Render HTML to PDF in Django site

For my django powered site, I am looking for an easy solution to convert dynamic html pages to pdf.

Pages include HTML and charts from Google visualization API (which is javascript based, yet including those graphs is a must).

128188 次浏览
小开
最佳答案

Try the solution from Reportlab.

Download it and install it as usual with python setup.py install

You will also need to install the following modules: xhtml2pdf, html5lib, pypdf with easy_install.

Here is an usage example:

First define this function:

import cStringIO as StringIO
from xhtml2pdf import pisa
from django.template.loader import get_template
from django.template import Context
from django.http import HttpResponse
from cgi import escape




def render_to_pdf(template_src, context_dict):
template = get_template(template_src)
context = Context(context_dict)
html  = template.render(context)
result = StringIO.StringIO()


pdf = pisa.pisaDocument(StringIO.StringIO(html.encode("ISO-8859-1")), result)
if not pdf.err:
return HttpResponse(result.getvalue(), content_type='application/pdf')
return HttpResponse('We had some errors<pre>%s</pre>' % escape(html))

Then you can use it like this:

def myview(request):
#Retrieve data or whatever you need
return render_to_pdf(
'mytemplate.html',
{
'pagesize':'A4',
'mylist': results,
}
)

The template:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
<html>
<head>
<title>My Title</title>
<style type="text/css">
@page {
size: \{\{ pagesize }};
margin: 1cm;
@frame footer {
-pdf-frame-content: footerContent;
bottom: 0cm;
margin-left: 9cm;
margin-right: 9cm;
height: 1cm;
}
}
</style>
</head>
<body>
<div>
{% for item in mylist %}
RENDER MY CONTENT
{% endfor %}
</div>
<div id="footerContent">
{%block page_foot%}
Page <pdf:pagenumber>
{%endblock%}
</div>
</body>
</html>

Hope it helps.

小开

I just whipped this up for CBV. Not used in production but generates a PDF for me. Probably needs work for the error reporting side of things but does the trick so far.

import StringIO
from cgi import escape
from xhtml2pdf import pisa
from django.http import HttpResponse
from django.template.response import TemplateResponse
from django.views.generic import TemplateView


class PDFTemplateResponse(TemplateResponse):


def generate_pdf(self, retval):


html = self.content


result = StringIO.StringIO()
rendering = pisa.pisaDocument(StringIO.StringIO(html.encode("ISO-8859-1")), result)


if rendering.err:
return HttpResponse('We had some errors<pre>%s</pre>' % escape(html))
else:
self.content = result.getvalue()


def __init__(self, *args, **kwargs):
super(PDFTemplateResponse, self).__init__(*args, mimetype='application/pdf', **kwargs)
self.add_post_render_callback(self.generate_pdf)




class PDFTemplateView(TemplateView):
response_class = PDFTemplateResponse

Used like:

class MyPdfView(PDFTemplateView):
template_name = 'things/pdf.html'
小开

You can use iReport editor to define the layout, and publish the report in jasper reports server. After publish you can invoke the rest api to get the results.

Here is the test of the functionality:

from django.test import TestCase
from x_reports_jasper.models import JasperServerClient


"""
to try integraction with jasper server through rest
"""
class TestJasperServerClient(TestCase):


# define required objects for tests
def setUp(self):


# load the connection to remote server
try:


self.j_url = "http://127.0.0.1:8080/jasperserver"
self.j_user = "jasperadmin"
self.j_pass = "jasperadmin"


self.client = JasperServerClient.create_client(self.j_url,self.j_user,self.j_pass)


except Exception, e:
# if errors could not execute test given prerrequisites
raise


# test exception when server data is invalid
def test_login_to_invalid_address_should_raise(self):
self.assertRaises(Exception,JasperServerClient.create_client, "http://127.0.0.1:9090/jasperserver",self.j_user,self.j_pass)


# test execute existent report in server
def test_get_report(self):


r_resource_path = "/reports/<PathToPublishedReport>"
r_format = "pdf"
r_params = {'PARAM_TO_REPORT':"1",}


#resource_meta = client.load_resource_metadata( rep_resource_path )


[uuid,out_mime,out_data] = self.client.generate_report(r_resource_path,r_format,r_params)
self.assertIsNotNone(uuid)

