Merge branch 'master' into fix-50--no-validate

src/mdformat/_cli.py

@@ -8,11 +8,10 @@
 from pathlib import Path
 import shutil
 import sys
-import textwrap
 
 import mdformat
 from mdformat._conf import DEFAULT_OPTS, InvalidConfError, read_toml_opts
-from mdformat._util import detect_newline_type, is_md_equal
+from mdformat._util import cached_textwrapper, detect_newline_type, is_md_equal
 import mdformat.plugins
 import mdformat.renderer
 
@@ -418,9 +417,7 @@ def wrap_paragraphs(paragraphs: Iterable[str]) -> str:
         wrap_width = terminal_width
     else:
         wrap_width = 80
-    wrapper = textwrap.TextWrapper(
-        break_long_words=False, break_on_hyphens=False, width=wrap_width
-    )
+    wrapper = cached_textwrapper(wrap_width)
     return "\n\n".join(wrapper.fill(p) for p in paragraphs) + "\n"
 
 

src/mdformat/_util.py

@@ -2,7 +2,9 @@
 
 from collections.abc import Iterable, Mapping
 from contextlib import nullcontext
+import functools
 import re
+import textwrap
 from types import MappingProxyType
 from typing import Any, Literal
 
@@ -45,6 +47,15 @@ def build_mdit(
     return mdit
 
 
+# Chars that markdown-it-py escapes when rendering code_inline:
+# https://github.com/executablebooks/markdown-it-py/blob/c5161b550f3c6c0a98d77e8389872405e8f9f9ee/markdown_it/common/utils.py#L138
+# Note that "&" is not included as it is used in the escape sequences of
+# these characters.
+_invalid_html_code_chars = '<>"'
+# a regex str that matches all except above chars
+_valid_html_code_char_re = rf"[^{re.escape(_invalid_html_code_chars)}]"
+
+
 def is_md_equal(
     md1: str,
     md2: str,
@@ -69,10 +80,11 @@ def is_md_equal(
         if codeformatters:
             langs_re = "|".join(re.escape(lang) for lang in codeformatters)
             html = re.sub(
-                rf'<code class="language-(?:{langs_re})">.*</code>',
+                rf'<code class="language-(?:{langs_re})">'
+                rf"{_valid_html_code_char_re}*"
+                r"</code>",
                 "",
                 html,
-                flags=re.DOTALL,
             )
 
         # Reduce all whitespace to a single space
@@ -114,3 +126,14 @@ def detect_newline_type(md: str, eol_setting: str) -> Literal["\n", "\r\n"]:
     if eol_setting == "crlf":
         return "\r\n"
     return "\n"
+
+
+@functools.lru_cache
+def cached_textwrapper(width: int) -> textwrap.TextWrapper:
+    return textwrap.TextWrapper(
+        break_long_words=False,
+        break_on_hyphens=False,
+        width=width,
+        expand_tabs=False,
+        replace_whitespace=False,
+    )

src/mdformat/plugins.py

@@ -2,12 +2,14 @@
 
 import argparse
 from collections.abc import Callable, Mapping
-from typing import Any, Protocol
+from typing import TYPE_CHECKING, Any, Protocol
 
 from markdown_it import MarkdownIt
 
 from mdformat._compat import importlib_metadata
-from mdformat.renderer.typing import Postprocess, Render
+
+if TYPE_CHECKING:
+    from mdformat.renderer.typing import Postprocess, Render
 
 
 def _load_entrypoints(

src/mdformat/renderer/_context.py

@@ -5,14 +5,14 @@
 from contextlib import contextmanager
 import logging
 import re
-import textwrap
 from types import MappingProxyType
 from typing import TYPE_CHECKING, Any, Literal, NamedTuple
 
 from markdown_it.rules_block.html_block import HTML_SEQUENCES
 
 from mdformat import codepoints
 from mdformat._conf import DEFAULT_OPTS
+from mdformat._util import cached_textwrapper
 from mdformat.renderer._util import (
     RE_CHAR_REFERENCE,
     decimalify_leading,
@@ -344,13 +344,7 @@ def _wrap(text: str, *, width: int | Literal["no"]) -> str:
     if width == "no":
         return _recover_preserve_chars(text, replacements)
 
-    wrapper = textwrap.TextWrapper(
-        break_long_words=False,
-        break_on_hyphens=False,
-        width=width,
-        expand_tabs=False,
-        replace_whitespace=False,
-    )
+    wrapper = cached_textwrapper(width)
     wrapped = wrapper.fill(text)
     wrapped = _recover_preserve_chars(wrapped, replacements)
     return wrapped

tests/test_util.py

@@ -21,3 +21,30 @@ def test_is_md_equal():
 paragr"""
     assert not is_md_equal(md1, md2)
     assert is_md_equal(md1, md2, codeformatters=("js", "go"))
+
+
+def test_is_md_equal__not():
+    md1 = """
+```js
+console.log()
+```
+
+paragr
+
+```js
+console.log()
+```
+"""
+    md2 = """
+```js
+bonsole.l()g
+```
+
+A different paragraph
+
+```js
+console.log()
+```
+"""
+    assert not is_md_equal(md1, md2)
+    assert not is_md_equal(md1, md2, codeformatters=("js",))