Use set instead of list for block level elements

Using a set allows for better performances when checking for membership of a tag within block level elements.

Issue-1507: #1507

Author: pawamoy

markdown/core.py

@@ -50,7 +50,7 @@ class Markdown:
     Attributes:
         Markdown.tab_length (int): The number of spaces which correspond to a single tab. Default: `4`.
         Markdown.ESCAPED_CHARS (list[str]): List of characters which get the backslash escape treatment.
-        Markdown.block_level_elements (list[str]): List of HTML tags which get treated as block-level elements.
+        Markdown.block_level_elements (set[str]): Set of HTML tags which get treated as block-level elements.
             See [`markdown.util.BLOCK_LEVEL_ELEMENTS`][] for the full list of elements.
         Markdown.registeredExtensions (list[Extension]): List of extensions which have called
             [`registerExtension`][markdown.Markdown.registerExtension] during setup.
@@ -113,7 +113,7 @@ def __init__(self, **kwargs):
         ]
         """ List of characters which get the backslash escape treatment. """
 
-        self.block_level_elements: list[str] = BLOCK_LEVEL_ELEMENTS.copy()
+        self.block_level_elements: set[str] = BLOCK_LEVEL_ELEMENTS.copy()
 
         self.registeredExtensions: list[Extension] = []
         self.docType = ""  # TODO: Maybe delete this. It does not appear to be used anymore.

markdown/extensions/md_in_html.py

@@ -42,7 +42,7 @@ class HTMLExtractorExtra(HTMLExtractor):
 
     def __init__(self, md: Markdown, *args, **kwargs):
         # All block-level tags.
-        self.block_level_tags = set(md.block_level_elements.copy())
+        self.block_level_tags = md.block_level_elements.copy()
         # Block-level tags in which the content only gets span level parsing
         self.span_tags = set(
             ['address', 'dd', 'dt', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', 'legend', 'li', 'p', 'summary', 'td', 'th']

markdown/util.py

@@ -24,6 +24,7 @@
 
 from __future__ import annotations
 
+from collections.abc import Callable
 import re
 import sys
 import warnings
@@ -44,7 +45,60 @@
 """
 
 
-BLOCK_LEVEL_ELEMENTS: list[str] = [
+class _BlockLevelElements(set):
+    # ----------------------------------
+    # Methods common to `list` and `set`
+    # ----------------------------------
+    def copy(self) -> set[str]:
+        return _BlockLevelElements(super().copy())
+
+    def pop(self, index: int | None = None, /) -> str:
+        if index is not None:
+            warnings.warn("The index argument is deprecated and will be removed in the future", DeprecationWarning)
+        try:
+            return self.pop()
+        except KeyError:
+            warnings.warn("`pop` will raise a `KeyError` in the future", DeprecationWarning)
+            raise IndexError("pop from an empty set") from None
+
+    def remove(self, element: object) -> None:
+        try:
+            return self.remove(element)
+        except KeyError:
+            warnings.warn("`remove` will raise a `KeyError` in the future", DeprecationWarning)
+            raise ValueError(f"{element!r} not in set") from None
+
+    # --------------------------
+    # Methods specific to `list`
+    # --------------------------
+    def append(self, element: str) -> None:
+        warnings.warn("method `append` will be removed in the future", DeprecationWarning)
+        self.add(element)
+
+    def count(self, value: str) -> int:
+        warnings.warn("method `count` will be removed in the future", DeprecationWarning)
+        return 1 if value in self else 0
+
+    def extend(self, elements: list[str]) -> None:
+        warnings.warn("method `extend` will be removed in the future", DeprecationWarning)
+        self.update(elements)
+
+    def index(self, value, start=0, stop=0, /) -> int:
+        warnings.warn("method `index` will be removed in the future", DeprecationWarning)
+        return 0 if value in self else -1
+
+    def insert(self, index: int, element: str) -> None:
+        warnings.warn("method `insert` will be removed in the future", DeprecationWarning)
+        self.add(element)
+
+    def reverse(self) -> None:
+        warnings.warn("method `reverse` will be removed in the future", DeprecationWarning)
+
+    def sort(self, /, *, key: Callable | None = None, reverse: bool = False) -> None:
+        warnings.warn("method `sort` will be removed in the future", DeprecationWarning)
+
+
+BLOCK_LEVEL_ELEMENTS: set[str] = _BlockLevelElements({
     # Elements which are invalid to wrap in a `<p>` tag.
     # See https://w3c.github.io/html/grouping-content.html#the-p-element
     'address', 'article', 'aside', 'blockquote', 'details', 'div', 'dl',
@@ -56,9 +110,9 @@
     'math', 'map', 'noscript', 'output', 'object', 'option', 'progress', 'script',
     'style', 'summary', 'tbody', 'td', 'textarea', 'tfoot', 'th', 'thead', 'tr', 'video',
     'center'
-]
+})
 """
-List of HTML tags which get treated as block-level elements. Same as the `block_level_elements`
+Set of HTML tags which get treated as block-level elements. Same as the `block_level_elements`
 attribute of the [`Markdown`][markdown.Markdown] class. Generally one should use the
 attribute on the class. This remains for compatibility with older extensions.
 """

tests/test_apis.py

@@ -920,7 +920,7 @@ class TestBlockAppend(unittest.TestCase):
     def testBlockAppend(self):
         """ Test that appended escapes are only in the current instance. """
         md = markdown.Markdown()
-        md.block_level_elements.append('test')
+        md.block_level_elements.add('test')
         self.assertEqual('test' in md.block_level_elements, True)
         md2 = markdown.Markdown()
         self.assertEqual('test' not in md2.block_level_elements, True)