python3 - TypeError: ord() expected string of length 1, but int found

[LeoFCardoso]

I am getting this error when using python3 and this simple code:

imagepdf = PdfFileReader(open(sys.argv[1], 'rb'), strict=False)
textpdf = PdfFileReader(open(sys.argv[2], 'rb'), strict=False)
for i in range(imagepdf.getNumPages()):
imagepage = imagepdf.getPage(i)
textpage = textpdf.getPage(i)
factor_x = textpage.mediaBox.upperRight[0] / imagepage.mediaBox.upperRight[0]
factor_y = textpage.mediaBox.upperRight[1] / imagepage.mediaBox.upperRight[1]
imagepage.scale(float(factor_x), float(factor_y))
textpage.mergePage(imagepage) # imagepage stay on top
textpage.compressContentStreams()
output.addPage(textpage)

Trace:

Traceback (most recent call last):
File "...", line 34, in
imagepage.scale(float(factor_x), float(factor_y))
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/pdf.py", line 2493, in scale
0, 0])
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/pdf.py", line 2479, in addTransformation
originalContent, self.pdf, ctm)
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/pdf.py", line 2180, in _addTransformationMatrix
contents = ContentStream(contents, pdf)
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/pdf.py", line 2641, in init
data += s.getObject().getData()
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/generic.py", line 837, in getData
decoded._data = filters.decodeStreamData(self)
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/filters.py", line 350, in decodeStreamData
data = LZWDecode.decode(data, stream.get("/DecodeParms"))
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/filters.py", line 255, in decode
return LZWDecode.decoder(data).decode()
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/filters.py", line 228, in decode
cW = self.nextCode();
File "/opt/local/Library/Frameworks/Python.framework/Versions/3.4/lib/python3.4/site-packages/PyPDF2/filters.py", line 205, in nextCode
nextbits=ord(self.data[self.bytepos])
TypeError: ord() expected string of length 1, but int found

Am I doing something wrong?

[mstamy2]

Any chance you can post/send me the PDF(s) you're working with?

Most likely this is a Python 3 type handling issue found in the LZW decoding algorithm, in which case it is easily fixable

[LeoFCardoso]

My script is
https://github.com/LeoFCardoso/pdf2pdfocr/blob/master/pdf2pdfocr_multibackground.py

Here is the call:

python3.4 pdf2pdfocr_multibackground.py first.pdf second.pdf result.pdf

second.pdf
first.pdf

Thanks!

[mstamy2]

5bbd5af should take care of these type issues

[mstamy2]

Let me know of any further issues!

[LeoFCardoso]

It works! Thanks!

[LeoFCardoso]

when new version will be available at https://pypi.python.org/pypi/PyPDF2?
Thanks!

[jguram]

199709222.pdf
I am getting the same issue.Can you please help me. @mstamy2

Here is my code

import PyPDF2
import os
import xlsxwriter

search_words = []
os.chdir(r'C:\Users\jui\Documents\Parabole')
with open('allwords.txt') as f:
for line in f:
search_words.append(line)
print(len(search_words))

p1 = 'C:\data\pdfContainer\test'
file_name = []
for filename in os.listdir(p1):
file_name.append(filename.lstrip())
print(file_name)

for i in file_name:
print(i)
os.chdir(r'C:\data\pdfContainer\test')
pdf_file = open(filename, 'rb')
print('opening pdf file')
read_pdf = PyPDF2.PdfFileReader(pdf_file)
print(read_pdf.isEncrypted)

number_of_pages = read_pdf.getNumPages()
print(number_of_pages)
pdf_content = ' '
for j in range(number_of_pages):
    page = read_pdf.getPage(j)
    print(page)
    j = +1
    page_content = page.extractText()
    print(page_content)
    pdf_content = pdf_content + page_content
#print(pdf_content)
new_dict = {}
for word in search_words:
    cnt_of_words = pdf_content.count(word)
    new_dict.update({word: cnt_of_words})
#print(new_dict)    
#for i in file_name:
print(i)
p = i + '.xlsx'
print(p)
workbook = xlsxwriter.Workbook(p)
worksheet = workbook.add_worksheet()
row = 0
col = 0
for key in new_dict.keys():
    row += 1
    worksheet.write(row, col, key)
    worksheet.write(row, col + 1, new_dict[key])
    row += 1

workbook.close()

allwords.txt

[jguram]

@mstamy2 I want to check the occurence of words present in allwords.txt in the PDF file mentioned and write it in excel

[rameessahlu]

5bbd5af should take care of these type issues

These fixes are still not merged.

[Jeff-Winchell]

I got multiple files that trigger this error.
I am currently running a loop looking at 177,000 pdf files. It blows up with that error on the extractText() function. Yes I am running python 3, which was shipped A DECADE AGO.

for index, row in Files.iterrows():
    try:
        filename=row.Filename
        pdffile=PdfFileReader(filename)
        for pagenum in range(pdffile.numPages):
            foo=[word.lower() for word in tokenizer.tokenize(pdffile.getPage(pagenum).extractText()) if word.lower() not in stopWords and not word.isdigit()]
    except:
        print(index,filename,pagenum)

The bug has been triggered 6 times in the first 1255 files, so I'm guessing the error rate is about 0.5%

[misilot]

Can a release be made that includes this fix?