fix regex to filter out *leading and trailing* punctation marks from …

…ocr text
LibraryOfCongress · Oct 31, 2013 · 8431f85 · 8431f85
1 parent 4266c99
commit 8431f85
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/core/ocr_extractor.py b/core/ocr_extractor.py
@@ -3,7 +3,7 @@
 from xml.sax.handler import ContentHandler, feature_namespaces
 from xml.sax import make_parser
 
-trailing_punctuation = re.compile('''[^a-zA-Z0-9]+$''')
+trailing_punctuation = re.compile('''^[^a-zA-Z0-9]+|[^a-zA-Z0-9]+$''')
 
 class OCRHandler(ContentHandler):