Mercurial > hg > python

annotate pdfComments.py @ 17:6c9e371b0325

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
doesn't break, what's there is good, but stuck part-way in 10a
author Henry S. Thompson <ht@inf.ed.ac.uk>
date Wed, 25 Mar 2020 19:21:38 +0000
parents e07789816ca5
children
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
rev   line source
2
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
1 import PyPDF2 as pyPdf, sys
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
2
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
3 if sys.argv[1]=='-v':
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
4 verbose=True
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
5 sys.argv.pop(1)
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
6 else:
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
7 verbose=False
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
8
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
9 f = open(sys.argv[1],'rb')
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
10
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
11 pdf = pyPdf.PdfFileReader(f)
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
12 pgs = pdf.getNumPages()
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
13 key = '/Annots'
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
14 uri = '/URI'
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
15 ank = '/A'
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
16
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
17 #print pdf.getNamedDestinations()
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
18
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
19 for pg in range(pgs):
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
20 print '#',pg
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
21 p = pdf.getPage(pg)
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
22 o = p.getObject()
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
23 #print o.keys()
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
24 if o.has_key(key):
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
25 ann = o[key]
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
26 #print key,ann
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
27 for a in ann:
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
28 u = a.getObject()
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
29 if '/Contents' in u:
e07789816ca5 adding more python files from lib/python on origen
Henry Thompson <ht@markup.co.uk>
parents:
diff changeset
30 print "%s: %s"%(u['/Subtype'],u['/Contents'])