opaf PDF stat maker

####################################################################
## felipe.andres.manzano@gmail.com  http://feliam.wordpress.com/  ##
## twitter.com/feliam        http://www.linkedin.com/in/fmanzano  ##
####################################################################
'''
= Opaf! It's an Open PDF Analysis Framework! =

python stats.py file1.pdf .... fileN.pdf

Total number of parsed files: 1
Total number of parsed bytes: 687662 [avg:687662.0]
Total number of indirect objects: 788 [avg:788.0]
Total number of streams: 89 [avg:89.0]
Total number of filtered streams: 65 [avg:65.0]
Total number of compressed objects: 684 [avg:684.0]
Object Filter frequencies: {'FlateDecode': 58, 'DCTDecode': 7}
Object Type frequencies: {'XObject': 8, 'XRef': 1, 'Group': 21, 'Encoding': 7, 'Pages': 1, 'Mask': 8, 'Catalog': 1, 'StructTreeRoot': 1, 'ExtGState': 13, 'FontDescriptor': 11, 'ObjStm': 7, 'Font': 15, 'Page': 6, 'Metadata': 17}


'''
import sys
from opaflib import *

if __name__ == '__main__':
    types = {}
    filters = {}
    bytes, iobjects, streams, fstreams, cobjects = [], [], [], [], []

    for filename in sys.argv[1:]:
        try:
            #read the pdf file
            pdf = file(filename,'rb').read()

            #parse it to xml
            xml_pdf = normalParser(pdf)
    #        xml_pdf = bruteParser(pdf)

            #find, expand and parse every ObjStm
            for objstm in xml_pdf.xpath(
     '//*[starts-with(local-name(),"indirect_object")]/dictionary/dictionary_entry/name[@payload=enc("Type")]/following-sibling::*[position()=1 and @payload=enc("ObjStm")]/../../..'):
                expand(objstm)
                expandObjStm(objstm)

            #count /Filters
            for xml_fi in xml_pdf.xpath('//indirect_object_stream/dictionary/dictionary_entry/name[@payload=enc("Filter")]/../*[position()=2]'):
                if xml_fi.tag == 'array':
                    fis = [payload(x) for x in xml_fi]
                elif xml_fi.tag == 'name':
                    fis = [payload(xml_fi)]
                else:
                    fis = []
                    print "Error parsing filter name at %s in file %s"%(filename,xml_fi.get('lexstart'))
                for fi in fis:
                    filters[fi] = filters.get(fi,0)+1

            #Count Object Types
            for ty in [payload(x) for x in xml_pdf.xpath('//*[starts-with(local-name(),"indirect_object")]/dictionary/dictionary_entry/name[@payload=enc("Type")]/following-sibling::*[1]')]:
                types[ty] = types.get(ty,0)+1

            #Other random counters...
            iobjects.append(len(xml_pdf.xpath('//indirect_object')))
            streams.append(len(xml_pdf.xpath('//indirect_object_stream')))
            fstreams.append(len(xml_pdf.xpath('//indirect_object_stream/dictionary/dictionary_entry/name[@payload=enc("Filter")]/../../..')))
            cobjects.append(len(xml_pdf.xpath('//indirect_object_stream//indirect_object')))
            bytes.append(len(pdf))
            print sum(bytes)
            if sum(bytes) >100000000:
                break
        except Exception,e:
            print e
            print ("Sorry couldn't parse file %s with sequential parser."%filename)
    def stats(L):
        total = sum(L)
        average = float(sum(L)) / len(L)
        return total,average

    #Print statistics to stdout.
    print "Total number of parsed files: %s"%len(bytes)

    if len(bytes) != 0 :
        print "Total number of parsed bytes: %s [avg:%s]"%stats(bytes)
        #print ("Total time: %s [avg:%s]"%stats(times))

        print "Total number of indirect objects: %s [avg:%s]"%stats(iobjects)
        print "Total number of streams: %s [avg:%s]"%stats(streams)
        print "Total number of filtered streams: %s [avg:%s]"%stats(fstreams)
        print "Total number of compressed objects: %s [avg:%s]"%stats(cobjects)

        print "Object Filter frequencies: %s"%repr(filters)
        print "Object Type frequencies: %s"%repr(types)

    print