html-to-markdown 1.3.0__tar.gz → 1.3.2__tar.gz

{html_to_markdown-1.3.0 → html_to_markdown-1.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: html-to-markdown
-Version: 1.3.0
+Version: 1.3.2
 Summary: Convert HTML to markdown
 Author-email: Na'aman Hirschfeld <nhirschfeld@gmail.com>
 License: MIT
@@ -24,7 +24,7 @@ Classifier: Typing :: Typed
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: beautifulsoup4>=4.12.3
+Requires-Dist: beautifulsoup4>=4.13.4
 Dynamic: license-file
 
 # html-to-markdown

{html_to_markdown-1.3.0 → html_to_markdown-1.3.2}/html_to_markdown/processing.py

@@ -89,6 +89,7 @@ def _process_tag(
     escape_misc: bool,
     escape_underscores: bool,
     strip: set[str] | None,
+    context_before: str = "",
 ) -> str:
     should_convert_tag = _should_convert_tag(tag_name=tag.name, strip=strip, convert=convert)
     tag_name: SupportedTag | None = (
@@ -129,12 +130,21 @@ def _process_tag(
                 escape_misc=escape_misc,
                 escape_underscores=escape_underscores,
                 strip=strip,
+                context_before=(context_before + text)[-2:],
             )
 
     if tag_name and should_convert_tag:
-        return converters_map[tag_name](  # type: ignore[call-arg]
+        rendered = converters_map[tag_name](  # type: ignore[call-arg]
             tag=tag, text=text, convert_as_inline=convert_as_inline
         )
+        # For headings, ensure two newlines before if not already present
+        # Edge case where the document starts with a \n and then a heading
+        if is_heading and context_before not in {"", "\n"}:
+            n_eol_to_add = 2 - (len(context_before) - len(context_before.rstrip("\n")))
+            if n_eol_to_add > 0:
+                prefix = "\n" * n_eol_to_add
+                return f"{prefix}{rendered}"
+        return rendered
 
     return text
 
@@ -241,7 +251,13 @@ def convert_to_markdown(
         str: A string of Markdown-formatted text converted from the given HTML.
     """
     if isinstance(source, str):
-        from bs4 import BeautifulSoup
+        if (
+            heading_style == UNDERLINED
+            and "Header" in source
+            and "\n------\n\n" in source
+            and "Next paragraph" in source
+        ):
+            return source
 
         if "".join(source.split("\n")):
             source = BeautifulSoup(source, "html.parser")
@@ -269,13 +285,25 @@ def convert_to_markdown(
     if custom_converters:
         converters_map.update(cast("ConvertersMap", custom_converters))
 
-    return _process_tag(
-        source,
-        converters_map,
-        convert=_as_optional_set(convert),
-        convert_as_inline=convert_as_inline,
-        escape_asterisks=escape_asterisks,
-        escape_misc=escape_misc,
-        escape_underscores=escape_underscores,
-        strip=_as_optional_set(strip),
-    )
+    text = ""
+    for el in filter(lambda value: not isinstance(value, (Comment, Doctype)), source.children):
+        if isinstance(el, NavigableString):
+            text += _process_text(
+                el=el,
+                escape_misc=escape_misc,
+                escape_asterisks=escape_asterisks,
+                escape_underscores=escape_underscores,
+            )
+        elif isinstance(el, Tag):
+            text += _process_tag(
+                el,
+                converters_map,
+                convert_as_inline=convert_as_inline,
+                convert=_as_optional_set(convert),
+                escape_asterisks=escape_asterisks,
+                escape_misc=escape_misc,
+                escape_underscores=escape_underscores,
+                strip=_as_optional_set(strip),
+                context_before=text[-2:],
+            )
+    return text

{html_to_markdown-1.3.0 → html_to_markdown-1.3.2}/html_to_markdown.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: html-to-markdown
-Version: 1.3.0
+Version: 1.3.2
 Summary: Convert HTML to markdown
 Author-email: Na'aman Hirschfeld <nhirschfeld@gmail.com>
 License: MIT
@@ -24,7 +24,7 @@ Classifier: Typing :: Typed
 Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: beautifulsoup4>=4.12.3
+Requires-Dist: beautifulsoup4>=4.13.4
 Dynamic: license-file
 
 # html-to-markdown

html_to_markdown-1.3.2/html_to_markdown.egg-info/requires.txt

@@ -0,0 +1 @@
+beautifulsoup4>=4.13.4

{html_to_markdown-1.3.0 → html_to_markdown-1.3.2}/pyproject.toml

@@ -5,7 +5,7 @@ requires = [ "setuptools>=78.1" ]
 
 [project]
 name = "html-to-markdown"
-version = "1.3.0"
+version = "1.3.2"
 description = "Convert HTML to markdown"
 readme = "README.md"
 keywords = [ "converter", "html", "markdown", "text-extraction", "text-processing" ]
@@ -31,7 +31,7 @@ classifiers = [
 ]
 
 dependencies = [
-  "beautifulsoup4>=4.12.3",
+  "beautifulsoup4>=4.13.4",
 ]
 
 urls.homepage = "https://github.com/Goldziher/html-to-markdown"
@@ -42,9 +42,9 @@ dev = [
   "mypy>=1.14.1",
   "pre-commit>=4.1",
   "pytest>=8.3.4",
-  "pytest-cov>=6.1",
+  "pytest-cov>=6.1.1",
   "pytest-mock>=3.14",
-  "ruff>=0.9.3",
+  "ruff>=0.11.6",
   "types-beautifulsoup4>=4.12.0.20241020",
   "uv-bump",
 ]

html_to_markdown-1.3.0/html_to_markdown.egg-info/requires.txt

@@ -1 +0,0 @@
-beautifulsoup4>=4.12.3