And here is an example of the invocation implementation:

from django.db import models
import requests
import sys
from xml.etree import ElementTree
import logging


# module logger definition
logger = logging.getLogger(__name__)


# Create your models here.
class JasperServerClient(models.Manager):


def __handle_exception(self, exception_root, exception_id, exec_info ):
type, value, traceback = exec_info
raise JasperServerClientError(exception_root, exception_id), None, traceback


# 01: REPORT-METADATA
#   get resource description to generate the report
def __handle_report_metadata(self, rep_resourcepath):


l_path_base_resource = "/rest/resource"
l_path = self.j_url + l_path_base_resource
logger.info( "metadata (begin) [path=%s%s]"  %( l_path ,rep_resourcepath) )


resource_response = None
try:
resource_response = requests.get( "%s%s" %( l_path ,rep_resourcepath) , cookies = self.login_response.cookies)


except Exception, e:
self.__handle_exception(e, "REPORT_METADATA:CALL_ERROR", sys.exc_info())


resource_response_dom = None
try:
# parse to dom and set parameters
logger.debug( " - response [data=%s]"  %( resource_response.text) )
resource_response_dom = ElementTree.fromstring(resource_response.text)


datum = ""
for node in resource_response_dom.getiterator():
datum = "%s<br />%s - %s" % (datum, node.tag, node.text)
logger.debug( " - response [xml=%s]"  %( datum ) )


#
self.resource_response_payload= resource_response.text
logger.info( "metadata (end) ")
except Exception, e:
logger.error( "metadata (error) [%s]" % (e))
self.__handle_exception(e, "REPORT_METADATA:PARSE_ERROR", sys.exc_info())




# 02: REPORT-PARAMS
def __add_report_params(self, metadata_text, params ):
if(type(params) != dict):
raise TypeError("Invalid parameters to report")
else:
logger.info( "add-params (begin) []" )
#copy parameters
l_params = {}
for k,v in params.items():
l_params[k]=v
# get the payload metadata
metadata_dom = ElementTree.fromstring(metadata_text)
# add attributes to payload metadata
root = metadata_dom #('report'):


for k,v in l_params.items():
param_dom_element = ElementTree.Element('parameter')
param_dom_element.attrib["name"] = k
param_dom_element.text = v
root.append(param_dom_element)


#
metadata_modified_text =ElementTree.tostring(metadata_dom, encoding='utf8', method='xml')
logger.info( "add-params (end) [payload-xml=%s]" %( metadata_modified_text )  )
return metadata_modified_text






# 03: REPORT-REQUEST-CALL
#   call to generate the report
def __handle_report_request(self, rep_resourcepath, rep_format, rep_params):


# add parameters
self.resource_response_payload = self.__add_report_params(self.resource_response_payload,rep_params)


# send report request


l_path_base_genreport = "/rest/report"
l_path = self.j_url + l_path_base_genreport
logger.info( "report-request (begin) [path=%s%s]"  %( l_path ,rep_resourcepath) )


genreport_response = None
try:
genreport_response = requests.put( "%s%s?RUN_OUTPUT_FORMAT=%s" %(l_path,rep_resourcepath,rep_format),data=self.resource_response_payload, cookies = self.login_response.cookies )
logger.info( " - send-operation-result [value=%s]"  %( genreport_response.text) )
except Exception,e:
self.__handle_exception(e, "REPORT_REQUEST:CALL_ERROR", sys.exc_info())




# parse the uuid of the requested report
genreport_response_dom = None


try:
genreport_response_dom = ElementTree.fromstring(genreport_response.text)


for node in genreport_response_dom.findall("uuid"):
datum = "%s" % (node.text)


genreport_uuid = datum


for node in genreport_response_dom.findall("file/[@type]"):
datum = "%s" % (node.text)
genreport_mime = datum


logger.info( "report-request (end) [uuid=%s,mime=%s]"  %( genreport_uuid, genreport_mime) )


return [genreport_uuid,genreport_mime]
except Exception,e:
self.__handle_exception(e, "REPORT_REQUEST:PARSE_ERROR", sys.exc_info())


