html-to-markdown 1.5.0__py3-none-any.whl → 1.8.0__py3-none-any.whl

html_to_markdown/utils.py

@@ -6,18 +6,25 @@ from html_to_markdown.constants import line_beginning_re
 
 
 def chomp(text: str) -> tuple[str, str, str]:
-    """If the text in an inline tag like b, a, or em contains a leading or trailing
-    space, strip the string and return a space as suffix of prefix, if needed.
+    """Simplified whitespace handling for inline elements.
+
+    For semantic markdown output, preserves leading/trailing spaces as single spaces
+    and normalizes internal whitespace.
 
     Args:
         text: The text to chomp.
 
     Returns:
-        A tuple containing the prefix, suffix, and the stripped text.
+        A tuple containing the prefix, suffix, and the normalized text.
     """
-    prefix = " " if text and text[0] == " " else ""
-    suffix = " " if text and text[-1] == " " else ""
+    if not text:
+        return "", "", ""
+
+    prefix = " " if text.startswith((" ", "\t")) else ""
+    suffix = " " if text.endswith((" ", "\t")) else ""
+
     text = text.strip()
+
     return prefix, suffix, text
 
 

{html_to_markdown-1.5.0.dist-info → html_to_markdown-1.8.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: html-to-markdown
-Version: 1.5.0
+Version: 1.8.0
 Summary: A modern, type-safe Python library for converting HTML to Markdown with comprehensive tag support and customizable options
 Author-email: Na'aman Hirschfeld <nhirschfeld@gmail.com>
 License: MIT
@@ -32,6 +32,9 @@ Requires-Python: >=3.9
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: beautifulsoup4>=4.13.4
+Requires-Dist: nh3>=0.2.21
+Provides-Extra: lxml
+Requires-Dist: lxml>=5; extra == "lxml"
 Dynamic: license-file
 
 # html-to-markdown
@@ -60,6 +63,28 @@ Python 3.9+.
 pip install html-to-markdown
 ```
 
+### Optional lxml Parser
+
+For improved performance, you can install with the optional lxml parser:
+
+```shell
+pip install html-to-markdown[lxml]
+```
+
+The lxml parser offers:
+
+- **~30% faster HTML parsing** compared to the default html.parser
+- Better handling of malformed HTML
+- More robust parsing for complex documents
+
+Once installed, lxml is automatically used by default for better performance. You can explicitly specify a parser if needed:
+
+```python
+result = convert_to_markdown(html)  # Auto-detects: uses lxml if available, otherwise html.parser
+result = convert_to_markdown(html, parser="lxml")  # Force lxml (requires installation)
+result = convert_to_markdown(html, parser="html.parser")  # Force built-in parser
+```
+
 ## Quick Start
 
 Convert HTML to Markdown with a single function call:
@@ -180,18 +205,19 @@ Custom converters take precedence over the built-in converters and can be used a
 
 ### Key Configuration Options
 
-| Option              | Type | Default          | Description                                            |
-| ------------------- | ---- | ---------------- | ------------------------------------------------------ |
-| `extract_metadata`  | bool | `True`           | Extract document metadata as comment header            |
-| `convert_as_inline` | bool | `False`          | Treat content as inline elements only                  |
-| `heading_style`     | str  | `'underlined'`   | Header style (`'underlined'`, `'atx'`, `'atx_closed'`) |
-| `highlight_style`   | str  | `'double-equal'` | Highlight style (`'double-equal'`, `'html'`, `'bold'`) |
-| `stream_processing` | bool | `False`          | Enable streaming for large documents                   |
-| `autolinks`         | bool | `True`           | Auto-convert URLs to Markdown links                    |
-| `bullets`           | str  | `'*+-'`          | Characters to use for bullet points                    |
-| `escape_asterisks`  | bool | `True`           | Escape * characters                                    |
-| `wrap`              | bool | `False`          | Enable text wrapping                                   |
-| `wrap_width`        | int  | `80`             | Text wrap width                                        |
+| Option              | Type | Default          | Description                                                     |
+| ------------------- | ---- | ---------------- | --------------------------------------------------------------- |
+| `extract_metadata`  | bool | `True`           | Extract document metadata as comment header                     |
+| `convert_as_inline` | bool | `False`          | Treat content as inline elements only                           |
+| `heading_style`     | str  | `'underlined'`   | Header style (`'underlined'`, `'atx'`, `'atx_closed'`)          |
+| `highlight_style`   | str  | `'double-equal'` | Highlight style (`'double-equal'`, `'html'`, `'bold'`)          |
+| `stream_processing` | bool | `False`          | Enable streaming for large documents                            |
+| `parser`            | str  | auto-detect      | BeautifulSoup parser (auto-detects `'lxml'` or `'html.parser'`) |
+| `autolinks`         | bool | `True`           | Auto-convert URLs to Markdown links                             |
+| `bullets`           | str  | `'*+-'`          | Characters to use for bullet points                             |
+| `escape_asterisks`  | bool | `True`           | Escape * characters                                             |
+| `wrap`              | bool | `False`          | Enable text wrapping                                            |
+| `wrap_width`        | int  | `80`             | Text wrap width                                                 |
 
 For a complete list of all 20+ options, see the [Configuration Reference](#configuration-reference) section below.
 
@@ -379,6 +405,17 @@ uv run python -m html_to_markdown input.html
 uv build
 ```
 
