CODEWITHSUNDEEP
pythonpdfjupyter-notebook
Curious2learn
Curious2learn

Reputation: 33598

View pdf image in an iPython Notebook

The following code allows me to view a png image in an iPython notebook. Is there a way to view pdf image? I don't need to use IPython.display necessarily. I am looking for a way to print a pdf image in a file to the iPython notebook output cell.

## This is for an `png` image
from IPython.display import Image

fig = Image(filename=('./temp/my_plot.png'))
fig

Thank you.

Upvotes: 54

Views: 87515

Answers (5)

T-Dog
T-Dog

Reputation: 153

I think there is a better way to do this, since

  • with wand you need to install the MagickWand library
  • with IFrame you are not flexible with the directories

I would do it like this:

    from PIL import Image
    import matplotlib.pyplot as plt
    import fitz

    your_path = # YOUR PATH
    doc = fitz.open(your_path)
    MAX_PAGES = 1

    zoom = 2 # to increase the resolution
    mat = fitz.Matrix(zoom, zoom)

    for i, page in enumerate(doc):
        pix = page.get_pixmap(matrix=mat)
        img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)

        # display images
        plt.figure(figsize=(7,7), facecolor="w")
        plt.xticks(color="white")
        plt.yticks(color="white")
        plt.tick_params(bottom = False)
        plt.tick_params(left = False)

        plt.imshow(img)

        if i > MAX_PAGES - 1:
            break

Upvotes: 1

Rve
Rve

Reputation: 410

In addition to Jakob's excellent answer recommending the Wand bindings for ImageMagick:

If your PDF contains vector graphics, use the resolution keyword to control the size of the rendered image. ImageMagick's default value is 72 dpi. Higher values yield more pixels.

from wand.image import Image as WImage
img = WImage(filename='hat.pdf', resolution=100) # bigger
img

Upvotes: 5

Levon
Levon

Reputation: 11914

To show pdf-s inside ipython/jupyter notebooks you can use IFrame

from IPython.display import IFrame
IFrame("./samples/simple3.pdf", width=600, height=300)

Here is the screenshot

pdf preview in ipython/jupyter notebook

Upvotes: 82

Jeremy Ellis
Jeremy Ellis

Reputation: 41

Assuming a multi-image pdf called Rplots.pdf

The following works in the jupyter notebook cell. For installation I used

pip install Wand

This code pastes into a cell

from wand.image import Image  

imageFromPdf = Image(filename='Rplots.pdf')  
pages = len(imageFromPdf.sequence)  

image = Image(  
  width=imageFromPdf.width,  
  height=imageFromPdf.height * pages  
)  
for i in range(pages):  
  image.composite(  
  imageFromPdf.sequence[i],  
  top=imageFromPdf.height * i,  
  left=0  
)  
image.format="png"  
image

Upvotes: 4

Jakob
Jakob

Reputation: 20801

The problem you (and others) face is that PDFs cannot be displayed directly in the browser. The only possible way to get something similar is to use an image-converter to create a PNG or JPG out of the PDF and display this one.
This could be done via imagemagick and a custom display function.

Update 1

A simple solution is to use wand (http://docs.wand-py.org) a python-imagemagick binding. I tried with Ubuntu 13.04:

wand session in ipython

In text form:

from wand.image import Image as WImage
img = WImage(filename='hat.pdf')
img

For a multi-page pdf, you can get e.g. the second page via:

img = WImage(filename='hat.pdf[1]')

Update 2

As recent browsers support to display pdfs with their embedded pdf viewer a possible alternative solution based on an iframe can be implemented as

class PDF(object):
  def __init__(self, pdf, size=(200,200)):
    self.pdf = pdf
    self.size = size

  def _repr_html_(self):
    return '<iframe src={0} width={1[0]} height={1[1]}></iframe>'.format(self.pdf, self.size)

  def _repr_latex_(self):
    return r'\includegraphics[width=1.0\textwidth]{{{0}}}'.format(self.pdf)

This class implements html and latex representations, hence the pdf will also survive a nbconversion to latex. It can be used like

PDF('hat.pdf',size=(300,250))

With Firefox 33 this results in enter image description here

Upvotes: 88

Related Questions