Adding images to PDFs is not much different than adding text. Here is how to add an image as a flowable:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

#!/usr/bin/env python

import os
import urllib2
from reportlab.lib.pagesizes import letter
from reportlab.platypus import SimpleDocTemplate, Image

filename = './python-logo.png'

def get_python_image():
    """ Get a python logo image for this example """
    if not os.path.exists(filename):
        response = urllib2.urlopen(
            'http://www.python.org/community/logos/python-logo.png')
        f = open(filename, 'w')
        f.write(response.read())
        f.close()

get_python_image()

doc = SimpleDocTemplate("image.pdf", pagesize=letter)
parts = []
parts.append(Image(filename))
doc.build(parts)

Ignore the get_python_image function. It is in there only to make this example easily runnable. As you see, I import the Image Flowable from platypus. Image takes a minimum of one argument, which is the path to the image to use. You can build it into a SimpleDocTemplate just like a Paragraph or a Spacer.

Writing images to a canvas is easy, but has one gotcha.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

#!/usr/bin/env python

import os
import urllib2
from reportlab.lib.pagesizes import letter
from reportlab.lib.units import inch
from reportlab.pdfgen import canvas

filename = './python-logo.png'

def get_python_image():
    """ Get a python logo image for this example """
    if not os.path.exists(filename):
        response = urllib2.urlopen(
            'http://www.python.org/community/logos/python-logo.png')
        f = open(filename, 'w')
        f.write(response.read())
        f.close()

get_python_image()

c = canvas.Canvas('imageabs.pdf', pagesize=letter)
width, height = letter
c.drawImage(filename, inch, height - 2 * inch) # Who needs consistency?
c.showPage()
c.save()

We use the drawImage method of our Canvas instance. Here is the gotcha. Unlike drawString and its siblings, drawImage takes its content, the path to the image, first and its coordinates second. This is poor design by the ReportLab developers. It is still easy to add the image though. Just remember that if you get an obtuse error like this:

AttributeError: 'float' object has no attribute 'jpeg_fh'

Or this:

AttributeError: 'int' object has no attribute 'jpeg_fh'

Then the arguments are out of order.

When making PDFs in python, we use reportlab for the most part. It is an extensive module that can do make about anything you would want in a PDF. Today, I will cover how to do the most basic function, putting some text somewhere on the page. Here is the simplest example:

1
2
3
4
5
6
7

#!/usr/bin/env python
from reportlab.pdfgen import canvas

c = canvas.Canvas('rldemo1.pdf')
c.drawString(100, 100, 'Hello, world!')
c.showPage()
c.save()

You will notice that reportlab takes care of opening the file when we create a new Canvas instance. The drawString method puts a piece of text in an interesting place. It starts 100 points from the left as expected, but also 100 points from the bottom. I expected the origin to be the top-left, but reportlab starts from the bottom-left. With showPage, we commit our changes and save will write the file to disc. If you run this, you will also notice that the page size is A4 by default, which is because ReportLab, the company, is based in the UK. So how do we change that and start from the top instead of the bottom in reportlab?

1
2
3
4
5
6
7
8
9

#!/usr/bin/env python
from reportlab.lib.pagesizes import letter
from reportlab.pdfgen import canvas

c = canvas.Canvas('rldemo2.pdf', pagesize=letter)
width, height = letter
c.drawString(100, height - 100, 'From the top!')
c.showPage()
c.save()

So here we supply Canvas with the letter pagesize. Just that simple. Now that we have the pagesize, which is but a tuple, we can extract the width and height. With the page height, we can subtract the offset from the top to have the text placed relative to the top of the page. Can I use something other than points? Real paper layouts are done in inches and centimeters.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13

#!/usr/bin/env python
from reportlab.lib.pagesizes import letter
from reportlab.lib.units import inch, cm
from reportlab.pdfgen import canvas

c = canvas.Canvas('rldemo3.pdf', pagesize=letter)
width, height = letter
c.drawString(inch, height - inch, '1 inch')
c.drawString(inch, height - 2 * inch, '2 inches')
c.drawString(cm, cm, '1 cm')
c.drawString(cm, 2 * cm, '2 cm')
c.showPage()
c.save()

Our reportlab module features a variety of constants for unit size, which are multiples of points. One gotcha with all of the text placement. The text is always drawn top and right of the coordinates given with drawString, which you can see if you run this example. You can write text from the left now, but what about centered text and right-aligned text?

1
2
3
4
5
6
7
8
9

#!/usr/bin/env python
from reportlab.lib.pagesizes import letter
from reportlab.lib.units import inch
from reportlab.pdfgen import canvas

c = canvas.Canvas('rldemo4.pdf', pagesize=letter)
width, height = letter
c.drawString(inch, height - inch, 'Left')
c.drawCentredString(width / 2.0, height - inch, 'Center') # Notice the UK

spelling :) c.drawRightString(width - inch, height - inch, 'Right') c.showPage() c.save()

There are two more methods for printing text, drawCentredString and drawRightString. The only difference from drawString is that these start drawing from the center and right. All three still go from the bottom up.