Hi,

So after much mucking about I've found that when the gadget fails in
Chrome/Safari the init function is called and setStateCallback is set
but the stateChanged function is never called. This happens sometimes
on Chrome and never on Firefox (Mac). It seems to occur more often
when the Wave is larger with more gadgets.

It seems the issue is the way that I use the Wave library (since I am
using mode changes, I need wave-preview). If you use wave-preview as
per the documentation errors arise in Chrome and Safari (on the Mac)
but your gadget will work in Firefox. I've reported the bug.
http://code.google.com/p/google-wave-resources/issues/detail?id=333

However, the work around is to remove <require wave and <require
wave-preview from your gadget specification and add the line:
<script src="https://wave-api.appspot.com/public/wave.js";
type="text/javascript"></script>
to the top of the HTML of your gadget.

Hopefully this might solve your problem too Chris and Vision Jinx.

Pamela, what is the deal with
https://wave-api.appspot.com/public/wave.js. I noticed some of the
gadget examples use this method of accessing the wave api (which is
how I found it). Is this recommended? Do you know why this solves my
problem?

Thanks,
Jonny

On Tue, Oct 20, 2009 at 12:38 PM, pamela (Google Employee)
<pamela...@gmail.com> wrote:
> Hey Jonathan-
> Yes, sorry that I haven't been able to spend more time on your example yet.
> Gadgets debugging can be really frustrating, due to the nature of them being
> embedded.
> We're working on re-enabling debug log, that issue is here:
> http://code.google.com/p/google-wave-resources/issues/detail?id=325
> That may not help you as much as you'd desire however -- it's most useful
> for just logging state deltas and other messages.
> If it's a bug that affects API development, you can log it in the API issue
> tracker and we will try to respond within a few days.
> For a bug like that (the error w/multiple gadgets), it's helpful if you can
> add one of us to the Wave and specify the URL in the issue.
> I just tried a quick replication, in case we broke something badly in last
> release, and I couldn't get the error.
> - pamela
> On Tue, Oct 20, 2009 at 1:11 PM, Jonathan Hunt <j...@42quarks.com> wrote:
>>
>> Hi Pamela,
>>
>> I have been working on giving you a minimal example but I'm now
>> finding the "Uncaught TypeError's" on almost any wave that has
>> multiple gadgets inserted in Chrome or Safari (i.e. yes no maybe
>> gadgets or see my minimal example
>> http://waveyscience.appspot.com/eqygadget/test.xml). However, the
>> example gadgets seem to keep working after this error occurs but mine
>> don't for some reason. This is the latest Chrome/Safari on Mac (when
>> Safari doesn't crash outright ;-)). I don't always get these errors,
>> but often, and particularly on waves which have lots of gadgets.
>>
>> Where is the best place to report Wave bugs? Getting these errors with
>> Yes/No/Maybe gadget makes me think this is not my bug.
>>
>> Also, I don't mean to complain, but I've put a significant amount of
>> effort into trying to narrow down the source of the problem, build a
>> simple test case etc. Google Wave JS has debug versions available (I
>> can see them if I change &debug=0 to &debug=1) but I can't put Wave in
>> debug mode which means I can't use them to get a line number or much
>> useful information about where in the Wave js code the error is
>> occurring. I understand this is a dev preview so I don't expect no
>> problems, but for both of our productivities, it would be nice to have
>> access to a debug mode (on a related note I got an email telling me to
>> choose a username for the sandbox but never heard back, although from
>> what I hear sandbox doesn't have debug mode anymore).
>>
>> Thanks for any help.
>> Jonny
>>
>> On Tue, Oct 20, 2009 at 1:39 AM, pamela (Google Employee)
>> <pamela...@gmail.com> wrote:
>> > Hmm. Maybe you or Jonny can try to put together a minimal example that
>> > shows
>> > this? It might be easy to see if you draw a box (DIV) inside a gadget of
>> > particular heights. Jonny's example is a bit harder to test with because
>> > of
>> > the difference in mode switch, where it sometimes shows a form and
>> > sometimes
>> > nothing at all.
>> >
>> > On Tue, Oct 20, 2009 at 2:15 AM, Vision Jinx <vjn...@gmail.com> wrote:
>> >>
>> >> I have had the same issue since last week or more.
>> >>
>> >> I have a gadget that uses Dynamic Height, I asked Pamela Fox (on
>> >> Twitter) if there were any issues with the lib, she replied saying
>> >> none that she knew of, I have the same issue in Wave and iGoogle.
>> >>
>> >> Works fine with Firefox but on Mac w/Safari and Google chrome (on
>> >> Windows) I have a lot of extra white space. I checked the rendered
>> >> code (on iGoogle) and the Gadget is being loaded at 1157px (on
>> >> average) which is many times the height of the gadgets content.
>> >>
>> >> I sent an email also to the iGoogle dev team but holding true to their
>> >> form no reply and in my exp they are more interested in breaking
>> >> things then fixing them or addressing developer concerns/issues (which
>> >> is why they lost all their Gurus ~nuff said).
>> >>
>> >> Regards,
>> >> Vision Jinx
>> >>
>> >>
>> >> On Oct 18, 11:30 pm, Chris Searle <chrisdsea...@gmail.com> wrote:
>> >> > Have a gadget with the dynamic-height feature.
>> >> >
>> >> > The last line of the stateUpdated method is
>> >> > gadgets.window.adjustHeight();
>> >> >
>> >> > However, I see that the gadget stays at 200px quite often. Leaving
>> >> > and
>> >> > re-opening the wave sometimes helps, not always. If you have the
>> >> > gadget several times in a wave some resize some don't. It all seems a
>> >> > little random.
>> >> >
>> >> > Is this simply a preview problem (can't say that I've seen it in the
>> >> > developer sandbox) ?
>> >>
>> >
>> >
>> > >
>> >
>>
>>
>>
>> --
>> Jonathan J Hunt <j...@42quarks.com>
>> Homepage: http://www.42quarks.com
>> (Further contact details there)
>> "Physics isn't the most important thing. Love is." Richard Feynman
>>
>>
>
>
> >
>



-- 
Jonathan J Hunt <j...@42quarks.com>
Homepage: http://www.42quarks.com
(Further contact details there)
"Physics isn't the most important thing. Love is." Richard Feynman

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"Google Wave API" group.
To post to this group, send email to google-wave-api@googlegroups.com
To unsubscribe from this group, send email to 
google-wave-api+unsubscr...@googlegroups.com
For more options, visit this group at 
http://groups.google.com/group/google-wave-api?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to