universal Feed Parser` is not meant
-to run standalone; it is a module for you to use as part of a larger
-:program:`Python` program.
+:program:`Universal Feed Parser` is not meant to run standalone;
+it is a module for you to use as part of a larger :program:`Python` program.
 
 :program:`Universal Feed Parser` is easy to use; it has one primary public
 function, ``parse``.  ``parse`` takes a number of arguments, but only one is

feedparser/__init__.py

@@ -38,7 +38,7 @@
 
 __author__ = "Kurt McKee <contactme@kurtmckee.org>"
 __license__ = "BSD 2-clause"
-__version__ = "6.0.12"
+__version__ = "6.0.13"
 
 # If you want feedparser to automatically resolve all relative URIs, set this
 # to 1.

feedparser/html.py

@@ -28,20 +28,7 @@
 import html.entities
 import re
 
-# 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,
-)
+from .sgml import sgmllib
 
 _cp1252 = {
     128: "\u20ac",  # euro sign
@@ -121,25 +108,8 @@ def _shorttag_replace(self, match):
             return "<" + tag + " />"
         return "<" + tag + "></" + tag + ">"
 
-    # By declaring these methods and overriding their compiled code
-    # with the code from sgmllib, the original code will execute in
-    # feedparser's scope instead of sgmllib's. This means that the
-    # `tagfind` and `charref` regular expressions will be found as
-    # they're declared above, not as they're declared in sgmllib.
-    def goahead(self, i):
-        raise NotImplementedError
-
-    # Replace goahead with SGMLParser's goahead() code object.
-    goahead.__code__ = sgmllib.SGMLParser.goahead.__code__
-
-    def __parse_starttag(self, i):
-        raise NotImplementedError
-
-    # Replace __parse_starttag with SGMLParser's parse_starttag() code object.
-    __parse_starttag.__code__ = sgmllib.SGMLParser.parse_starttag.__code__
-
     def parse_starttag(self, i):
-        j = self.__parse_starttag(i)
+        j = super().parse_starttag(i)
         if self._type == "application/xhtml+xml":
             if j > 2 and self.rawdata[j - 2 : j] == "/>":
                 self.unknown_endtag(self.lasttag)

feedparser/sgml.py

@@ -27,43 +27,12 @@
 
 import re
 
-import sgmllib  # type: ignore[import]
+import feedparser.sgmllib as sgmllib
 
 __all__ = [
     "sgmllib",
-    "charref",
-    "tagfind",
-    "attrfind",
-    "entityref",
-    "incomplete",
-    "interesting",
-    "shorttag",
-    "shorttagopen",
-    "starttagopen",
-    "endbracket",
 ]
 
-# sgmllib defines a number of module-level regular expressions that are
-# insufficient for the XML parsing feedparser needs. Rather than modify
-# the variables directly in sgmllib, they're defined here using the same
-# 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]*")
-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
-
 
 class _EndBracketRegEx:
     def __init__(self):
@@ -95,4 +64,4 @@ def start(self, n):
         return self.match.end(n)
 
 
-endbracket = _EndBracketRegEx()
+sgmllib.endbracket = _EndBracketRegEx()  # type: ignore[assignment]

pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "feedparser"
-version = "6.0.11"
+version = "6.0.13"
 description = "Parse Atom/RSS/JSON feeds in Python"
 readme = "README.rst"
 requires-python = ">=3.10"
@@ -15,7 +15,7 @@ classifiers = [
     "Topic :: Text Processing :: Markup",
 ]
 dependencies = [
-    "sgmllib3k==1.0.0",
+    "feedparser-sgmllib (>=1, <2)",
     "requests>=2.20.0",
 ]
 

tests/conftest.py

@@ -4,6 +4,32 @@
 import responses
 
 
+# TODO: tag=bad-layout
+#
+#       The `sys.path` must be modified to guarantee that
+#       the repository's root directory IS NOT present
+#       at the start of the import path list.
+#
+#       If the path appears at the beginning of `sys.path`,
+#       it will breaks imports because of the namespace package
+#       feedparser-sgmllib.
+#
+#       After migrating to a `src/` layout, remove this code block.
+#
+# ---- 8< ----
+def rewrite_sys_path():
+    import pathlib
+    import sys
+
+    root = str(pathlib.Path(__file__).parent.parent.absolute())
+    if sys.path[0] == root:
+        sys.path.pop(0)
+
+
+rewrite_sys_path()
+# ---- >8 ----
+
+
 @pytest.fixture
 def use_loose_parser(monkeypatch):
     import feedparser.api

tox.ini

@@ -46,7 +46,16 @@ deps =
     -r requirements/test/requirements.txt
     chardet: chardet
 commands =
-    coverage run -m pytest {posargs:}
+    # TODO: tag=bad-layout
+    #
+    #       `--import-mode=append` is required to prevent pytest
+    #       from inserting the repository's root directory
+    #       at the beginning of `sys.path`, which breaks imports
+    #       because of the namespace package feedparser-sgmllib.
+    #
+    #       After migrating to a `src/` layout, remove the option.
+    #
+    coverage run -m pytest --import-mode=append {posargs:}
 
 
 [testenv:coverage_base]