This work was originally inspired by relatorio and diverged in implementation, scope and ambitions. Relatorio has many more features when py3o.template is mainly focused on ODT, ODS and text output.
py3o is an elegant and scalable solution to design reports using LibreOffice. py3o.template is the templating component that takes care of merging your data sets with a corresponding templated OpenOffice document.
It is plateform independent and does not require LibreOffice/OpenOffice itself to generate an ODF file.
If you want to generate a PDF or any other supported output format you will then need to have a server with either LibreOffice or OpenOffice and to install the py3o.renderserver on it. We also provide a docker image on the docker hub
If you want to have templating fusion & document conversion in one single web service usable from any language with just HTTP/POST you can install py3o.fusion server. Which also exists as a docker image
Python 3 support
py3o.template is python3 ready. But, yes there is a but... alas!, you'll need to install a trunk version of Genshi:
$ # activate your python env... $ svn checkout http://svn.edgewall.org/repos/genshi/trunk genshi_trunk $ cd genshi_trunk $ python setup.py build $ python setup.py install
We tested this with revision 1271. When genshi 0.8 is released we can officially say we support Python3 out of the box.
Below is an example that you can find in the source code inside the examples directory.
from py3o.template import Template t = Template("py3o_example_template.odt", "py3o_example_output.odt") t.set_image_path('staticimage.logo', 'images/new_logo.png') class Item(object): pass items = list() item1 = Item() item1.val1 = 'Item1 Value1' item1.val2 = 'Item1 Value2' item1.val3 = 'Item1 Value3' item1.Currency = 'EUR' item1.Amount = '12345.35' item1.InvoiceRef = '#1234' items.append(item1) for i in xrange(1000): item = Item() item.val1 = 'Item%s Value1' % i item.val2 = 'Item%s Value2' % i item.val3 = 'Item%s Value3' % i item.Currency = 'EUR' item.Amount = '6666.77' item.InvoiceRef = 'Reference #%04d' % i items.append(item) document = Item() document.total = '9999999999999.999' data = dict(items=items, document=document) t.render(data)
0.9.12 Unreleased yet
- Better error reporting for cases when the link is not positionned on its own line.
- Better error reporting for cases when an instruction link does not contain the equal sign.
0.9.11 Dec. 16 2016
- Added support for escaping False values. Values evaluated as False are replaced with an empty string during template rendering. This functionality is disabled by default and can be enabled with a new parameter on the Template object.
0.9.10 Oct. 3 2016
- Fixed a packaging issue (genshi requirement was unintentionally dropped in last release)
- Added more tests
- Added more support for types formatting
0.9.9 Apr. 13 2016
- Template parser: fix multiple loops on the same list of objects
- Template parser: fix multiple for loops on a variable that corresponds to a name (and not an attribute) in the data source.
- Template parser: when parsing for loops, extract the iterator even when the assigned variables are not used anywhere in the body.
- Template parser: now works with ODS templates.
- Template parser: fix a bug when parsing an unknown ast expression.
- Template parser: manage tuple, list and set literals.
- Template parser: improve the parsing of loops where the iterator is a function.
- Template parser: implement specific iteration behavior for enumerate. The second unpacked variable is linked to the function argument, as if the latter was used as the iterable object for the loop.
- Image injection: fix an exception that was raided when the image is not present in the data dictionary, even when it shouldn't (template initialized with ignore_undefined_variables=True)
- 100 % coverage, with many new unit tests and fixes on the old ones.
0.9.7.1 Feb. 19 2016
- This release fix a bug in the formatting function format_date where unicode date strings were not correctly managed.
0.9.7 Dec. 8 2015
- This release focuses on the parser. If you don't use the template parser you can skip this release as it does not change anything else.
- Fix a bug where the body of the if blocks were not evaluated by the parser
- Refactoring tests for clearer reading
- fix some missing/untested functions
- remove NotImplementedError from coverage
0.9.6 Nov. 20 2015
- Fixed formatting function
- Fixed if="" functionality that was not correctly parsed
- Display empty string instead of "False" when browsing relational field
- Code formatting in the ast parser
0.9.5 Oct. 30 2015
- 0.9.5 is just a rerelease of 0.9.4 with this README correctly amended...
- Added support for formatting functions in the template in order to control date number localization
- Added reference to original inspiration in the README
- Added example source code for ODS output.
- Fixed a memory usage in case of big output file size that was introduced by the dynamic image feature.
- Introduced an experimental TextTemplate
- Fixed support for special cases where page breaks removals removed trailing text (thanks to agix)
0.9.3 Jun. 26 2015
- Introduced dynamic image support (ie: pass in image data in the input iterator) and use it to insert images on the fly in the report)
0.9.2 Jun. 26 2015
WARNING: if you used the old image replacement system, you need to update your code to set static image from this:# old way t.set_image_path('logo', 'images/new_logo.png') # You should now set it like this instead # new way. Note the 'staticimage.' prefix added just before 'logo' t.set_image_path('staticimage.logo', 'images/new_logo.png')
WARNING: after updating your code you must also update your template to prefix your static images names with the 'staticimage.' prefix.
Added dynamic images support. You can now add images instructions inside for loops or anywhere in your templates and pass the image data inside your objects attributes.
Added support for image data (dynamic ones) being passed-in as base64 data. This is useful for some clients like Odoo's report_py3o because they store image data as base64 encoded fields.
0.9.1 Jun. 3 2015
- Fixed parser to read from the var name instead of its description, if you dont use the helper tool to introspect your reports this release is of no importance to you as it only fixes this particular point.
0.9 Jan. 8 2015
- Added support for soft page breaks
0.8 Nov. 19 2014
- Added better unit tests
- Fixed corner cases in the variable introspection mechanism
- Better handling of "ignore_undefined" that now also allows undefined images
0.7 Oct. 15 2014
- Added Python3 support
- Fixed a problem with validity of output in case the template contains a text:list inside a for loop
- Added new public methods to help report servers introspect the template data dictionary
- Added real unit tests (96% coverage ATM, way to go test team!)