fix: type error string indices bug (#1940)

Fix TypeError: string indices must be integers. The `annotation_dict`
variable is conditioned to be `None` if instance type is not dict. Then
we add logic to skip the attempt if the value is `None`.

CHANGELOG.md

@@ -23,6 +23,7 @@
 * **Fix ingest pipeline to be able to use chunking and embedding together** Problem: When ingest pipeline was using chunking and embedding together, embedding outputs were empty and the outputs of chunking couldn't be re-read into memory and be forwarded to embeddings. Fix: Added CompositeElement type to TYPE_TO_TEXT_ELEMENT_MAP to be able to process CompositeElements with unstructured.staging.base.isd_to_elements
 * **Fix unnecessary mid-text chunk-splitting.** The "pre-chunker" did not consider separator blank-line ("\n\n") length when grouping elements for a single chunk. As a result, sections were frequently over-populated producing a over-sized chunk that required mid-text splitting.
 * **Fix frequent dissociation of title from chunk.** The sectioning algorithm included the title of the next section with the prior section whenever it would fit, frequently producing association of a section title with the prior section and dissociating it from its actual section. Fix this by performing combination of whole sections only.
+* **Fix PDF attempt to get dict value from string.** Fixes a rare edge case that prevented some PDF's from being partitioned. The `get_uris_from_annots` function tried to access the dictionary value of a string instance variable. Assign `None` to the annotation variable if the instance type is not dictionary to avoid the erroneous attempt.
 
 ## 0.10.27
 

test_unstructured/partition/pdf_image/test_pdf.py

@@ -19,6 +19,7 @@ from unstructured.documents.elements import (
     Title,
 )
 from unstructured.partition import ocr, pdf, strategies
+from unstructured.partition.pdf import get_uris_from_annots
 from unstructured.partition.utils.constants import UNSTRUCTURED_INCLUDE_DEBUG_METADATA
 
 
@@ -981,6 +982,51 @@ def test_ocr_language_passes_through(strategy, ocr_func):
     assert kwargs["lang"] == "kor"
 
 
+@pytest.mark.parametrize(
+    ("annots", "height", "coordinate_system", "page_number", "expected"),
+    [
+        (["BS", "BE"], 300, PixelSpace(300, 300), 1, 0),
+        (
+            [
+                {
+                    "Type": "/'Annot'",
+                    "Subtype": "/'Link'",
+                    "A": {
+                        "Type": "/'Action'",
+                        "S": "/'URI'",
+                        "URI": "b'https://layout-parser.github.io'",
+                    },
+                    "BS": {"S": "/'S'", "W": 1},
+                    "Border": [0, 0, 1],
+                    "C": [0, 1, 1],
+                    "H": "/'I'",
+                    "Rect": [304.055, 224.156, 452.472, 234.368],
+                },
+                {
+                    "Type": "/'Annot'",
+                    "Subtype": "/'Link'",
+                    "A": {"S": "/'GoTo'", "D": "b'cite.harley2015evaluation'"},
+                    "BS": {"S": "/'S'", "W": 1},
+                    "Border": [0, 0, 1],
+                    "C": [0, 1, 0],
+                    "H": "/'I'",
+                    "Rect": [468.305, 128.081, 480.26, 136.494],
+                },
+            ],
+            792,
+            PixelSpace(612, 792),
+            1,
+            2,
+        ),
+    ],
+)
+def test_get_uris_from_annots_string_annotation(
+    annots, height, coordinate_system, page_number, expected
+):
+    annotation_list = get_uris_from_annots(annots, height, coordinate_system, page_number)
+    assert len(annotation_list) == expected
+
+
 @pytest.mark.parametrize("file_mode", ["filename", "rb", "spool"])
 @pytest.mark.parametrize(
     ("filename", "is_image"),

unstructured/partition/pdf.py

@@ -921,8 +921,14 @@ def get_uris_from_annots(
     """
     annotation_list = []
     for annotation in annots:
-        annotation_dict = try_resolve(annotation)
-        if str(annotation_dict["Subtype"]) != "/'Link'" or "A" not in annotation_dict:
+        annotation_dict = (
+            try_resolve(annotation) if isinstance(try_resolve(annotation), dict) else None
+        )
+        if (
+            not annotation_dict
+            or str(annotation_dict["Subtype"]) != "/'Link'"
+            or "A" not in annotation_dict
+        ):
             continue
         x1, y1, x2, y2 = rect_to_bbox(annotation_dict["Rect"], height)
         uri_dict = try_resolve(annotation_dict["A"])