Fix: avoid elements sharing the same memory address (#2940)

This PR attempts to fix a memory issue, which resulted in errors like
this: #2931
The root cause seems to be in how ListItems are being combined, not in
how hashes or parent IDs are updated.

When `assign_and_map_hash_ids()` is called and elements (or elements'
metadata) do not have unique memory addresses, then updating the
parent_id of one element will also overwrite the parent_id of some other
element.

---------

Co-authored-by: cragwolfe <[email protected]>

Author: micmarty-deepsense

CHANGELOG.md

@@ -1,11 +1,12 @@
-## 0.13.5-dev0
+## 0.13.5
 
 ### Enhancements
 
 ### Features
 
 ### Fixes
 
+* **KeyError raised when updating parent_id** In the past, combining `ListItem` elements could result in reusing the same memory location which then led to unexpected side effects when updating element IDs.
 * **Bump unstructured-inference==0.7.29**: table transformer predictions are now removed if confidence is below threshold
 
 ## 0.13.4

example-docs/a1977-backus-p21.pdf

@@ -887,12 +887,20 @@ def test_partition_pdf_word_bbox_not_char(
     filename=example_doc_path("interface-config-guide-p93.pdf"),
 ):
     try:
-        elements = pdf.partition_pdf(filename=filename)
+        elements = pdf.partition_pdf(filename=filename, strategy="fast")
     except Exception as e:
         raise ("Partitioning fail: %s" % e)
     assert len(elements) == 17
 
 
+def test_partition_pdf_fast_no_mapping_errors(
+    filename=example_doc_path("a1977-backus-p21.pdf"),
+):
+    """Verify there is no regression for https://github.com/Unstructured-IO/unstructured/pull/2940,
+    failing to map old parent_id's to new"""
+    pdf.partition_pdf(filename=filename, strategy="fast")
+
+
 def test_partition_pdf_raises_TypeError_for_invalid_languages():
     filename = example_doc_path("chevron-page.pdf")
     with pytest.raises(TypeError):

test_unstructured/partition/pdf_image/test_pdf.py

@@ -1 +1 @@
-__version__ = "0.13.5-dev0"  # pragma: no cover
+__version__ = "0.13.5"  # pragma: no cover

unstructured/__version__.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import contextlib
+import copy
 import io
 import os
 import re
@@ -867,7 +868,7 @@ def _combine_coordinates_into_element1(
         points=points,
         system=coordinate_system,
     )
-    return element1
+    return copy.deepcopy(element1)
 
 
 def convert_pdf_to_images(