> On Nov. 6, 2012, 7:27 a.m., Matt Franklin wrote: > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/container/container.js, > > line 472 > > <https://reviews.apache.org/r/7857/diff/1/?file=185211#file185211line472> > > > > This enforces a contract for the PRELOAD_TOKENS property of the preload > > data where the token response MUST have a TTL if a token exists for a > > particular gadget. > > > > If this is the intent, that is fine. If not, it might be better to > > fall back to the metadata response TTL in the event that a token doesn't > > have a TTL.
I think the implied contract is fine with PRELOAD_TOKENS because the format of the preloaded data is that of a TokenResponse, which by definition has a TTL attribute. - Stanton ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: https://reviews.apache.org/r/7857/#review13129 ----------------------------------------------------------- On Nov. 4, 2012, 3:40 p.m., Stanton Sievers wrote: > > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > https://reviews.apache.org/r/7857/ > ----------------------------------------------------------- > > (Updated Nov. 4, 2012, 3:40 p.m.) > > > Review request for shindig, Dan Dumont and Matt Franklin. > > > Description > ------- > > From the Jira: > When using the "preloadMetadatas" and "preloadTokens" optional config when > constructing an osapi.container.Container, the token refresh schedule will be > based on the TTL of whatever is in "preloadMetadatas" and not "preloadTokens". > > > This addresses bug SHINDIG-1880. > https://issues.apache.org/jira/browse/SHINDIG-1880 > > > Diffs > ----- > > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/container/container.js > 1405426 > > http://svn.apache.org/repos/asf/shindig/trunk/features/src/test/javascript/features/container/container_test.js > 1405426 > > Diff: https://reviews.apache.org/r/7857/diff/ > > > Testing > ------- > > Wrote some new JSUnits to cover this use case. These are included in the > patch. Some of the "window.setTimeout" code in the JSUnit is odd, but it's > the only way I could get it to not puke with reference errors saying that > "setTimeout" is undefined when it is called in > osapi.container.Container.scheduleRefreshTokens_. > > > Thanks, > > Stanton Sievers > >
