Indepth Example#

In this page I will detail the way I created the reports that can be found in the examples directory.

Let’s start with the content of common.py, this file stores the definition of an invoice that will be used to create the different reports. The invoice is a simple python dictionary with some methods added for the sake of simplicity:

from os.path import dirname, join


class Invoice(dict):

    @property
    def total(self):
        return sum(line['amount'] for line in self['lines'])

    @property
    def vat(self):
        return self.total * 0.21


inv = Invoice(customer={'name': 'John Bonham',
                        'address': {'street': 'Smirnov street',
                                    'zip': 1000,
                                    'city': 'Montreux'}},
              lines=[{'item': {'name': 'Vodka 70cl',
                               'reference': 'VDKA-001',
                               'price': 10.34},
                      'quantity': 7,
                      'amount': 7 * 10.34},
                     {'item': {'name': 'Cognac 70cl',
                               'reference': 'CGNC-067',
                               'price': 13.46},
                      'quantity': 12,
                      'amount': 12 * 13.46},
                     {'item': {'name': 'Sparkling water 25cl',
                               'reference': 'WATR-007',
                               'price': 4},
                      'quantity': 1,
                      'amount': 4},
                     {'item': {'name': 'Good customer',
                               'reference': 'BONM-001',
                               'price': -20},
                      'quantity': 1,
                      'amount': -20},
                    ],
              id='MZY-20080703',
              status='late',
              bottle=(open(join(dirname(__file__), 'bouteille.png'), 'rb'),
                  'image/png'))

Create a simple OpenOffice Writer template#

Let’s start with the simple template defined in basic.odt.

_images/basic.png

This report will be created and rendered with the following three line of code:

from relatorio.templates.opendocument import Template
basic = Template(source='', filepath='basic.odt')
file('bonham_basic.odt', 'wb').write(basic.generate(o=inv).render().getvalue())

Notice that the dictionary passed to generate is used to bind names to make them accessible to the report. So you can access the data of the invoice with a Text Placeholder containing o.customer.name. This is where you can see our Genshi heritage. In fact, all reports using relatorio are subclasses of Genshi’s Template. Thus you can use most of the goodies provided by Genshi.

To iterate over a list you must use an hyperlink (created through ‘Insert > Hyperlink’) and encode as the target the Genshi expression to use. The URL-scheme used must be relatorio. You can use whatever text you want as the link text, but we find it way more explicit to display the Genshi code used. Here is the example of the for loop.

_images/hyperlink.png

And thus here is our invoice, generated through relatorio:

_images/basic_generated.png

Advanced zip options#

It can happen that the document being generated result in a file which uncompressed size is bigger than 2GiB. The default settings of Python’s zipfile library do not allow to generate those files.

To circumvent this issue, opendocument templates support additional parameters to the generate method. Those parameters are:

_relatorio_compresslevel:

This parameter defines the compresslevel parameter of the underlying Zipfile call.

_relatorio_chunksize:

This parameter defines the size of each chunk of data streamed to the underlying Zipfile object.

_relatorio_zip64:

This parameter forces the handling of ZIP64 extensions. It should be set to true if the size of the file may be bigger than 2GiB.

_relatorio_compression_method:

This parameter defines the compression paramater of the underlygin Zipfile call.

One step further: OpenOffice Calc and OpenOffice Impress templates#

Just like we defined a Writer template it is just as easy to define a Calc/Impress template. Let’s take a look at pivot.ods.

_images/pivot.png

As usual you can see here the different way to make a reference to the content of the invoice object:

  • through the Text Placeholder interpolation of Genshi

  • or through the hyperlink specification I explained earlier.

Note that there is another tab in this Calc file used to make some data aggregation thanks to the data pilot possibilities of OpenOffice.

And so here is our rendered template:

_images/pivot_rendered.png

Note that the type of data is correctly set even though we did not have anything to do.

Everybody loves charts#

Now we would like to make our basic report a bit more colorful, so let’s add a little chart. We are using PyCha to generate them from our pie_chart template:

options:
    width: 600
    height: 400
    background: {hide: true}
    legend: {hide: true}
    axis: {labelFontSize: 14}
    padding: {bottom: 10, left: 10, right: 10, top: 10}
chart:
    type: pie
    output_type: png
    dataset:
    {% for line in o.lines %}
      - - ${line.item.name}
        - - [0, $line.amount]
    {% end %}

Once again we are using the same syntax as Genshi but this time this is a TextTemplate. This file follow the YAML format thus we can render it into a data structure that will be sent to PyCha:

  • the options dictionary will be sent to PyCha as-is

  • the dataset in the chart dictionary is sent to PyCha through its .addDataset method.

If you need more information about those go to the pycha website.

And here is the result:

_images/pie.png

A (not-so) real example#

Now that we have everything to start working on our complicated template invoice.odt, we will go through it one step at a time.

_images/complicated.png

In this example, you can see that not only the openoffice plugin supports the for directive, it also supports the if directive and the choose directive that way you can choose to render or not some elements.

The next step is to add images programmatically, all you need to do is to create frame (‘Insert > Frame’) and name it image: expression just like in the following example:

_images/frame.png

The expression when evaluated must return a couple whose first element is a file object containing the image and second element is its mimetype. Note that if the first element of the couple is an instance of relatorio.Report then this report is rendered (using the same arguments as the originating template) and used as the source for the file definition.

This kind of setup gives us a nice report like that:

_images/complicated_rendered.png