Hi all, now that we have Zope 3.4.0 finally behind us, let's look forward. As I said in the release notes, I am really willing to switch to a 6 months release cycle again. I think there are three areas that we can work on: - Python 2.6 support. I think this should be straight forward. We have to update packages to not use Python 2.6 keywords and review the AST changes. - Dependency reduction. Well, we are in the middle of this work already. - Improve project setup As Shane pointed out, it is hard to get started. The solution could be a combination of documentation and tools, like zopeproject. I am going to setup a Zope 3.5 development KGS shortly. Jim has already created the directory for me on the server. Thoughts? Regards, Stephan -- Stephan Richter Web Software Design, Development and Training Google me. "Zope Stephan Richter"
Hey, Stephan Richter wrote: [snip]
- Improve project setup
As Shane pointed out, it is hard to get started. The solution could be a combination of documentation and tools, like zopeproject.
It's not hard to get started as long as you use Zope 3 in the form of Grok. :) Grok has done a lot of work on getting a project set up, Windows support, and the like, and what you get when you use grokproject is essentially a Zope 3 project (with the extra Grok packages). grokproject has a lot of interesting features. It looks for the URL for the versions of the most recent release of Grok. It's now installing paster-based setup. It's also using things like z3c.recipe.eggbasket to pull in a whole bunch of eggs in a single tarball (from a single URL which means less potential for failure in theory). Etc. It might be worthwhile to review that work and reuse bits of it. Regards, Martijn
On Sun, Feb 1, 2009 at 07:51, Stephan Richter <srichter@cosmos.phy.tufts.edu> wrote:
- Improve project setup
As Shane pointed out, it is hard to get started. The solution could be a combination of documentation and tools, like zopeproject.
zc.zope3recipes is very interesting, but with a "deficiency": no default configuration for app & instance part. A good thing could be to integrate zopeproject.
Thoughts?
I dream to see SQLAlchemy DB as first class citizen in Zope3. We use Zope3 for developing our products and it's the main failure for us. -- Sebastien Douche <sdouche@gmail.com>
2009/2/2 Sebastien Douche <sdouche@gmail.com>:
On Sun, Feb 1, 2009 at 07:51, Stephan Richter <srichter@cosmos.phy.tufts.edu> wrote:
- Improve project setup
As Shane pointed out, it is hard to get started. The solution could be a combination of documentation and tools, like zopeproject.
zc.zope3recipes is very interesting, but with a "deficiency": no default configuration for app & instance part. A good thing could be to integrate zopeproject.
I also just committed my little sandbox setup that can be used to take a quick look at Zope3. I personally use it for playing with some eggs and for showing "zope3 the application server" to other people. It's based only on zc.buildout, zc.zope3recipes, and pure zcml setup. It doesn't use paste or any z3c package - just plain old zope3 on twisted with rotterdam and so on. It also contains an example of code that creates a new content type (with doctests). If we want people to start learning Zope3 through ZMI, may be something like that could be developed and used as a demo instead of z3c.formdemo :) I quite like it because it contains very few files and is rather easy to understand by newbie. Next, we may want to point people to packages like z3c.pagelet and z3c.form, so some tutorial should be written on how to add support for that to their current setup (z3c.rotterdam could be used as an easy way). Here it is: http://svn.zope.org/Sandbox/nadako/zopesandbox/ -- WBR, Dan Korostelev
Am Sonntag 01 Februar 2009 07:51:43 schrieb Stephan Richter:
Hi all,
now that we have Zope 3.4.0 finally behind us, let's look forward. As I said in the release notes, I am really willing to switch to a 6 months release cycle again.
I think there are three areas that we can work on:
- Python 2.6 support. - Dependency reduction. - Improve project setup
Yes, I'd also suggest this. What I'd also like to have is: - More focus on z3c.form/z3c.pagelet and less on the ZMI. - More focus on SQLAlchemy integration. And something I wonder from time to time is if it would be possible to integrate a recent Twisted release into Zope3, or, even better, directly use the current twisted egg with Zope3. Best Regards, Hermann -- hermann@qwer.tk GPG key ID: 299893C7 (on keyservers) FP: 0124 2584 8809 EF2A DBF9 4902 64B4 D16B 2998 93C7
2009/2/8 Hermann Himmelbauer <dusty@qwer.tk>:
And something I wonder from time to time is if it would be possible to integrate a recent Twisted release into Zope3, or, even better, directly use the current twisted egg with Zope3.
The problem is that the Twisted egg doesn't contain the "web2" module, used by zope's twisted-based WSGI server, so we can only try to integrate the latest release into zope.app.twisted itself, downloading the Twisted release using svn:external as we do now. I tried do do that before, but tests failed and I was unable to fix that, as I don't know twisted at all (shame on me). -- WBR, Dan Korostelev
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hermann Himmelbauer wrote:
Am Sonntag 01 Februar 2009 07:51:43 schrieb Stephan Richter:
Hi all,
now that we have Zope 3.4.0 finally behind us, let's look forward. As I said in the release notes, I am really willing to switch to a 6 months release cycle again.
I think there are three areas that we can work on:
- Python 2.6 support. - Dependency reduction. - Improve project setup
Yes, I'd also suggest this. What I'd also like to have is:
- More focus on z3c.form/z3c.pagelet and less on the ZMI. - More focus on SQLAlchemy integration.
And something I wonder from time to time is if it would be possible to integrate a recent Twisted release into Zope3, or, even better, directly use the current twisted egg with Zope3.
I would rather move the Twisted support out into a non-core package, and focus on making the Zope3 components play nicely with *any* WSGI-compliant server. The fact that we still have a forked Twisted in Zope3 is directly tied to the absence of a crucial component in the released version of Twisted. Tres. - -- =================================================================== Tres Seaver +1 540-429-0999 tseaver@palladion.com Palladion Software "Excellence by Design" http://palladion.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFJkEjl+gerLs4ltQ4RAl4pAKCET8mNIdimYzpuiPegRY4gpeXBaQCfXdzC JDsA2cClJ1OfStL5QqA6ZKA= =/OZS -----END PGP SIGNATURE-----
2009/2/9 Tres Seaver <tseaver@palladion.com>:
I would rather move the Twisted support out into a non-core package, and focus on making the Zope3 components play nicely with *any* WSGI-compliant server. The fact that we still have a forked Twisted in Zope3 is directly tied to the absence of a crucial component in the released version of Twisted.
Well, it's already not so bad. The zope.app.twisted is not the core package, and the zope.app.wsgi makes it easy to get the WSGI application of "Zope". I've also just checked in an app_factory for PasteDeploy for it, so it can be used as an application component in the PasteDeploy pipeline without any additional python code. The zope.publisher also provides a simple WSGI application for use with paste. -- WBR, Dan Korostelev
participants (6)
-
Dan Korostelev -
Hermann Himmelbauer -
Martijn Faassen -
Sebastien Douche -
Stephan Richter -
Tres Seaver