getobj can raise PDFObjectNotFound

pdfparanoia/plugins/jstor.py

@@ -10,6 +10,8 @@ from ..eraser import (
 )
 from ..plugin import Plugin
 
+from pdfminer.pdftypes import PDFObjectNotFound
+
 class JSTOR(Plugin):
     """
     JSTOR
@@ -50,6 +52,7 @@ class JSTOR(Plugin):
         # check each object in the pdf
         for objid in objids:
             # get an object by id
+            try:
                 obj = pdf.getobj(objid)
 
                 if hasattr(obj, "attrs"):
@@ -92,6 +95,8 @@ class JSTOR(Plugin):
                             replacements.append([objid, better_content])
 
                             page_id += 1
+            except PDFObjectNotFound, e:
+                print >>sys.stderr, 'Missing object: %r' % e
 
         if verbose >= 1 and replacements:
             sys.stderr.write("%s: Found objects %s with %r; omitting..." % (cls.__name__, [deets[0] for deets in replacements], cls.requirements))