Fix for issue Unstructured-IO#3815

Removed redundent patch as it is already merged in pdfminer/pdfminer.six#885
PhorstenkampFuzzy · Dec 17, 2024 · 19509ed · 19509ed
1 parent 9a9bf4c
commit 19509ed
Show file tree

Hide file tree

Showing 3 changed files with 0 additions and 30 deletions.
diff --git a/unstructured/partition/pdf.py b/unstructured/partition/pdf.py
@@ -11,7 +11,6 @@
 
 import numpy as np
 import wrapt
-from pdfminer import psparser
 from pdfminer.layout import LTContainer, LTImage, LTItem, LTTextBox
 from pdfminer.utils import open_filename
 from pi_heif import register_heif_opener
@@ -96,16 +95,11 @@
     PartitionStrategy,
 )
 from unstructured.partition.utils.sorting import coord_has_valid_points, sort_page_elements
-from unstructured.patches.pdfminer import parse_keyword
 from unstructured.utils import first, requires_dependencies
 
 if TYPE_CHECKING:
     pass
 
-# NOTE(alan): Patching this to fix a bug in pdfminer.six. Submitted this PR into pdfminer.six to fix
-# the bug: https://github.com/pdfminer/pdfminer.six/pull/885
-psparser.PSBaseParser._parse_keyword = parse_keyword  # type: ignore
-
 RE_MULTISPACE_INCLUDING_NEWLINES = re.compile(pattern=r"\s+", flags=re.DOTALL)
 
 

diff --git a/unstructured/patches/__init__.py b/unstructured/patches/__init__.py
diff --git a/unstructured/patches/pdfminer.py b/unstructured/patches/pdfminer.py