[Grok-dev] Re: Neanderthal sprint topics

Martijn Faassen faassen at startifact.com
Sun Sep 30 03:51:40 EDT 2007


Martin Aspeli wrote:
> Philipp von Weitershausen wrote:
> 
>> By the way, I looked at integrating Genshi into Zope 3 the other day. 
>> It's shockingly easy. Takes you about 3 lines of code. I wouldn't mind 
>> ditching PageTemplates (at least as the preferred option) altogether 
>> since Genshi
>>
>> * has more appealing templating syntax (I think it's much faster)
>>
>> * is much more push-oriented (rather than ZPT's orientation towards pull)
>>
>> * has support for theming and pipelining.
>>
>> I know this sounds a bit revolutionary, but a prototype would be cool.
> 
> +1 for prototype. I've used Genshi, and it's very good.
> 
> That said, don't make using ZPTs too difficult, please, and make sure 
> that things still interoperate. I don't see why they wouldn't, but you 
> need to be able to re-use a viewlet that's ZPT based at the very least. 
> There would also be some impedance mis-match if e.g. you had components 
> that presented UI with macros and slots for you to plug into, but that's 
> really more of a problem for a pluggable system such as Plone than a 
> start-from-scratch Grok app.

+1 too (as people already knew). I think the sprinters, if they pick 
this topic, should concentrate on:

* testing basic use cases with Genshi (static content, equivalent of 
macros, calling up zope 3 views on 'context', viewlets, etc, etc)

* Basic ZPT interoperability of some kind

Buffet is rather framework bound, so isn't straightforwardly reusable. 
The proposed successor hasn't gotten off the ground yet. So while this 
is definitely still interesting to me, let's see about getting Genshi 
integrated first.

Some hints from Philipp (a branch?) would be interesting to see.

Regards,

Martijn



More information about the Grok-dev mailing list