AbbrTreeprocessor now sorts the abbreviation

list by length before processing the element tree

This ensures that multi-word abbreviations are
implemented even if an abbreviation exists for
one of those component words.

To test this,
added `test_abbr_superset_vs_subset()` to
`tests.test_syntax.extensions.test_abbr`.

Author: nbanyan

docs/changelog.md

@@ -23,6 +23,10 @@ better reflects what it is. `AbbrPreprocessor` has been deprecated.
 
 A call to `Markdown.reset()` now clears all previously defined abbreviations.
 
+Abbreviations are now sorted by length before executing `AbbrTreeprocessor`
+to ensure that multi-word abbreviations are implemented even if an abbreviation
+exists for one of those component words.
+
 ### Fixed
 
 * Fixed links to source code on GitHub from the documentation (#1453).

markdown/extensions/abbr.py

@@ -92,7 +92,9 @@ def run(self, root: etree.Element) -> etree.Element | None:
             # No abbreviations defined. Skip running processor.
             return
         # Build and compile regex
-        self.RE = re.compile(f"\\b(?:{ '|'.join(re.escape(key) for key in self.abbrs) })\\b")
+        abbr_list = list(self.abbrs.keys())
+        abbr_list.sort(key=len, reverse=True)
+        self.RE = re.compile(f"\\b(?:{ '|'.join(re.escape(key) for key in abbr_list) })\\b")
         # Step through tree and modify on matches
         self.iter_element(root)
 

tests/test_syntax/extensions/test_abbr.py

@@ -383,6 +383,24 @@ def test_abbr_with_attr_list(self):
             extensions=['abbr', 'attr_list']
         )
 
+    def test_abbr_superset_vs_subset(self):
+        self.assertMarkdownRenders(
+            self.dedent(
+                """
+                abbr, SS, and abbr-SS should have different definitions.
+                
+                *[abbr]: Abbreviation Definition
+                *[abbr-SS]: Abbreviation Superset Definition
+                *[SS]: Superset Definition
+                """
+            ),
+            self.dedent(
+                """
+                <p><abbr title="Abbreviation Definition">abbr</abbr>, <abbr title="Superset Definition">SS</abbr>, and <abbr title="Abbreviation Superset Definition">abbr-SS</abbr> should have different definitions.</p>
+                """
+            )
+        )
+
     def test_abbr_reset(self):
         ext = AbbrExtension()
         md = Markdown(extensions=[ext])