[md_inline_extension] Broken with Markdown >= 3.4 #1373
Description
Markdown 3.2 deprecated markdown.util.etree, which md_inline_extension currently uses. Markdown 3.4 removed the deprecated etree symbol, which causes md_inline_extension plugin to fail to load with this error message:
Markdown is not installed - inline Markdown extension disabled
Updating the import is not enough to get md_inline_extension working again, after that it fails with:
PelicanInlineMarkdownExtension.extendMarkdown() missing 1 required positional argument: 'md_globals'
Removing md_globals parameter from PelicanInlineMarkdownExtension.extendMarkdown takes us to yet another error:
'Registry' object has no attribute 'add'
Which is caused by this line in pelican_inline_markdown_extension.py. Updating it from .add to .register takes us to the following error (taken with DEBUG=1 for useful stacktraces):
'<' not supported between instances of 'str' and 'int'
Traceback (most recent call last):
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/pelican/generators.py", line 740, in generate_context
page = self.readers.read_file(
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/pelican/readers.py", line 580, in read_file
content, reader_metadata = reader.read(path)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/pelican/readers.py", line 342, in read
self._md = Markdown(**self.settings['MARKDOWN'])
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/core.py", line 96, in __init__
self.registerExtensions(extensions=kwargs.get('extensions', []),
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/core.py", line 125, in registerExtensions
ext.extendMarkdown(self)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/extensions/footnotes.py", line 82, in extendMarkdown
md.inlinePatterns.register(FootnoteInlineProcessor(FOOTNOTE_RE, self), 'footnote', 175)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/util.py", line 331, in register
self.deregister(name)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/util.py", line 343, in deregister
index = self.get_index_for_name(name)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/util.py", line 307, in get_index_for_name
self._sort()
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/util.py", line 357, in _sort
self._priority.sort(key=lambda item: item.priority, reverse=True)
TypeError: '<' not supported between instances of 'str' and 'int'
Which boils down to register() taking an integer as priority and not a string as Registry.add (now deprecated) did.
Hopeful that this was the last error I try again, but alas another error:
'str' object has no attribute 'ANCESTOR_EXCLUDES'
Traceback (most recent call last):
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/pelican/generators.py", line 740, in generate_context
page = self.readers.read_file(
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/pelican/readers.py", line 580, in read_file
content, reader_metadata = reader.read(path)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/pelican/readers.py", line 344, in read
content = self._md.convert(text)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/core.py", line 268, in convert
newRoot = treeprocessor.run(root)
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/treeprocessors.py", line 367, in run
self.__handleInline(text), child
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/treeprocessors.py", line 127, in __handleInline
data, matched, startIndex = self.__applyPattern(
File "/home/fidel/Code/blog.fidelramos.net/virtualenv/lib/python3.10/site-packages/markdown/treeprocessors.py", line 258, in __applyPattern
for exclude in pattern.ANCESTOR_EXCLUDES:
AttributeError: 'str' object has no attribute 'ANCESTOR_EXCLUDES'
At this point I gave up, I will leave a PR in case someone wants to continue, but clearly the Markdown project doesn't care much about backwards-compatibility. :(