[openspending-dev] Genshi -> Jinja

[Friedrich Lindenberg]

Hey Nigel,

it's really cool you're going to take this on, even though I still don't
fully grok your hatred of Genshi. It validates, which is very validating :)
In any case, I'd argue against a partial conversion as well - there's just
a large risk it won't get done if there is no breakage and then we really
have two problems...

The only really hard-core thing that I could think of for the templating
engine is building in support for Ghost.py (http://jeanphix.me/Ghost.py/):
see when the Google comes, render things out via Ghost and make sure our
ajaxy bits are indexed as well.

That's not going to make OS more stable, but in terms of SEO it may be
fairly crucial. *If* you do this, then there's also really no argument
against moving a lot more stuff into the API and to increasingly deprecate
the web UI in favour of a well-docced API with some JS to use it. So: why
not move to Handlebars? Jinja is so 2009!

Cheers,

 - Friedrich


On Tue, Feb 5, 2013 at 6:30 PM, Nigel Babu <nigel.babu at okfn.org> wrote:

> Hi Friedrich,
>
> I had a chat with Martin earlier today and as part of it, I looked up the
> feasibility of converting OpenSpending to Jinja2.
>
> It seems that we can do it part by part and it would probably take some
> time, but it is possible. Pylons does support two templating engines at the
> same time, but for the duration that we do the conversation, it's going to
> be hell because some code will definitely be duplicated.
>
> What are your thoughts before I engage in this foolish move (in my free
> time)?
>
> Nigel Babu
> Developer, Open Knowledge Foundation
> http://www.okfn.org/
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/openspending-dev/attachments/20130205/85230bdb/attachment.html>