feat: keep img tag's class attr (#4050)

This change affects partition html.

Previously when there is a table in the html, we clean any tags inside
the table of their `class` and `id` attributes. However, sometimes there
are images, `img` tags, present in a table and its `class` attribute
identifies some important information about the image. This change
preserves the `class` attribute for `img` tags inside a table. This
change is reflected in a table element's `metadata.text_as_html`
attribute.

Author: badGarnet

CHANGELOG.md

@@ -1,8 +1,9 @@
-## 0.18.5-dev1
+## 0.18.5
 
 ### Enhancements
 
 - **Bump dependencies and remove lingering Python 3.9 artifacts** Cleaned up some references to 3.9 that were left When we dropped Python 3.9 support.
+- **`text_as_html` for Table element now keeps `img` tag's `class` attribute** Previously in partition HTML any tag inside a table is stripped of its `class` attribute. Now this attribute is preserved for `img` tag in the table element's `metadata.text_as_html`.
 
 ### Features
 

test_unstructured/documents/test_ontology_to_unstructured_parsing.py

@@ -1,6 +1,7 @@
 from pathlib import Path
 
 import pytest
+from bs4 import BeautifulSoup
 
 from unstructured.chunking.basic import chunk_elements
 from unstructured.chunking.title import chunk_by_title
@@ -13,6 +14,7 @@
     Paragraph,
     Section,
     Table,
+    remove_ids_and_class_from_table,
 )
 from unstructured.embed.openai import OpenAIEmbeddingConfig, OpenAIEmbeddingEncoder
 from unstructured.partition.html import partition_html
@@ -24,6 +26,37 @@
 from unstructured.staging.base import elements_from_json
 
 
+def test_remove_ids_and_class_from_table():
+    html_text = """
+    <table>
+        <tr class="TableRow">
+            <td><img class="Signature" alt="cell 1"/></td>
+            <td>cell 2</td>
+        </tr>
+        <tr>
+            <td><IMG class="Signature" alt="cell 3"/></td>
+            <td>cell 4</td>
+        </tr>
+    </table>
+    """
+    soup = BeautifulSoup(html_text, "html.parser")
+    assert (
+        str(remove_ids_and_class_from_table(soup))
+        == """
+<table>
+<tr>
+<td><img alt="cell 1" class="Signature"/></td>
+<td>cell 2</td>
+</tr>
+<tr>
+<td><img alt="cell 3" class="Signature"/></td>
+<td>cell 4</td>
+</tr>
+</table>
+"""
+    )
+
+
 def test_page_number_is_passed_correctly():
     ontology = Document(
         children=[

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.18.5-dev1"  # pragma: no cover
+__version__ = "0.18.5"  # pragma: no cover

unstructured/documents/ontology.py

@@ -147,12 +147,23 @@ def page_number(self) -> int | None:
         return None
 
 
-def remove_ids_and_class_from_table(soup: BeautifulSoup):
+def remove_ids_and_class_from_table(soup: BeautifulSoup) -> BeautifulSoup:
+    """
+    Remove id and class attributes from tags inside tables,
+    except preserve class attributes for img tags.
+
+    Args:
+        soup: BeautifulSoup object containing the HTML
+
+    Returns:
+        BeautifulSoup: Modified soup with attributes removed
+    """
     for tag in soup.find_all(True):
-        if tag.name == "table":  # type: ignore
+        if tag.name.lower() == "table":  # type: ignore
             continue  # We keep table tag
-        tag.attrs.pop("class", None)  # type: ignore
         tag.attrs.pop("id", None)  # type: ignore
+        if tag.name.lower() != "img":  # type: ignore
+            tag.attrs.pop("class", None)  # type: ignore
     return soup