Creating PDFs with django (wkhtmltopdf)

Question

Could someone please provide me with a comprehensive example of how to get a view in django to return a PDF using wkhtmltopdf. There are limited number of examples that come with django-wkhtmltopdf and they presume a level of knowledge I just don't have. I have looked through the source code but I can't make heads or tails of how to use it (for example whats the difference between PDFTemplateView and PDFTemplateResponse?!?)

I would be very grateful for any help.

BTW(I'm using templates for the header and footer as well)

EDIT

def some_view(request,sID):
    something = get_object_or_404(Something,id=sID)
    return render_to_response('something.html', {'something':something}, context_instance=RequestContext(request))

How would I get the following simple view to provide me with a pdf instead of an html file?

EDIT 2

I am currently playing around with:

def pdf_view(request,sID):
    template = 'pdf.html'
    something = get_object_or_404(Something,id=sID)
    context = {
        'something' : Something,
        'object_for_header_and_footer': something.object_for_header_and_footer,
    }
    cmd_options = settings.WKHTMLTOPDF_CMD_OPTIONS

    return PDFTemplateResponse(request=request,
        context=context,
        template=template,
        filename='something',
        header_template='header.html',
        footer_template='footer.html',
        cmd_options=cmd_options)

but I am getting 'str' object has no attribute 'update' in /usr/local/lib/python2.7/dist-packages/wkhtmltopdf/utils.py in wkhtmltopdf, line 74. I don't know whether to starting hacking wkhtmltopdf?!?!

The normal view and PDFTemplateResponse is working for me. I didn't specify the WKHTMLTOPDF_CMD_OPTIONS. — allcaps, Aug 19 '13 at 11:41

allcaps · Accepted Answer · 2013-08-19T14:35:34.487

The difference between PDFTemplateView and PDFTemplateResponse is that the view is a class-based view. And PDFTemplateResponse renders the pdf data and sets the right response headers. To add header and footer:

# urls.py

from django.conf.urls.defaults import *
from wkhtmltopdf.views import PDFTemplateView


    urlpatterns = patterns('',
        ...
        url(r'^pdf/$', PDFTemplateView.as_view(template_name='my_template.html',
                filename='my_pdf.pdf', 
                header_template='my_header_template.html', 
                footer_template='my_footer_template.html', 
                ...
                ), name='pdf'),
    )

Opening pdf/ in your browser will start a download of my_pdf.pdf based on the my_template.html, my_header_template.html and my_footer_template.html.

The advanced example shows how to subclass PDFTemplateView extending and changing the logic of PDFTemplateView. To understand what happens read Using class based views.

Like header_template and footer_template you can define a response_class. Because PDFTemplateResponse is the default, you don't have to define it.

EDIT

The following simple view provides you with a pdf instead of an html. This is not using django-wkhtmltopdf. You could use wkhtmltopdf in your html2pdf function.

def some_view(request):
    t = loader.get_template('myapp/template.html')
    c = RequestContext(request, {'foo': 'bar'})
    html = t.render(c)
    pdf_data = html2pdf(html) # Your favorite html2pdf generator
    response = HttpResponse(pdf_data, content_type='application/pdf')
    response['Content-Disposition'] = 'attachment; filename="some_filename.pdf"'
    return response

EDIT 2

A simple view with context:

template.html

<!DOCTYPE html>
<html>
<head>
    <meta charset="utf-8" />
    <title>Untitled</title>
</head>
<body>
    <h1>{{ title }}</h1>
</body>
</html>

urls.py

from views import MyPDFView

urlpatterns = patterns('',
    (r'^pdf/', MyPDFView.as_view()),
)

views.py

from django.views.generic.base import View
from wkhtmltopdf.views import PDFTemplateResponse

class MyPDFView(View):
    template='template.html'
    context= {'title': 'Hello World!'}

    def get(self, request):
        response = PDFTemplateResponse(request=request,
                                       template=self.template,
                                       filename="hello.pdf",
                                       context= self.context,
                                       show_content_in_browser=False,
                                       cmd_options={'margin-top': 50,},
                                       )
        return response

EDIT 3

If you use a DetailView, you can add the object to context:

url(r'^books/(?P<pk>\d+)/$', MyPDFView.as_view(), name='book-detail'),


class MyPDFView(DetailView):
    template='pdftestapp/template.html'    
    context= {'title': 'Hello World!'}
    model = Book

    def get(self, request, *args, **kwargs):        
        self.context['book'] = self.get_object()

        response=PDFTemplateResponse(request=request,
                                     template=self.template,
                                     filename ="hello.pdf",
                                     context=self.context,
                                     show_content_in_browser=False,
                                     cmd_options={'margin-top': 50,}
                                     )
        return response

That's fine for rendering static `.html` (I presume) but what happens if you want `def something.views.something_view(request,someID)` to render a `pdf` rather then `html`? — Robert Johnstone, Aug 14 '13 at 21:20
(& on a side note why are static `*.html` files being passed to attributes with `template` in them. It doesn't make sense?!? surely if an attribute has `template` in it it is asking for a `view`. Isn't it?!?) — Robert Johnstone, Aug 14 '13 at 21:22
I agree, passing a view to render a pdf would be nice. You could open a issue on their GitHub page. — allcaps, Aug 15 '13 at 11:10
How do I then pass an id so I can get an object inside the `get` function? (i.e. so I can use `{% url something.views.something_view something.id %}` — Robert Johnstone, Aug 19 '13 at 11:54
And yes, you can just put {% url ... %} in your templates. It worked for me. — allcaps, Aug 19 '13 at 12:48
got it working!!! Now just need to get the right version of `wkhtmltopdf` (can't do headers and footers with my version :( — Robert Johnstone, Aug 19 '13 at 12:53

score 2 · Answer 2 · answered Aug 19 '13 at 11:26

Hmm the error indicates that you're passing string somewhere that you shouldn't.

After checking its source code, I guess in settings.py you have WKHTMLTOPDF_CMD_OPTIONS as a string, something like

WKHTMLTOPDF_CMD_OPTIONS = 'some_option_here'

But you should assign a dict there:

WKHTMLTOPDF_CMD_OPTIONS = {
    'quiet': True,
}

Otherwise your code should work fine.

Creating PDFs with django (wkhtmltopdf)

2 Answers2

Linked