Re: [Grok-dev] Grok-dev Digest, Vol 90, Issue 3
Paul Sephton
Let me at the outset express my thanks for everyones answers regarding my grok.Layout query. And thanks to everyone who have and continue to make zope and grok a great product.
My personal take on ZCML is that if I had ever been forced to use it, I would never have built anything at all using the Zope toolkit. Yes, I feel that strongly about it.
So what is my problem with ZCML? The fact that I cannot imagine having to maintain a configuration file for views and adapters separate from my code, regardless of the format of the configuration file. To me, this would mean switching context while coding, and I hate that. There are two places to check if things go wrong, and the ZCML is not available from the debugger.
Probably the biggest thing that Grok does for me is to make me not have to bother with ZCML, and turns Zope into a true pythonic technology. Whats not to love about it?
Anyway, I think you can do a ZMI release which integrates with grokcore.admin without having to release a whole new grokproject. After it is tested for a while one might consider adding it to the core. It is best to keep dependencies to an absolute minimum.
Regards,
Paul
Christopher Lozinski
On 10/23/15 1:05 PM, Paul Sephton wrote:
> y personal take on ZCML is that if I had ever been forced to use it, I
> would never have built anything at all using the Zope toolkit. Yes, I
> feel that strongly about it.
are preaching to the choir. I am not sure the best way to get this
message out. The Plone guys do not want to hear it. The Pyramid guys
do not want to hear it. New python users who are interested in
ZODB might want to hear it. I wonder what it takes to get grok
supporters on this mailing list to be more active users?
I did manage to put grok on the ZODB map.
http://blogory.org/group/zodb-ecosystem
When you compare grok with ZCML and with Pyramid View Decorators and
View Functions, clearly grok is the way to go.
But I have no ability to change other people's thinking.
I do have to release a new version of grokproject. Probably call it
zopache.grok-install. Just a very simple install. Not as nice as
grok.project. The current version pins to a buggy zope.copypastemove.
I need a newer version.
I will probably release a zopache.containerview. I can do it with
ZCML. That gets lots of Plone users. That can't hurt.
And then the zopache-grok install will include zopache.zmi which has the
zopache application. Add a zopache application to the grok root, and
you have a nice container view.
So writing up my plans did help my thinking about what I had to do.
I am also getting very interested in jade templating.
http://jade-lang.com/
I have a browser based jade editor.
Chris
I
_______________________________________________
Grok-dev mailing list
Grok...@zope.org
https://mail.zope.org/mailman/listinfo/grok-dev