Not in this case, actually. It turns out that caja.js can't be minified due
to an error that gets produced in html_sanitizer. I'll contact Mike Samuel
to see what we can do about this.
For now I've excluded all caja javascript from the minification process.
On Wed, Feb 20, 2008 at 10:14 AM, Cassie <[EMAIL PROTECTED]> wrote:
> I vote for focusing on some tests for the sample container. Then we could
> require the tests to all pass for checkins (a presubmit trigger or
> whatever
> svn's equivalent is). This would require us to get selenium or some other
> browser testing framework up and running.
But there's no infrastructure for this -- browser based testing tools can
only verify that something works on the browsers you have installed on your
machine.
Tests would be great, but I don't think there's any way to ensure that
they're actually being run (or that they're trustworthy).
>
>
> If someone helps get the framework setup I think I have a selenium test
> that
> will work.
>
> - Cassie
>
>
> On Wed, Feb 20, 2008 at 9:58 AM, Paul Lindner <[EMAIL PROTECTED]> wrote:
>
> > I have seen similar problems. It appears that the Java 1.6 developer
> > preview on Mac OS X has the same class loader problem that occurs on
> > Java 1.5.
> >
> > I had to edit features/core/core.js to have this:
> >
> > var gadgets = gadgets || {};
> >
> > otherwise the dynamic-height js gets loaded first and that gets
> > overwritten. On Fedora 8 with openjdk 1.7 things run a bit more
> > smoothly. However rpc is not working at all in FF 3 or Safari 3.
> >
> > As far as samplecontainer.html goes, it still references ifpc.js, so
> > it's not even using the new rpc code.
> >
> > Can we please get some more eyes on fixing the sample apps? This is
> > the first thing that new users see, and if it doesn't work out of the
> > box people are going to get very confused and send lots of support
> > requests to the mailing list.
> >
> > With that in mind perhaps it's time to start publishing snapshots for
> > end users, and perhaps add a requirement that all checkins must not
> > break the samples, or at the very least add big warning messages
> > to the sample to designate that it's known broken...
> >
> > On Wed, Feb 20, 2008 at 09:19:05AM -0800, Chak Nanga wrote:
> > > Hi,
> > >
> > > I updated my Shindig source (having been away for a week) and I¹m
> unable
> > to
> > > load/render even the SocialHelloWorld.xml gadget.
> > >
> > > Firebug reports the following errors when I bring up
> > samplecontainer.html
> > > (and when it tries to load the SocialHelloWorld.xml gadget by
> default):
> > >
> > > gadgets.container has no properties
> > >
> http://localhost:8080/gadgets/files/samplecontainer/samplecontainer.html
> > > Line 93
> > >
> > > gadgets.rpc has no properties
> > > http://localhost:8080/gadgets/files/container/gadgets.js Line 226
> > >
> > > gadgets.container has no properties
> > >
> http://localhost:8080/gadgets/files/samplecontainer/samplecontainer.html
> > > Line 50
> > >
> > > I cleared the browser cache(to make sure I was getting the latest
> > > samplecontainer) and retried I get the same errors.
> > > I pulled a brand new snapshot of the trunk (into a new directory),
> built
> > it
> > > and pointed my browser to the samplecontainer.html same errors.
> > >
> > > Anyone else seeing this? If not, any pointers to get this resolved
> would
> > be
> > > helpful.
> > >
> > > Thanks
> > > Chak
> > >
> >
> > --
> > Paul Lindner
> > hi5 Architect
> > [EMAIL PROTECTED]
> >
>
--
~Kevin
If you received this email by mistake, please delete it, cancel your mail
account, destroy your hard drive, silence any witnesses, and burn down the
building that you're in.
