|
Boost Testing : |
From: Beman Dawes (bdawes_at_[hidden])
Date: 2007-09-25 13:45:16
Rene Rivera wrote:
> Sebastian Redl wrote:
>> Phil Endecott wrote:
>>> Suggestion: abandon this stuff, and write it again in Python or C++.
>>> I speak from my own previous bad experience of this language, which I
>>> will not use again.
>>>
>> Aside from being complex, it's also rather big. And an imperative
>> version would be even bigger. So I'm not sure if that's such a good idea.
>
> From personal experience I can say that's likely not true :-) ... I did
> such a port for the doxygen2boostbook.xsl (49.7K)
> <http://svn.boost.org/trac/boost/browser/trunk/tools/boostbook/xsl/doxygen/>
> to Python (35.7K)
> <http://svn.boost.org/trac/boost/browser/trunk/tools/build/v2/tools/doxproc.py>.
> A 14K difference (~30%), and that includes a larger comment/code ratio
> on the Python code.
>
> Regardless the ideal, which has been mentioned before, is to go directly
> to a database and generate the pages dynamically.
That really seems like a better bet for the long run.
--Beman