Where does the error message live? This seems like the perfect time to make it a bit more helpful, even if it just suggests to people that they check for both browser and server caching.
2009/9/21 Miguel Méndez <[email protected]> > I've also run into this problem... > > > On Fri, Sep 18, 2009 at 8:37 AM, Marko Vuksanovic < > [email protected]> wrote: > >> >> I had the same problem.. After struggling for 2 days I figured out >> that the embedded jetty loads the hosted.html file form cache (C: >> \Documents and Settings\YOUR_USERNAME\Local Settings\Temporary >> Internet Files\Content.IE5 - on windows xp) - after manually deleting >> that cache everything started working normally. >> >> Hope this saves somebody else at least 2 days :) >> >> On Aug 12, 5:52 pm, Ray Ryan <[email protected]> wrote: >> > We could make the error message a bit more helpful. If I'm following the >> > conversation correctly I think I've bumped into this due to browser >> caching. >> > Suggesting that possibility in the error message, or offering a link to >> > force a reload in a browser busting way (e.g. adding a query param wtih >> a >> > date stamp value) could save folks a lot of time. >> > >> > >> > >> > On Tue, Aug 11, 2009 at 9:33 PM, Fred Sauer <[email protected]> wrote: >> > >> > > On Tue, Aug 11, 2009 at 5:47 PM, John Tamplin <[email protected]> >> wrote: >> > >> > >> On Tue, Aug 11, 2009 at 8:22 PM, Fred Sauer <[email protected]> >> wrote: >> > >> > >>> I'm not sure I understand the issue. If you have different versions >> of >> > >>>> hosted.html and GWT, things are likely to not work and that is why >> that >> > >>>> check was added in 1.6. I don't see how you could have the >> incorrect >> > >>>> contents of hosted.html but the correct version, unless you >> update/rollback >> > >>>> parts of GWT separately, in which case you are already likely to >> break >> > >>>> things if you don't know what you are doing. >> > >> > >>> Yep, that's exactly the scenario :). Knowing that you have to watch >> out >> > >>> for this stuff is one thing. Being bit by it every now and again and >> going >> > >>> down a rabbit hole, is (mildly) annoying. >> > >> > >>> Not a huge deal as this affect contributors only, but could be a >> good >> > >>> sanity check to save time down the road. >> > >> > >> There are so many ways you can screw things up by mixing different >> > >> versions of parts of GWT, I don't see how this is any different. For >> > >> example, if you rollback an old TypeOracleMediator but don't roll >> back >> > >> related TypeOracle changes, things are going to break horribly. I >> think >> > >> trying to add code inside GWT to detect such situations is >> counterproductive >> > >> and unlikely to be effective anyway. >> > >> > > You had me at "There are so many ways you can screw things up" >> > >> > > Thanks >> > > Fred >> > >> > >> I would hope that anyone knows building a version of GWT that is not >> at a >> > >> consistent revision across the board means they better know exactly >> what >> > >> they are doing or they will get weird breakages. >> > >> > >> -- >> > >> John A. Tamplin >> > >> Software Engineer (GWT), Google >> > >> > > -- >> > > Fred Sauer >> > > Developer Advocate >> > > Google Inc. 1600 Amphitheatre Parkway >> > > Mountain View, CA 94043 >> > > [email protected] >> >> >> > > > -- > Miguel > > > > > --~--~---------~--~----~------------~-------~--~----~ http://groups.google.com/group/Google-Web-Toolkit-Contributors -~----------~----~----~----~------~----~------~--~---
