Add table caption support.

unstructured/documents/html.py

@@ -37,7 +37,7 @@
 )
 from unstructured.utils import htmlify_matrix_of_cell_texts
 
-TEXT_TAGS: Final[List[str]] = ["p", "a", "td", "span", "font"]
+TEXT_TAGS: Final[List[str]] = ["p", "a", "td", "span", "font", "caption"]
 LIST_ITEM_TAGS: Final[List[str]] = ["li", "dd"]
 LIST_TAGS: Final[List[str]] = ["ul", "ol", "dl"]
 HEADING_TAGS: Final[List[str]] = ["h1", "h2", "h3", "h4", "h5", "h6"]
@@ -159,7 +159,7 @@ def _parse_pages_from_element_tree(self) -> List[Page]:
         for article in articles:
             descendanttag_elems: Tuple[etree._Element, ...] = ()
             for tag_elem in article.iter():
-                if tag_elem in descendanttag_elems:
+                if tag_elem in descendanttag_elems and tag_elem.tag != "caption":
                     # Prevent repeating something that's been flagged as text as we chase it
                     # down a chain
                     continue