Repository navigation
Closed
Labels
bugSomething isn't workingSomething isn't workingcomponent: document-decodersize: small<= 2d human effort<= 2d human effort
Description
2.4.29. Suddenly started appearing on a standard test doc?
2026-05-26 11:23:58,061 - document-decoder - ERROR - Message processing exception: module 'matplotlib.ft2font' has no attribute '__path__'
Traceback (most recent call last):
File "/usr/local/lib/python3.13/site-packages/trustgraph/base/consumer.py", line 239, in handle_one_from_queue
await self.handler(msg, self, self.flow)
File "/usr/local/lib/python3.13/site-packages/trustgraph/decoding/universal/processor.py", line 421, in on_message
elements = self.extract_elements(blob, mime_type)
File "/usr/local/lib/python3.13/site-packages/trustgraph/decoding/universal/processor.py", line 200, in extract_elements
elements = partition(**kwargs)
File "/usr/local/lib/python3.13/site-packages/unstructured/partition/auto.py", line 215, in partition
partition_pdf = partitioner_loader.get(file_type)
File "/usr/local/lib/python3.13/site-packages/unstructured/partition/auto.py", line 371, in get
self._partitioners[file_type] = self._load_partitioner(file_type)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^
File "/usr/local/lib/python3.13/site-packages/unstructured/partition/auto.py", line 389, in _load_partitioner
partitioner_module = importlib.import_module(file_type.partitioner_module_qname)
File "/usr/lib64/python3.13/importlib/__init__.py", line 88, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1395, in _gcd_import
File "<frozen importlib._bootstrap>", line 1360, in _find_and_load
File "<frozen importlib._bootstrap>", line 1331, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 935, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 1023, in exec_module
File "<frozen importlib._bootstrap>", line 488, in _call_with_frames_removed
File "/usr/local/lib/python3.13/site-packages/unstructured/partition/pdf.py", line 55, in <module>
from unstructured.partition.pdf_image.pdfminer_processing import (
...<4 lines>...
)
File "/usr/local/lib/python3.13/site-packages/unstructured/partition/pdf_image/pdfminer_processing.py", line 17, in <module>
from unstructured.partition.pdf_image.pdf_image_utils import remove_control_characters
File "/usr/local/lib/python3.13/site-packages/unstructured/partition/pdf_image/pdf_image_utils.py", line 17, in <module>
from unstructured_inference.inference.layout import convert_pdf_to_image as render_pdf_to_image
File "/usr/local/lib/python3.13/site-packages/unstructured_inference/inference/layout.py", line 23, in <module>
from unstructured_inference.visualize import draw_bbox
File "/usr/local/lib/python3.13/site-packages/unstructured_inference/visualize.py", line 7, in <module>
import matplotlib.pyplot as plt
File "/usr/local/lib64/python3.13/site-packages/matplotlib/pyplot.py", line 57, in <module>
import matplotlib.image
File "/usr/local/lib64/python3.13/site-packages/matplotlib/image.py", line 25, in <module>
from matplotlib.backend_bases import FigureCanvasBase
File "/usr/local/lib64/python3.13/site-packages/matplotlib/backend_bases.py", line 49, in <module>
from matplotlib import (
_api, backend_tools as tools, cbook, colors, _docstring, text,
_tight_bbox, transforms, widgets, is_interactive, rcParams)
File "/usr/local/lib64/python3.13/site-packages/matplotlib/text.py", line 18, in <module>
from .textpath import TextPath, TextToPath # noqa # Logically located here
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib64/python3.13/site-packages/matplotlib/textpath.py", line 7, in <module>
from matplotlib import _text_helpers, dviread
File "/usr/local/lib64/python3.13/site-packages/matplotlib/_text_helpers.py", line 10, in <module>
from .ft2font import FT2Font, Kerning, LoadFlags
RuntimeError: module 'matplotlib.ft2font' has no attribute '__path__'
Metadata
Metadata
Assignees
Labels
bugSomething isn't workingSomething isn't workingcomponent: document-decodersize: small<= 2d human effort<= 2d human effort
Activity
cybermaggedon commentedon May 26, 2026
Turns out the cause was an HTML doc fed in as a PDF doc. The test data doc was removed and so URL just returns an HTML error page
cybermaggedon commentedon May 26, 2026
Maybe the real problem is that some validation should have taken place.
jmolz commentedon Jun 7, 2026
Opened PR #977 to handle this: #977
The patch validates decoded PDF bytes before invoking the PDF loader, so HTML error pages or other non-PDF content are ignored without emitting page documents. Added unit coverage for librarian-backed content that is labeled as a PDF but contains HTML.