+## Performance
+
+The library is optimized for performance with several key features:
+
+- **Efficient ancestor caching**: Reduces repeated DOM traversals using context-aware caching
+- **Streaming support**: Process large documents in chunks to minimize memory usage
+- **Optional lxml parser**: ~30% faster parsing for complex HTML documents
+- **Optimized string operations**: Minimizes string concatenations in hot paths
+
+Typical throughput: ~2 MB/s for regular processing on modern hardware.
+
 ## License
 
 This library uses the MIT license.

html_to_markdown-1.8.0.dist-info/RECORD

@@ -0,0 +1,16 @@
+html_to_markdown/__init__.py,sha256=TzZzhZDJHeXW_3B9zceYehz2zlttqdLsDr5un8stZLM,653
+html_to_markdown/__main__.py,sha256=DJyJX7NIK0BVPNS2r3BYJ0Ci_lKHhgVOpw7ZEqACH3c,323
+html_to_markdown/cli.py,sha256=8xlgSEcnqsSM_dr1TCSgPDAo09YvUtO78PvDFivFFdg,6973
+html_to_markdown/constants.py,sha256=8vqANd-7wYvDzBm1VXZvdIxS4Xom4Ov_Yghg6jvmyio,584
+html_to_markdown/converters.py,sha256=COC2KqPelJlMCY5eXUS5gdiPOG8Yzx0U719FeXPw3GA,55514
+html_to_markdown/exceptions.py,sha256=s1DaG6A23rOurF91e4jryuUzplWcC_JIAuK9_bw_4jQ,1558
+html_to_markdown/preprocessor.py,sha256=S4S1ZfLC_hkJVgmA5atImTyWQDOxfHctPbaep2QtyrQ,11248
+html_to_markdown/processing.py,sha256=wkbhLg42U3aeVQSZFuzGt5irtN037XzRKpCE71QYZXI,36520
+html_to_markdown/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+html_to_markdown/utils.py,sha256=QgWPzmpZKFd6wDTe8IY3gbVT3xNzoGV3PBgd17J0O-w,2066
+html_to_markdown-1.8.0.dist-info/licenses/LICENSE,sha256=3J_HR5BWvUM1mlIrlkF32-uC1FM64gy8JfG17LBuheQ,1122
+html_to_markdown-1.8.0.dist-info/METADATA,sha256=6pgiK4p0A77axLfD8MH1EGgzifP06koVV8KWS_5-iYk,17175
+html_to_markdown-1.8.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+html_to_markdown-1.8.0.dist-info/entry_points.txt,sha256=xmFijrTfgYW7lOrZxZGRPciicQHa5KiXKkUhBCmICtQ,116
+html_to_markdown-1.8.0.dist-info/top_level.txt,sha256=Ev6djb1c4dSKr_-n4K-FpEGDkzBigXY6LuZ5onqS7AE,17
+html_to_markdown-1.8.0.dist-info/RECORD,,

html_to_markdown-1.5.0.dist-info/RECORD

@@ -1,14 +0,0 @@
-html_to_markdown/__init__.py,sha256=ZfPBBhhxQJTFQiOX-5OtgSMP2xFs5UUJeYmLL-AawoQ,265
-html_to_markdown/__main__.py,sha256=DJyJX7NIK0BVPNS2r3BYJ0Ci_lKHhgVOpw7ZEqACH3c,323
-html_to_markdown/cli.py,sha256=WzQVr97jKECEZwW-xIJofSl3v4EhqU-De7XRQjmgc08,7179
-html_to_markdown/constants.py,sha256=8vqANd-7wYvDzBm1VXZvdIxS4Xom4Ov_Yghg6jvmyio,584
-html_to_markdown/converters.py,sha256=xEVT0rQGWBU4V-HBF7Mmm-2XGPB1cboAmKlF6vcxS4k,59456
-html_to_markdown/processing.py,sha256=nqpPiRZu5B--E9dJ9AOwH2r1alg-ynv7ie63rtIb9Ls,28661
-html_to_markdown/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-html_to_markdown/utils.py,sha256=HJUDej5HSpXRtYv-OkCyD0hwnPnVfQCwY6rBRlIOt9s,1989
-html_to_markdown-1.5.0.dist-info/licenses/LICENSE,sha256=3J_HR5BWvUM1mlIrlkF32-uC1FM64gy8JfG17LBuheQ,1122
-html_to_markdown-1.5.0.dist-info/METADATA,sha256=nGVi7PSapoEUNTn5WGBW2g744dZTxaXCcFxl_ILeb9s,15641
-html_to_markdown-1.5.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-html_to_markdown-1.5.0.dist-info/entry_points.txt,sha256=xmFijrTfgYW7lOrZxZGRPciicQHa5KiXKkUhBCmICtQ,116
-html_to_markdown-1.5.0.dist-info/top_level.txt,sha256=Ev6djb1c4dSKr_-n4K-FpEGDkzBigXY6LuZ5onqS7AE,17
-html_to_markdown-1.5.0.dist-info/RECORD,,