Rewrite emphasis handling

docs/changelog.md

@@ -20,6 +20,7 @@ See the [Contributing Guide](contributing.md) for details.
   performance for repeated inline patterns (#1619).
 * Officially support Python 3.15 and drop support for Python 3.10
 * Walk backtick runs in `BacktickInlineProcessor` without a regex (#1620).
+* Complete rework of emphasis handling to imporove nested emphasis handling better (#1632).
 
 ### Fixed
 

docs/extensions/api.md

@@ -349,7 +349,7 @@ Here are some convenience functions and other examples:
 
 | Class                                                                                | Kind      | Description                                                   |
 | -------------------------------------------------------------------------------------|-----------|---------------------------------------------------------------|
-| [`AsteriskProcessor`][markdown.inlinepatterns.AsteriskProcessor]                     | built-in  | Emphasis processor for handling strong and em matches inside asterisks |
+| [`DelimiterProcessor`][markdown.inlinepatterns.DelimiterProcessor]                    | built-in  | Emphasis processor for handling strong and em matches        |
 | [`WikiLinksInlineProcessor`][markdown.extensions.wikilinks.WikiLinksInlineProcessor] | extension | Link `[[article names]]` to wiki given in metadata            |
 | [`FootnoteInlineProcessor`][markdown.extensions.footnotes.FootnoteInlineProcessor]   | extension | Replaces footnote in text with link to footnote div at bottom |
 

markdown/core.py

@@ -28,7 +28,7 @@
 from .preprocessors import build_preprocessors
 from .blockprocessors import build_block_parser
 from .treeprocessors import build_treeprocessors
-from .inlinepatterns import build_inlinepatterns
+from .inlinepatterns import build_inlinepatterns, DelimiterProcessor
 from .postprocessors import build_postprocessors
 from .extensions import Extension
 from .serializers import to_html_string, to_xhtml_string
@@ -106,6 +106,7 @@ def __init__(self, **kwargs):
 
         """
 
+        self.last_run: float = 0.0
         self.tab_length: int = kwargs.get('tab_length', 4)
 
         self.ESCAPED_CHARS: list[str] = [
@@ -118,6 +119,7 @@ def __init__(self, **kwargs):
         self.registeredExtensions: list[Extension] = []
         self.docType = ""  # TODO: Maybe delete this. It does not appear to be used anymore.
         self.stripTopLevelTags: bool = True
+        self.delimiters: DelimiterProcessor | None = None
 
         self.build_parser()
 
@@ -270,6 +272,11 @@ def reset(self) -> Markdown:
         self.htmlStash.reset()
         self.references.clear()
 
+        if self.delimiters is not None:
+            self.delimiters.reset()
+            if self.delimiters not in self.inlinePatterns:
+                self.delimiters = None
+
         for extension in self.registeredExtensions:
             if hasattr(extension, 'reset'):
                 extension.reset()

markdown/extensions/legacy_em.py

@@ -14,29 +14,7 @@
 from __future__ import annotations
 
 from . import Extension
-from ..inlinepatterns import UnderscoreProcessor, EmStrongItem, EM_STRONG2_RE, STRONG_EM2_RE
-import re
-
-# _emphasis_
-EMPHASIS_RE = r'(_)([^_]+)\1'
-
-# __strong__
-STRONG_RE = r'(_{2})(.+?)\1'
-
-# __strong_em___
-STRONG_EM_RE = r'(_)\1(?!\1)([^_]+?)\1(?!\1)(.+?)\1{3}'
-
-
-class LegacyUnderscoreProcessor(UnderscoreProcessor):
-    """Emphasis processor for handling strong and em matches inside underscores."""
-
-    PATTERNS = [
-        EmStrongItem(re.compile(EM_STRONG2_RE, re.DOTALL | re.UNICODE), 'double', 'strong,em'),
-        EmStrongItem(re.compile(STRONG_EM2_RE, re.DOTALL | re.UNICODE), 'double', 'em,strong'),
-        EmStrongItem(re.compile(STRONG_EM_RE, re.DOTALL | re.UNICODE), 'double2', 'strong,em'),
-        EmStrongItem(re.compile(STRONG_RE, re.DOTALL | re.UNICODE), 'single', 'strong'),
-        EmStrongItem(re.compile(EMPHASIS_RE, re.DOTALL | re.UNICODE), 'single', 'em')
-    ]
+from ..inlinepatterns import DelimiterProcessor
 
 
 class LegacyEmExtension(Extension):
@@ -45,14 +23,17 @@ class LegacyEmExtension(Extension):
     def extendMarkdown(self, md):
         """ Register the processor.
 
-        | Class Instance                                                | Registry                                                         | Name   | Priority |
-        | ------------------------------------------------------------- | ---------------------------------------------------------------- | ------ | :------: |
-        | [`LegacyUnderscoreProcessor`][markdown.extensions.legacy_em.LegacyUnderscoreProcessor] | [`inlinepatterns`][markdown.inlinepatterns.build_inlinepatterns] | `em_strong2` | `50` |
+        | Class Instance                                                     | Registry                                                         | Name         | Priority |
+        | ------------------------------------------------------------------ | ---------------------------------------------------------------- | ------------ | :------: |
+        | [`DelimiterProcessor`][markdown.inlinepatterns.DelimiterProcessor] | [`inlinepatterns`][markdown.inlinepatterns.build_inlinepatterns] | `em_strong`  | `60`     |
 
         """
-        # flake8: noqa: E501 48-50
-        md.inlinePatterns.register(LegacyUnderscoreProcessor(r'_'), 'em_strong2', 50)
+        # flake8: noqa: E501 27-29
 
+        if md.delimiters is not None:
+            md.delimiters.add('_', 'strong,em')
+        else:
+            md.inlinePatterns.register(DelimiterProcessor('_', 'strong,em', md), 'em_strong', 60)
 
 def makeExtension(**kwargs):  # pragma: no cover
     """ Return an instance of the `LegacyEmExtension` """