Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

What is this (cid:51) in the output of pdf2txt?

So i'm trying to extract the text from a pdf file, I need its position, width, height, font.

I have tried many, but the most useful and complete solution looks to be PDFMiner, and in this case, more exactly pdf2txt.py.

I have followed the doc and the examples and tried to extract the text Learn More from my pdf using this command:

pdf2txt.py -Y normal -t xml -o buttons.xml buttons.pdf

And the output buttons.xml looks like that:

<?xml version="1.0" encoding="utf-8" ?>
  <pages>
      <page id="1" bbox="0.000,0.000,799.900,449.944" rotate="0">
      <textbox id="0" bbox="164.979,213.240,247.680,235.944">
          <textline bbox="164.979,213.240,247.680,235.944">
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="164.979,213.240,178.978,235.944" size="22.704">(cid:51)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="173.280,213.240,187.278,235.944" size="22.704">(cid:76)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="181.315,213.240,195.313,235.944" size="22.704">(cid:72)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="189.350,213.240,203.348,235.944" size="22.704">(cid:89)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="194.795,213.240,208.793,235.944" size="22.704">(cid:85)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="203.096,213.240,217.094,235.944" size="22.704">(cid:3)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="206.987,213.240,220.986,235.944" size="22.704">(cid:52)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="219.684,213.240,233.682,235.944" size="22.704">(cid:86)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="228.237,213.240,242.235,235.944" size="22.704">(cid:89)</text>
              <text font="KZNUUP+HelveticaNeue-Bold" bbox="233.682,213.240,247.680,235.944" size="22.704">(cid:76)</text>
              <text></text>
          </textline>
          </textbox>
          <textgroup bbox="164.979,213.240,419.659,235.944">
              <textbox id="0" bbox="164.979,213.240,247.680,235.944" />
          </textgroup>
      </page>
  </pages>

The first character should be a L and 51 (cid:51) doesn't seem to match any of the character i have in my sentence, regarding the ascii table and the utf-8 table

So as the title says, I wonder what is it, and how to use these (cid:51)... ?


EDIT

So I found that instead of getting the real character the program write (cid:%d) because he doesn't recognize that it's a unicode string.

It first call this function to write the char:

def render_char(self, matrix, font, fontsize, scaling, rise, cid):
    try:
        text = font.to_unichr(cid)
        assert isinstance(text, unicode), text
    except PDFUnicodeNotDefined:
        text = self.handle_undefined_char(font, cid)

But the assert fail and fire the event PDFUnicodeNotDefined which is caught and calls:

def handle_undefined_char(self, font, cid):
    if self.debug:
        print >>sys.stderr, 'undefined: %r, %r' % (font, cid)
    return '(cid:%d)' % cid

And that's how I end with a file containing all these (cid:%d).

I'm fairly new to python and I try to figure out a way to recognize these chars, it should be one no ? Does anyone has any idea ?

like image 808
Micka Avatar asked May 13 '13 13:05

Micka


1 Answers

to understand how to interpret the cid you need to know a pair of things:

  1. The Registry-Ordering-Supplement (ROS) information for the font in question. It's usually something like 'Adobe-Japan1-5' and is an informational property stored in the font. The ROS determines how the CIDs are to be interpreted.

  2. Armed with the ROS info, select a compatible CMap and decode through that.You can find CMap files for the Adobe-defined ROSes at http://sourceforge.net/projects/cmap.adobe/files/

More information on CID and CMaps direct from the inventors is available at http://www.adobe.com/content/dam/Adobe/en/devnet/font/pdfs/5014.CIDFont_Spec.pdf

check decode CID font codes to equivalent ASCII characters for more information

like image 163
Andre Avatar answered Sep 30 '22 19:09

Andre