misc updates

.github/workflows/ci.yml

@@ -15,14 +15,16 @@ jobs:
       fail-fast: false
       matrix:
         include:
-            - os: ubuntu-latest
-              python-version: '3.9'
             - os: ubuntu-latest
               python-version: '3.10'
             - os: ubuntu-latest
               python-version: '3.11'
             - os: ubuntu-latest
               python-version: '3.12'
+            - os: ubuntu-latest
+              python-version: '3.13'
+            - os: ubuntu-latest
+              python-version: '3.14'
 
     runs-on: ${{ matrix.os }}
     timeout-minutes: 10

CHANGES.rst

@@ -1,25 +1,35 @@
 Change Log
 ==========
 
+Version 0.12.0 (not released yet)
+---------------------------------
+
+Other changes:
+
+- support Python 3.13 and 3.14, drop support for Python 3.9
+- switch to SPDX license identifier
+- use native ``pyproject.toml`` support in tox
+- add scripts for sdist, signing, and PyPI upload
+
 Version 0.11.0 (2024-04-05)
 ---------------------------
 
 Fixes:
 
-- html parser: fix extraction of encoding from meta element
+- HTML parser: fix extraction of encoding from meta element
 - use raw string for regex, fixes "DeprecationWarning: invalid escape
   sequence \s" and others.
 
 New features:
 
-- include version infos into emeraldtree.__init__
+- include version info into emeraldtree.__init__
 
 Other changes:
 
 - packaging modernized / enhanced: pyproject.toml, MANIFEST.in, setuptools-scm
 - requires Python >= 3.9 now (dropped Python 2.x, removed "six")
-- add github actions CI, remove travis CI config
-- rst markup fixes / clean ups
+- add GitHub Actions CI, remove Travis CI config
+- rst markup fixes / cleanups
 - use tox for testing
 
 Version 0.10.0 (2015-06-10)
@@ -28,7 +38,7 @@ Version 0.10.0 (2015-06-10)
 Fixes:
 
 - fix setup.py - platform and no download_url
-- invalid output from HTML converter parsing preformatted code, multiline
+- fix invalid output from HTML converter parsing preformatted code, multiline
   paragraphs: part 2 of 2 fixes moin2 #516
 
 Other changes:

pyproject.toml

@@ -5,23 +5,23 @@ readme = "README.rst"
 authors = [{name="Bastian Blank", email="bblank@thinkmo.de"}, ]
 maintainers = [{name="Thomas Waldmann", email="tw@waldmann-edv.de"}, ]
 description = "EmeraldTree - a light-weight XML object model for Python."
-requires-python = ">=3.9"
+requires-python = ">=3.10"
 keywords = ["xml", "html", "html5", "polyglot", "element", "tree", "dom", "unicode"]
 classifiers = [
     "Development Status :: 4 - Beta",
     "Operating System :: OS Independent",
     "Programming Language :: Python",
     "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
-    "License :: OSI Approved :: Python Software Foundation License",
+    "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
     "Intended Audience :: Developers",
     "Topic :: Text Processing :: Markup :: HTML",
     "Topic :: Text Processing :: Markup :: XML",
 ]
-license = {text="Python (MIT style)"}
+license = "HPND-sell-variant"  # aka "Python (MIT Style)" or "Elementtree" license
 
 [project.urls]
 "Homepage" = "https://github.com/moinwiki/emeraldtree"
@@ -30,18 +30,18 @@ license = {text="Python (MIT style)"}
 where = ["src"]
 
 [build-system]
-requires = ["setuptools", "wheel", "setuptools_scm[toml]"]
+requires = ["setuptools>=78.1.1", "wheel", "setuptools_scm[toml]"]
 build-backend = "setuptools.build_meta"
 
 [tool.setuptools_scm]
 write_to = "src/emeraldtree/_version.py"
 
 [tool.tox]
-legacy_tox_ini = """
-[tox]
-envlist = py{39,310,311,312}
+requires = ["tox>=4"]
+env_list = ["py310", "py311", "py312", "py313", "py314"]
 
-[testenv]
-deps = pytest
-commands = pytest -rs --pyargs {posargs:emeraldtree}
-"""
+[tool.tox.env.py]
+deps = ["pytest"]
+commands = [
+    ["pytest", "-rs", "--pyargs", "{posargs:emeraldtree}"]
+]

src/emeraldtree/html.py

@@ -93,9 +93,10 @@ def handle_starttag(self, tag, attrs):
                 elif k == "content":
                     content = v
             if http_equiv == "content-type" and content:
-                import cgi
-                _, params = cgi.parse_header(content)
-                encoding = params.get('charset')
+                from email.message import Message
+                msg = Message()
+                msg['content-type'] = content
+                encoding = msg.get_param('charset', header='content-type')
                 if encoding:
                     self.encoding = encoding
         if tag.name in self.AUTOCLOSE: