Skip to content

Commit 0c9c5eb

Browse files
committed
Integrate the version unstructured-inference 0.8.7
1 parent 6cede68 commit 0c9c5eb

File tree

1 file changed

+0
-18
lines changed

1 file changed

+0
-18
lines changed

unstructured/partition/pdf.py

Lines changed: 0 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -1171,24 +1171,6 @@ def document_to_element_list(
11711171
page_elements.extend(element)
11721172
translation_mapping.extend([(layout_element, el) for el in element])
11731173
continue
1174-
1175-
# TODO(klaijan) - isalnum() only works with A-Z, a-z and 0-9
1176-
# will need to switch to some pattern matching once we support more languages
1177-
if not word:
1178-
isalnum = char.isalnum()
1179-
if word and char.isalnum() != isalnum:
1180-
isalnum = char.isalnum()
1181-
words.append(
1182-
{"text": word, "bbox": (x1, y1, x2, y2), "start_index": start_index},
1183-
)
1184-
word = ""
1185-
1186-
if len(word) == 0:
1187-
text_len + index
1188-
character.x0
1189-
height - character.y0
1190-
character.x1
1191-
height - character.y1
11921174
else:
11931175

11941176
element.metadata.links = (

0 commit comments

Comments
 (0)