Thanks for offering your help Thomas. I had tried cleaning everything before with no success.
We're using a Tokenizer Factory, and noticed that if we commented out the last Place tokenizer method we added (in one of the branches merged) and if we created the Tokenizer statically within the Place object instead so that we add it to our HistoryMapper using @WithTokenizers, then we can compile and the app works fine. I did debug the gwt code and was able to see that within PlaceHistoryMapperGenerator.java, the prefix comes as null for the second Place we have in our tokenizer factory (which is not the last one we added, but it always fails on this one, or at least this is the first one that comes as null). Other properties are not null however, such as placeTypeName, which also comes from the PlaceHistoryGeneratorContext. As another data point, all methods in the factory have a @Prefix defined, but I know at least one of the tokenizers provided in @WithTokenizers does not. Since we have a workaround, I probably won't dig any further, but I'm wondering if there's somehow a limit to the number of Places we can have through a Tokenizer Factory. Previous to adding the last place that gave us problems, we had 54 there, plus 6 on the @WithTokenizers annotation. I know there's nothing bad or special about the last place added, because I tried adding a completely different one and I got the same issue. I don't see a quick way to provide a repro case, as I've only seen this issue after adding more than 60 places and with a custom tokenizer factory, otherwise I'd provide one, or if I find myself with extra time... I've already spent more time on this than I wanted ;-( On Tuesday, May 6, 2014 8:40:00 AM UTC-4, Thomas Broyer wrote: > > Things to try: clean your project (all outputs) and remove the > .gwt-unitCache. > Also, set a breakpoint where the NPE is thrown to try to find what can be > the cause. > If you can provide me a repro case, I could debug myself if you prefer. > > On Tuesday, May 6, 2014 1:55:28 PM UTC+2, Rogelio Flores wrote: >> >> After a big git merge with conflicting files, our project doesn't >> gwt-compile anymore. It gives this error which I don't remember seeing >> before: >> >> [INFO] Computing all possible rebind results for >> 'com.mycomp.HistoryMapper' >> [INFO] Rebinding com.mycomp.HistoryMapper >> [INFO] Invoking generator >> com.google.gwt.place.rebind.PlaceHistoryMapperGenerator >> [INFO] [ERROR] Generator >> 'com.google.gwt.place.rebind.PlaceHistoryMapperGenerator' threw an >> exception while rebinding 'com.mycomp.HistoryMapper' >> [INFO] java.lang.NullPointerException >> [INFO] at >> com.google.gwt.place.rebind.PlaceHistoryMapperGenerator.writeGetPrefixAndToken(PlaceHistoryMapperGenerator.java:122) >> >> I haven't found any related info here or elsewhere online. There's the >> possibility of course that one or more of our files were corrupted during >> the merge, but there's nothing obvious that is wrong since there's no >> errors highlighted in the IDE. Regular (non-gwt) compile works, and I >> manually checked all conflicting files. The only related reported issue >> I've found is this one: >> >> https://code.google.com/p/google-web-toolkit/issues/detail?id=8036 >> >> but I don't think that issue fails during gwt compilation. Given the >> class where the NPE occurs, I suspect it has something to do with my >> Places/Tokenizers, but turning on DEGUG logLevel doesn't provide more info. >> Does anyone have a clue seeing the above snippet? >> >> >> -- You received this message because you are subscribed to the Google Groups "Google Web Toolkit" group. To unsubscribe from this group and stop receiving emails from it, send an email to google-web-toolkit+unsubscr...@googlegroups.com. To post to this group, send email to google-web-toolkit@googlegroups.com. Visit this group at http://groups.google.com/group/google-web-toolkit. For more options, visit https://groups.google.com/d/optout.
