Refactor abbr escaping

A alternate fix to #1444. This does not exclude the use of carrots or square brackets in abbreviations. It still excludes backslashse, however. I played with backslashes and it just doesn't make sense to support them as they have special meaning in the Markdown, not because of their use in regular expressions.

Author: waylan

docs/changelog.md

@@ -34,7 +34,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 * Include `scripts/*.py` in the generated source tarballs (#1430).
 * Ensure lines after heading in loose list are properly detabbed (#1443).
 * Give smarty tree processor higher priority than toc (#1440).
-* Explicitly omit carrot (`^`) and backslash (`\`) from abbreviations (#1444).
+* Permit carrots (`^`) and square brackets (`]`) but explicitly exclude
+ backslashes (`\`) from abbreviations (#1444).
 
 ## [3.5.2] -- 2024-01-10
 

docs/extensions/abbreviations.md

@@ -36,13 +36,9 @@ will be rendered as:
 is maintained by the <abbr title="World Wide Web Consortium">W3C</abbr>.</p>
 ```
 
-The following three characters are not permitted in an abbreviation. Any
-abbreviation definitions which include one will not be recognized as an
-abbreviation definition.
-
-1. carrot (`^`)
-2. backslash (`\`)
-3. left square bracket (`]`)
+The backslash (`\`) is not permitted in an abbreviation. Any abbreviation
+definitions which include one or more backslashes between the square brackets
+will not be recognized as an abbreviation definition.
 
 Usage
 -----

markdown/extensions/abbr.py

@@ -41,7 +41,7 @@ def extendMarkdown(self, md):
 class AbbrPreprocessor(BlockProcessor):
  """ Abbreviation Preprocessor - parse text for abbr references. """
 
- RE = re.compile(r'^[*]\[(?P<abbr>[^\]\^\\]*)\][ ]?:[ ]*\n?[ ]*(?P<title>.*)$', re.MULTILINE)
+ RE = re.compile(r'^[*]\[(?P<abbr>[^\\]*?)\][ ]?:[ ]*\n?[ ]*(?P<title>.*)$', re.MULTILINE)
 
  def test(self, parent: etree.Element, block: str) -> bool:
  return True
@@ -72,16 +72,8 @@ def run(self, parent: etree.Element, blocks: list[str]) -> bool:
  return False
 
  def _generate_pattern(self, text: str) -> str:
- """
- Given a string, returns a regex pattern to match that string.
-
- 'HTML' -> r'(?P<abbr>\b[H][T][M][L]\b)'
-
- Note: we force each char as a literal match via a character set (in brackets)
- as we don't know what they will be beforehand.
-
- """
- return f"(?P<abbr>\\b{ ''.join(f'[{ c }]' for c in text) }\\b)"
+ """ Given a string, returns a regex pattern to match that string. """
+ return f"(?P<abbr>\\b{ re.escape(text) }\\b)"
 
 
 class AbbrInlineProcessor(InlineProcessor):

tests/test_syntax/extensions/test_abbr.py

@@ -24,6 +24,7 @@
 
 
 class TestAbbr(TestCase):
+ maxDiff = None
 
  default_kwargs = {'extensions': ['abbr']}
 
@@ -260,28 +261,19 @@ def test_abbr_single_quoted(self):
  )
  )
 
- def test_abbr_ignore_special_chars(self):
+ def test_abbr_ignore_backslash(self):
  self.assertMarkdownRenders(
  self.dedent(
  r"""
- [^] [\\] [\]] []]
+ \\foo
 
- *[^]: Not an abbreviation
-
- *[\\]: Not an abbreviation
-
- *[\]]: Not an abbreviation
-
- *[]]: Not an abbreviation
+ *[\\foo]: Not an abbreviation
  """
  ),
  self.dedent(
  r"""
- <p>[^] [\] []] []]</p>
- <p>*[^]: Not an abbreviation</p>
- <p>*[\]: Not an abbreviation</p>
- <p>*[]]: Not an abbreviation</p>
- <p>*[]]: Not an abbreviation</p>
+ <p>\foo</p>
+ <p>*[\foo]: Not an abbreviation</p>
  """
  )
  )
@@ -301,3 +293,35 @@ def test_abbr_hyphen(self):
  """
  )
  )
+
+ def test_abbr_carrot(self):
+ self.assertMarkdownRenders(
+ self.dedent(
+ """
+ ABBR^abbr
+
+ *[ABBR^abbr]: Abbreviation
+ """
+ ),
+ self.dedent(
+ """
+ <p><abbr title="Abbreviation">ABBR^abbr</abbr></p>
+ """
+ )
+ )
+
+ def test_abbr_bracket(self):
+ self.assertMarkdownRenders(
+ self.dedent(
+ """
+ ABBR]abbr
+
+ *[ABBR]abbr]: Abbreviation
+ """
+ ),
+ self.dedent(
+ """
+ <p><abbr title="Abbreviation">ABBR]abbr</abbr></p>
+ """
+ )
+ )