There are some unnecessary dimension that are introducing xN factors
in the number of permutations, e.g. ClientTypes.

I'm also in favour of only building for Webkit (user.agent = safari)
so long as WebSocket is the only functioning communication channel,
and maybe Firefox 4 (user.agent = gecko1_8).  But the iphone, android,
and IE permutations aren't worth building right now.

And yes, having separate development and production GWT builds would
make things much better.  Disabling aggressive optimization for a dev
build will also significantly improve compile time.

This should be quite doable by splitting WebClient.gwt.xml into three
(a base module, a production extension, and a development extension),
and splitting the ant target into two (one for prod, one for dev).  I
can probably prepare that change later today, unless someone else gets
to it first.

On Oct 19, 7:44 pm, Alex North <[email protected]> wrote:
>    The WebClient.gwt.xml was recently fixed to produce all the necessary
>    permutations (29) for a production build. This is necessary for good 
> browser
>    support, etc, but is really slow for development (with typically just one
>    browser most of the time). The compile_gwt ant target adds -style PRETTY
>    which makes the JS larger than it should be.
>
>    Is it possible to set GWT properties in the ant file, so we could have a
>    build target that restricts to chrome/safari?
>
>    - compile_gwt: unrestricted, no logging
>       - compile_gwt_dev: safari, PRETTY, logging enabled

-- 
You received this message because you are subscribed to the Google Groups "Wave 
Protocol" 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/wave-protocol?hl=en.

Reply via email to