Hi everybody,
I'm currently in the process of moving those two libraries off to their
new git repositories. If you are currently monitoring the hg
repositories, go over to the new github pages where not only the code
is, but also the bug trackers and a better way to provide patches.
This will also finally make it possible to decide on the future of
Werkzeug because I will maintain with the 0.7 release a proper
maintenance branch where bugfixes go and an experimental branch that
should clean up code for 1.0 and remove a few features that really
should be in their own package.
A big issue with Werkzeug currently is the magical import system. A
couple of users raised concerns there and I am in favor of changing
this. In case you are unaware of how it works, this is the process in a
nutshell: Werkzeug replaces itself on import with a lazy object that
resolves attribute lookups. The idea of this originally was to unify
the import paths to one module, hiding the module that actually
implements the object.
This however turns out to be a problem for two reasons: first of all,
pickle still looks at the original location for the object which is why
we still have to support old import paths internally if one does not
want to break pickle.
Secondly it causes a lot of slowdown on systems like GAE. Instead it
would have been wiser to support imports from
"werkzeug.themodulethatimplementsit" directly for all objects.
Because this is a huge change I want to have a votum on that first. A
possible upgrade path would be to have a couple of more Werkzeug
releases until 1.0 which still supports the lazy imports from werkzeug
itself.
Regards,
Armin
--
You received this message because you are subscribed to the Google Groups
"pocoo-libs" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/pocoo-libs?hl=en.
