fix: chunking dropping table content (#4352)

## Preserve mixed-content tail text in table HTML

### Summary
`HtmlTable.from_html_text` compactification cleared every element's
`.tail`, silently dropping any text between inline children of a `<td>`.
In practice this deleted real content like `": do not..."` after
`<b>NOTICE</b>`, bullet lines following each `<br/>`, and any inter-tag
prose in rich table cells — visible whenever a table was chunked (any
strategy: `basic`, `by_title`, `by_page`).

Pure-whitespace tails are still stripped (those are just pretty-print
indent). Tails carrying content are now kept, with internal whitespace
runs collapsed and a single leading/trailing space preserved so adjacent
text doesn't run together.

### Changes
- `unstructured/common/html_table.py`: replace the blanket `e.tail =
None` with whitespace-aware normalization.
- `test_unstructured/common/test_html_table.py`: two regression tests
covering mixed inline markup, `<br/>`-separated lines, and whitespace
collapsing.
- Bump to `0.22.28` + changelog entry.

<!-- CURSOR_SUMMARY -->
---

> [!NOTE]
> **Low Risk**
> Low risk: localized change to table HTML minification that preserves
previously-dropped text; main risk is minor whitespace/serialization
differences in downstream table chunking output.
> 
> **Overview**
> Fixes `HtmlTable.from_html_text()` compactification to **stop deleting
meaningful mixed-content tail text** (text between inline children like
`<b>foo</b> bar` or between `<br/>` siblings), while still dropping pure
formatting/indent whitespace and collapsing internal whitespace runs.
> 
> Adds regression tests for mixed inline markup and `<br/>`-separated
cell content, and bumps the library to `0.22.28` with a changelog entry
describing the fix.
> 
> <sup>Reviewed by [Cursor Bugbot](https://cursor.com/bugbot) for commit
fe27d35. Bugbot is set up for automated
code reviews on this repo. Configure
[here](https://www.cursor.com/dashboard/bugbot).</sup>
<!-- /CURSOR_SUMMARY -->

Author: badGarnet

CHANGELOG.md

@@ -1,3 +1,9 @@
+## 0.22.28
+
+### Fixes
+
+- **Preserve mixed-content tail text in table HTML**: `HtmlTable` compactification previously cleared every element's `.tail`, silently dropping real text between inline children. Pure-whitespace tails are still removed, but tails carrying content are now kept with internal whitespace collapsed.
+
 ## 0.22.27
 
 ### Fixes

test_unstructured/common/test_html_table.py

@@ -125,6 +125,28 @@ def it_removes_any_extra_whitespace_between_elements_and_normalizes_whitespace_i
         )
         assert html_table.html == "<table><tr><td>abc def ghi</td></tr></table>"
 
+    def it_preserves_tail_text_in_mixed_content_cells(self):
+        """Tail text (between inline children) carries real content and must not be dropped."""
+        html_table = HtmlTable.from_html_text(
+            "<table><tr>"
+            "<td><b>foo</b> bar <b>baz</b></td>"
+            "<td><b>x</b><br/>y<br/>z</td>"
+            "</tr></table>"
+        )
+        assert html_table.html == (
+            "<table><tr>"
+            "<td><b>foo</b> bar <b>baz</b></td>"
+            "<td><b>x</b><br/>y<br/>z</td>"
+            "</tr></table>"
+        )
+
+    def and_it_normalizes_whitespace_within_tail_text(self):
+        """Tail whitespace runs are collapsed, but leading/trailing spaces are kept."""
+        html_table = HtmlTable.from_html_text(
+            "<table><tr><td><b>a</b>   b\n  c  <b>d</b></td></tr></table>"
+        )
+        assert html_table.html == "<table><tr><td><b>a</b> b c <b>d</b></td></tr></table>"
+
     def it_can_serialize_the_table_element_to_str_html_text(self):
         table = fragment_fromstring("<table><tr><td>foobar</td></tr></table>")
         html_table = HtmlTable(table)

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.22.27"  # pragma: no cover
+__version__ = "0.22.28"  # pragma: no cover

unstructured/common/html_table.py

@@ -104,9 +104,19 @@ def from_html_text(cls, html_text: str) -> HtmlTable:
             if e.text:
                 e.text = " ".join(e.text.split())
 
-            # -- remove all tails, those are newline + indent if anything --
+            # -- normalize tails. A tail is the text between an element's closing tag and the
+            # -- start of the next sibling. Pure-whitespace tails are pretty-printing noise and
+            # -- can be dropped, but tails can also carry real content (e.g. mixed inline
+            # -- markup like `<b>foo</b> bar <b>baz</b>` or text between `<br/>` tags), which
+            # -- must be preserved.
             if e.tail:
-                e.tail = None
+                parts = e.tail.split()
+                if not parts:
+                    e.tail = None
+                else:
+                    prefix = " " if e.tail[0].isspace() else ""
+                    suffix = " " if e.tail[-1].isspace() else ""
+                    e.tail = prefix + " ".join(parts) + suffix
 
         return cls(table, header_row_idxs=header_row_idxs, source_row_htmls=source_row_htmls)