wordlift-sdk 2.8.0__tar.gz → 2.9.0__tar.gz

{wordlift_sdk-2.8.0 → wordlift_sdk-2.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wordlift-sdk
-Version: 2.8.0
+Version: 2.9.0
 Summary: 
 Author: David Riccitelli
 Author-email: david@wordlift.io

{wordlift_sdk-2.8.0 → wordlift_sdk-2.9.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "wordlift-sdk"
-version = "2.8.0"
+version = "2.9.0"
 description = ""
 authors = ["David Riccitelli <david@wordlift.io>"]
 readme = "README.md"

{wordlift_sdk-2.8.0 → wordlift_sdk-2.9.0}/wordlift_sdk/utils/__init__.py

@@ -1,10 +1,13 @@
 from .create_dataframe_from_google_sheets import create_dataframe_from_google_sheets
 from .create_dataframe_of_entities_by_types import create_dataframe_of_entities_by_types
-from .create_dataframe_of_entities_with_embedding_vectors import create_dataframe_of_entities_with_embedding_vectors
+from .create_dataframe_of_entities_with_embedding_vectors import (
+    create_dataframe_of_entities_with_embedding_vectors,
+)
 from .create_dataframe_of_url_iri import create_dataframe_of_url_iri
 from .create_entity_patch_request import create_entity_patch_request
 from .delayed import create_delayed
 from .get_me import get_me
+from .html_converter import HtmlConverter
 
 __all__ = [
     "create_dataframe_from_google_sheets",
@@ -13,5 +16,6 @@ __all__ = [
     "create_dataframe_of_url_iri",
     "create_entity_patch_request",
     "create_delayed",
-    "get_me"
+    "get_me",
+    "HtmlConverter",
 ]

wordlift_sdk-2.9.0/wordlift_sdk/utils/html_converter.py

@@ -0,0 +1,56 @@
+"""HTML to XHTML conversion utility."""
+
+from __future__ import annotations
+
+import re
+from typing import Any
+
+_INVALID_XML_CHARS_RE = re.compile(r"[\x00-\x08\x0B\x0C\x0E-\x1F\x7F]")
+_XML_NAME_RE = re.compile(r"^[A-Za-z_][A-Za-z0-9_.:-]*$")
+
+
+class HtmlConverter:
+    """Converts HTML to XHTML."""
+
+    def convert(self, html: str) -> str:
+        """
+        Convert an HTML string to a valid XHTML string.
+
+        Args:
+            html: The raw HTML string.
+
+        Returns:
+            A sanitized XHTML string.
+        """
+        html = re.sub(r"<!DOCTYPE[^>]*>", "", html, flags=re.IGNORECASE)
+        html = self._strip_invalid_xml_chars(html)
+        try:
+            from lxml import html as lxml_html
+        except ImportError as exc:
+            raise ImportError(
+                "lxml is required for XHTML output. Install with: pip install lxml"
+            ) from exc
+
+        try:
+            parser = lxml_html.HTMLParser(encoding="utf-8", recover=True)
+            doc = lxml_html.document_fromstring(html, parser=parser)
+            self._sanitize_xhtml_tree(doc)
+            xhtml = lxml_html.tostring(doc, encoding="unicode", method="xml")
+            return self._strip_invalid_xml_chars(xhtml)
+        except Exception as exc:
+            raise RuntimeError("Failed to convert HTML to XHTML.") from exc
+
+    def _strip_invalid_xml_chars(self, value: str) -> str:
+        return _INVALID_XML_CHARS_RE.sub("", value)
+
+    def _sanitize_xhtml_tree(self, doc: Any) -> None:
+        for element in doc.iter():
+            if not hasattr(element, "attrib"):
+                continue
+            for attr in list(element.attrib):
+                if not _XML_NAME_RE.match(attr):
+                    del element.attrib[attr]
+                    continue
+                value = element.attrib.get(attr)
+                if isinstance(value, str):
+                    element.attrib[attr] = self._strip_invalid_xml_chars(value)