# 04: REPORT-RETRIEVE RESULTS
def __handle_report_reply(self, genreport_uuid ):




l_path_base_getresult = "/rest/report"
l_path = self.j_url + l_path_base_getresult
logger.info( "report-reply (begin) [uuid=%s,path=%s]"  %( genreport_uuid,l_path) )


getresult_response = requests.get( "%s%s/%s?file=report" %(self.j_url,l_path_base_getresult,genreport_uuid),data=self.resource_response_payload, cookies = self.login_response.cookies )
l_result_header_mime =getresult_response.headers['Content-Type']


logger.info( "report-reply (end) [uuid=%s,mime=%s]"  %( genreport_uuid, l_result_header_mime) )
return [l_result_header_mime, getresult_response.content]


# public methods ---------------------------------------


# tries the authentication with jasperserver throug rest
def login(self, j_url, j_user,j_pass):
self.j_url= j_url


l_path_base_auth = "/rest/login"
l_path = self.j_url + l_path_base_auth


logger.info( "login (begin) [path=%s]"  %( l_path) )


try:
self.login_response = requests.post(l_path , params = {
'j_username':j_user,
'j_password':j_pass
})


if( requests.codes.ok != self.login_response.status_code ):
self.login_response.raise_for_status()


logger.info( "login (end)" )
return True
# see http://blog.ianbicking.org/2007/09/12/re-raising-exceptions/


except Exception, e:
logger.error("login (error) [e=%s]" % e )
self.__handle_exception(e, "LOGIN:CALL_ERROR",sys.exc_info())
#raise


def generate_report(self, rep_resourcepath,rep_format,rep_params):
self.__handle_report_metadata(rep_resourcepath)
[uuid,mime] = self.__handle_report_request(rep_resourcepath, rep_format,rep_params)
# TODO: how to handle async?
[out_mime,out_data] = self.__handle_report_reply(uuid)
return [uuid,out_mime,out_data]


@staticmethod
def create_client(j_url, j_user, j_pass):
client = JasperServerClient()
login_res = client.login( j_url, j_user, j_pass )
return client




class JasperServerClientError(Exception):


def __init__(self,exception_root,reason_id,reason_message=None):
super(JasperServerClientError, self).__init__(str(reason_message))
self.code = reason_id
self.description = str(exception_root) + " " + str(reason_message)
def __str__(self):
return self.code + " " + self.description
小开

https://github.com/nigma/django-easy-pdf

Template:

{% extends "easy_pdf/base.html" %}


{% block content %}
<div id="content">
<h1>Hi there!</h1>
</div>
{% endblock %}

View:

from easy_pdf.views import PDFTemplateView


class HelloPDFView(PDFTemplateView):
template_name = "hello.html"

If you want to use django-easy-pdf on Python 3 check the solution suggested here.

小开

Try wkhtmltopdf with either one of the following wrappers

django-wkhtmltopdf or python-pdfkit

This worked great for me,supports javascript and css or anything for that matter which a webkit browser supports.

For more detailed tutorial please see this blog post

小开

After trying to get this to work for too many hours, I finally found this: https://github.com/vierno/django-xhtml2pdf

It's a fork of https://github.com/chrisglass/django-xhtml2pdf that provides a mixin for a generic class-based view. I used it like this:

    # views.py
from django_xhtml2pdf.views import PdfMixin
class GroupPDFGenerate(PdfMixin, DetailView):
model = PeerGroupSignIn
template_name = 'groups/pdf.html'


# templates/groups/pdf.html
<html>
<style>
@page { your xhtml2pdf pisa PDF parameters }
</style>
</head>
<body>
<div id="header_content"> (this is defined in the style section)
<h1>\{\{ peergroupsignin.this_group_title }}</h1>
...

