pyJasper

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.

Usage

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 
             --form sign_keyname={{ alias of the certificate in the KeyStore }} #optional
             --form sign_reason={{ reason to be send w/ certificate }} #optional
             --form callback={{ callback_URL }} #optional
             http://localhost:8080/pyJasper/jasper.py > test.pdf

test.pdf should now contain a rendered PDF document. If you provided sign_keyname and sign_reason AND your backend installation knows this keyname, you get a signed pdf.

With parameter callback you tell the server to generate the pdf and send it to the given URL instead of immediate return.

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.

Installation

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 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 supervise tool for running the Jetty server.

exampledeployment on Amazon EC2

Create an new instance via AWS Management console

Launch instance
- Choose an AMI
  - Community AMI
  - e.g. ubuntu-images-eu/ubuntu-lucid-10.04-i386-server-20100923.manifest.xml
Instance details
Create Keypair
- Create a new keypair or choose one of your existing
Firewall
- Custom: TCP Port 5555 - 5555
- This custom option is not choosable in this setup-wizard but you can change the firewall settings later

Install and configure pyJasper

ssh -A ubuntu@dns-name-of-instance.com
sudo aptitude update
sudo aptitude safe-upgrade
sudo aptitude install openjdk-6-jre daemontools daemontools-run git-core
cd /usr/local
sudo git clone https://github.com/hudora/pyJasper.git
cd /usr/local/pyJasper

Now you have to add a command to /etc/crontab to delete old files from /tmp/pyJasper/ to keep the temporary files small. For example you can use this line to delete all files older than 7 days in /tmp/pyJasper/.

0  1    * * *   root    /usr/bin/find /tmp/pyJasper/ -mtime +7 -type f -exec /bin/rm {} \;

Optional: Now you have to upload a keystore file for signing your documents and set it in /usr/local/pyJasper/pyjasper/backend/pyJasper-httpd.sh

export PYJASPER_KEYSTORE_PASSWORD="thesecretpassphrase"
export PYJASPER_KEYSTORE_FILE="/usr/local/pyJasper/signature-keystore-file.ks"

Finally configure daemontools to run the following command as a service:

/usr/local/pyJasper/pyjasper/backend/pyJasper-httpd.sh -Xms128m -Xmx1G

History

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)
.01 one Java process per document (2006)

Files

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

mix86/pyJasper