I'll add some context:

ForPlay has an optional AppCacheLinker, which extends the
DirectInstallLinker.
ForPlay's SVN repository has a pair of svn:externals links, so it can track
changes in GWT trunk linker changes, since the latest DirectInstallLinker
fixes have not yet been released into a GWT SDK.

If your ForPlay project does not need the AppCacheLinker your .gwt.xml file
need to contain a linker override, i.e. something like this:
  <add-linker name="appcachelinker"/>


On Thu, Jun 16, 2011 at 11:15 AM, unnurg <[email protected]> wrote:

> Hi Russel - tryMarkerScript was removed from the new xsiframe linker.
> It's still present in the old, default linkers, but the new xsiframe
> linker needs to support Late Loading by default, which crashes when
> you do a document.write() and so we decided to remove this detection.
> Apologies that the comments didn't make it into the review thread, it
> was a rather lengthy discussion that we ended up having offline with
> various members of the team.
>
> I believe that you saw the break recently, because ForPlay was
> recently switched from the old linker to one of the new xsiframe based
> ones (I believe it uses the direct_install one, or the xsiframe one).
> I believe that switch was made because the new linker also supports
> CrossSite by default, and ForPlay needed this functionality.
>
> Note that if you don't do anything fancy, the other mechanisms should
> catch the right path, so in *most* cases, tryMarkerScript is not
> needed.  However, if you've renamed something and none of the other
> functions are triggering, then the easiest thing is to simply set a
> base or meta tag that will make sure that computeScriptBase works.
>
> I've also added Fred to the thread, since I don't know that much about
> ForPlay, and he's the one that switched ForPlay to the new linker, so
> he may have additional info and/or corrections.
>
> - Unnur
>
>
> On Jun 16, 8:55 am, Russell <[email protected]> wrote:
> > I think this might be a GWT issue.  I've tracked it as far as the
> > removal of tryMarkerScript() from computeScriptBase.js r10045:
> >
> > http://code.google.com/p/google-web-toolkit/source/browse/trunk/dev/c...
> >
> > If I add that function back in, everything works fine.  I'm not sure
> > why the other methods ofdirectorydetectionaren't working, but I
> > assume there was a reason this function was present in the original
> > code.  The code review on r10045 raises some questions that haven't
> > been answered; I guess I don't understand the Google review process.
> >
> > Thanks.
>



-- 
Fred Sauer
Developer Advocate
Google Inc.
1600 Amphitheatre Parkway
Mountain View, CA 94043
[email protected]

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" 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/google-web-toolkit?hl=en.

Reply via email to