> On 2011-12-09 14:25:51, Jesse Ciancetta wrote: > > LGTM > > > > This is an interesting capability -- its too bad we couldn't find a way to > > get it done without two requests though. > > > > I did actually have one other thought about an alternative that could work > > (but again has its own share of drawbacks -- many of which are the same as > > the head approach) but thought I'd share anyway in case it spurs some > > thoughts from others too. I was thinking there might be a way to add > > another proxy to shindig for cases like this -- maybe something like > > IframeProxyServlet -- so the iframe src would point to > > shindig-server/iframeProxy?url=http://example.com/theIframeContent.html -- > > and then on the shindig side it could fetch the content, add a script > > include for the shindig RPC library and then a script block that would fire > > a message to the parent container as soon as it's processed. On the parent > > page you could listen for the RPC call and when fired you'd know the iframe > > loaded. You could probably even do the same thing for the failure case on > > the shindig side -- you return a page with a friendly error message saying > > that the content failed to load and fire the RPC to the parent page with a > > message letting it know that the iframe failed -- and you could even > > include any details you needed to as to exactly what failed. > > > > Again -- just throwing out some thoughts here -- there would be plenty of > > issues with this approach too (besides being much more work) like the > > iframe domain would be that of shindig instead of the real host, relative > > links in documents probably wouldn't work unless they were rewritten (which > > I guess shindig already knows how to do) etc -- but thought it was still > > worth sharing in case anyone else was interested.
Thanks Jesse. This is def something to consider for a future enhancement. - Ryan ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/3037/#review3779 ----------------------------------------------------------- On 2011-12-08 17:58:33, Ryan Baxter wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/3037/ > ----------------------------------------------------------- > > (Updated 2011-12-08 17:58:33) > > > Review request for shindig, Dan Dumont and Stanton Sievers. > > > Summary > ------- > > When you call commoncontainer.navigateUrl if the URL cannot be reached the > caller has no way of knowing if the URL was navigated successfully or not. To > solve this we make a head request to the URL we are navigating to and add a > callback to the API. > > It is important to note that we will not be caching the response of the head > request. While this could possibly give us better performance we have no way > of guaranteeing the server will still be up next time and everything may > fail. This is different from the gadget case where we have the gadget XML > cached on the server. > > > This addresses bug SHINDIG-1669. > https://issues.apache.org/jira/browse/SHINDIG-1669 > > > Diffs > ----- > > > http://svn.apache.org/repos/asf/shindig/trunk/content/samplecontainer/examples/embeddedexperiences/PhotoList.xml > 1211913 > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/container.util/util.js > 1211913 > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/container/container.js > 1211913 > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/embeddedexperiences/embedded_experiences_container.js > 1211913 > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/open-views/viewenhancements-container.js > 1211913 > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/test/javascript/features/container.url/container_url_test.js > 1211913 > > Diff: https://reviews.apache.org/r/3037/diff > > > Testing > ------- > > Tested in container as well as updating unit tests. > > > Thanks, > > Ryan > >