Use the model name you defined in your view in all lowercase when populating the template fields. Because its a GCBV, you can just call it as '.as_view' in your urls.py:

    # urls.py (using url namespaces defined in the main urls.py file)
url(
regex=r"^(?P<pk>\d+)/generate_pdf/$",
view=views.GroupPDFGenerate.as_view(),
name="generate_pdf",
),
小开

I get the code to generate the PDF from html template :

    import os


from weasyprint import HTML


from django.template import Template, Context
from django.http import HttpResponse




def generate_pdf(self, report_id):


# Render HTML into memory and get the template firstly
template_file_loc = os.path.join(os.path.dirname(__file__), os.pardir, 'templates', 'the_template_pdf_generator.html')
template_contents = read_all_as_str(template_file_loc)
render_template = Template(template_contents)


#rendering_map is the dict for params in the template
render_definition = Context(rendering_map)
render_output = render_template.render(render_definition)


# Using Rendered HTML to generate PDF
response = HttpResponse(content_type='application/pdf')
response['Content-Disposition'] = 'attachment; filename=%s-%s-%s.pdf' % \
('topic-test','topic-test', '2018-05-04')
# Generate PDF
pdf_doc = HTML(string=render_output).render()
pdf_doc.pages[0].height = pdf_doc.pages[0]._page_box.children[0].children[
0].height  # Make PDF file as single page file
pdf_doc.write_pdf(response)
return response


def read_all_as_str(self, file_loc, read_method='r'):
if file_exists(file_loc):
handler = open(file_loc, read_method)
contents = handler.read()
handler.close()
return contents
else:
return 'file not exist'
小开

If you have context data along with css and js in your html template. Than you have good option to use pdfjs.

In your code you can use like this.

from django.template.loader import get_template
import pdfkit
from django.conf import settings


context={....}
template = get_template('reports/products.html')
html_string = template.render(context)
pdfkit.from_string(html_string, os.path.join(settings.BASE_DIR, "media", 'products_report-%s.pdf'%(id)))

In your HTML you can link extranal or internal css and js, it will generate best quality of pdf.

小开
  • This is for Django >=3
  • This code converts HTML template to pdf file for any page. For example: post/1/new1, post/2/new2
  • pdf file name is last part in url. For example for post/2/new2, file name is new2

First install xhtml2pdf

pip install xhtml2pdf

urls.py

from .views import generatePdf as GeneratePdf
from django.urls import re_path
urlpatterns = [
#...
re_path(r'^pdf/(?P<cid>[0-9]+)/(?P<value>[a-zA-Z0-9 :._-]+)/$', GeneratePdf, name='pdf'),
#...
]

views.py

from django.template.loader import get_template
from .utils import render_to_pdf
# pdf
def generatePdf(request,cid,value):
print(cid,value)
pdf = render_to_pdf('myappname/pdf/your.html',cid)
return HttpResponse(pdf, content_type='application/pdf')

utils.py

from io import BytesIO #A stream implementation using an in-memory bytes buffer
# It inherits BufferIOBase


from django.http import HttpResponse
from django.template.loader import get_template


#pisa is a html2pdf converter using the ReportLab Toolkit,
#the HTML5lib and pyPdf.


from xhtml2pdf import pisa
#difine render_to_pdf() function
from .models import myappname
from django.shortcuts import get_object_or_404




def render_to_pdf(template_src,cid, context_dict={}):
template = get_template(template_src)
node = get_object_or_404(myappname, id =cid)
context = {'node':node}
context_dict=context
html  = template.render(context_dict)
result = BytesIO()


#This part will create the pdf.
pdf = pisa.pisaDocument(BytesIO(html.encode("ISO-8859-1")), result)
if not pdf.err:
return HttpResponse(result.getvalue(), content_type='application/pdf')
return None

Structure:

myappname/
|___views.py
|___urls.py
|___utils.py
|___templates/myappname/your.html
小开

I tried the best answer in this thread and it didn't work for python3.8, hence I had to do some changes as follows ( for anyone working on python3.8 ) :

import io
from xhtml2pdf import pisa
from django.http import HttpResponse
from html import escape


from django.template.loader import render_to_string


def render_to_pdf(template_src, context_dict):
html = render_to_string(template_src, context_dict)
result = io.BytesIO()






pdf = pisa.pisaDocument(io.BytesIO (html.encode("utf-8")), result)
if not pdf.err:
return HttpResponse(result.getvalue(), content_type='application/pdf')
return HttpResponse('We had some errors<pre>%s</pre>' % escape(html))

I had to change cgi to html since cgi.escape is depricated, and I replaced StringIO with io.ByteIO() as for the rendering I used render_to_string instead of converting the dict to context which was throwing an error.


提交答案