Introduction¶
Texthon is a library that processes templates written in a mix of text and Python.
Getting Texthon¶
No prerequisites aside from Python. Texthon is primarily developed using Python 3.3, but it’s 2.7 compatible as well.
Installation¶
Run setup.py install to install the library and the command line script into your Python installation. The scripts will be placed under the “Scripts” subdirectory.
If only the library is desired, just copy the texthon subdirectory (where all the py sources live) to a path under PYTHONPATH.
Design Philosophy¶
There are a lot of Python template engines, why did I make this one?
Texthon lets you embed Python code directly.
Unlike libraries that sandbox template execution, Texthon does as little interpretation as possible of control logic, preferring instead to pass statements straight to the Python interpreter.
To ease the debugging of templates, Texthon translates tracebacks to their original template source lines. The command line script can also dump generated Python code for inspection.
Texthon’s implementation is small and portable.
No reliance on non-standard libraries. The core implementation is just a couple of files that the user can copy and deploy at will. Texthon runs under both Python 2 and 3.
Texthon compiles templates into Python-like functions and organizes them into Python-like modules.
A template definition transforms into a callable function: it takes some parameters and returns a string. Templates in a file are grouped together and accessed much like a Python module.
One important feature is the support of mixins. Users can mix template modules together to form a new module. This allows the user to setup chains that’s almost like template inheritance.
The syntax of Texthon is inspired by Cheetah, which saved me from the unspeakable agony of having to rely on C preprocessors for game object serialization code. If only it wouldn’t bug me about that stupid Namemapper extension all the time.
Example¶
Here’s a simple template file, hello.tmpl.txt
#template main(who, count)
$who says:
#{for i in range(0, count):
hello world!
#}
#end template
using the script texthon with the following parameters:
texthon hello.tmpl.txt -P who=someone -P count=2
the following output is generated:
someone says:
hello world!
hello world!
For an example that’s closer to a real-world scenario, these templates demonstrate the use of Texthon to write C++ object reflection code: a task that would normally require tedious copy-and-pasting and/or C++ template/macro black magic.