JasperReports webserver and Python client
Python Shell
Pull request Compare This branch is 29 commits behind hudora:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Failed to load latest commit information.



pyJasper is a set of python based tools to handle JasperReports.
Since jasper reports is a Java application you need Java installed.

Jython (bundled with pyJasper) is used to access the JasperReports
library via HTTP. Your pure Python clients can transparently generate
JasperReport Documents.

See StackOverflow and this Blogposting to understand what problem pyJasper is trying to solve.


Servlet Interface

The servlet keeps no state at all. You have to supply it with an XML datasource, an XPath expression for that datasource and the JRXML report design. You get back the generated PDF or an plain text error message. The respective data has to be submitted via the form variables ‘xpath’, ‘design’ and ‘xmldata’.

To try it out you can use curl. E.g. do to pyjasper/backend and start the jetty servlet container (sh pyJasper-httpd.sh). Then use curl in another window to request rendering of a PDF.

curl -X POST —form xpath=//lieferscheine/lieferschein —form design=@reports/Lieferschein.jrxml —form xmldata=@sample-xml/Lieferschein.xml http://localhost:8080/pyJasper/jasper.py > test.pdf

test.pdf should now contain a rendered PDF document.

Python interface

You are expected to subclass pyjasper.JasperGenerator and call it’s generate_pdf() function. Usually you only have to overwrite the __init__() and generate_xml(self, ...) functions and use the the Python ElementTree API to generate an xml-tree. E.g.

class MyPDFGenerator(JasperGenerator): """Jasper-Generator for Greetingcards""" def init(self): super(MovementGenerator, self).init() self.reportname = ‘reports/Greeting.jrxml’ self.xpath = ‘/greetings/greeting’ self.root = ET.Element(‘gretings’) def generate_xml(self, tobegreeted): """Generates the XML File used by Jasperreports""" ET.SubElement(self.root, ‘generator’).text = revision for name in tobegreeted: xml_greeting = ET.SubElement(self.root, ‘greeting’) ET.SubElement(xml_greeting, “greeting_to”).text = unicode(name) ET.SubElement(xml_greeting, “greeting_from”).text = u"Max" return xmlroot

Now you can use MyPDFGenerator like this:

generator = MyPDFGenerator() pdf = generator.generate([‘nik’, ‘tobias’, ‘chris’, ‘daniel’]) open(‘/tmp/greetingcard.pdf’, ‘w’).write(pdf)

The Python client finds the URL of the Jasper Servlet by checking the PYJASPER_SERVLET_URL environment variable. It this Variable is not set, a default value of http://localhost:8080/pyJasper/jasper.py is used.


Get it at the Python Cheeseshop or at GitHub
To install the Python client interface just execute python setup.py install as administrator. This should install the requred dependency ([http://code.google.com/p/httplib2/ httplib2]) automatically. For the Server part there exist no automatic setup script. Just copy pyjasper/backend/ to a suitable location and start pyJasper-httpd.sh I use Dan Bernsteins [http://cr.yp.to/daemontools/supervise.html supervise] tool for running the Jetty server.


  • 0.2.1* public release (Summer 2008)
  • 0.2* release based on Jetty/Servlets (late 2007)
  • 0.1.1* public release on hosted-projects.com
  • 0.1* release based on long running Java Process (late 2006)
  • 0.01 one Java process per document (2006)


  • client.py – contains high-level Python functions for report generation. Should theoretically be able to work with report generators other than JasperReports. So far builds on JasperClient.py.
  • backend/ – contains tools to drive JasperReports and actually build reports. The backend was once based on XmlJasperInterface by Jonas Schwertfeger but hase benn moving away from it for some time.
  • backend/lib – JasperReports and Jython Java Libraries
  • reports – report source files (*.jrxml)
  • sample-xml – sample XML files to work with reports
  • sample-pdf – sample PDFs generated by using reports with sample XML