Improved html tag find/replace with selectolax.

pyproject.toml

@@ -45,6 +45,7 @@ dependencies = [
     "cydifflib>=1.2.0",
     "tiktoken>=0.9.0",
     "regex>=2024.11.6",
+    "selectolax>=0.3.32",
 ]
 
 [project.optional-dependencies]

src/chopdiff/html/__init__.py

@@ -1,11 +1,6 @@
 # flake8: noqa: F401
 
 from chopdiff.html.extractor import ContentNotFound, Extractor, Match
-from chopdiff.html.html_find_tags import (
-    TagMatch,
-    html_extract_attribute_value,
-    html_find_tag,
-)
 from chopdiff.html.html_in_md import (
     Attrs,
     ClassNames,
@@ -24,6 +19,13 @@
     tag_with_attrs,
 )
 from chopdiff.html.html_plaintext import html_to_plaintext, plaintext_to_html
+from chopdiff.html.html_tags import (
+    TagMatch,
+    html_extract_attribute_value,
+    html_find_tag,
+    rewrite_html_img_urls,
+    rewrite_html_tag_attr,
+)
 from chopdiff.html.timestamps import (
     TimestampExtractor,
     extract_timestamp,
@@ -39,6 +41,8 @@
     "TagMatch",
     "html_extract_attribute_value",
     "html_find_tag",
+    "rewrite_html_img_urls",
+    "rewrite_html_tag_attr",
     "Wrapper",
     "div_wrapper",
     "escape_md_html",