I don't think it's an issue that blocks since people are most likely to use SDM in uncompiled mode, and the 'error' can always be suppressed and turned into a warning. At best, it might inhibit a const optimization in Closure where it sees the same variable declared/assigned twice I think.
On Tue, Jun 9, 2015 at 10:27 AM, 'Roberto Lublinerman' via GWT Contributors <[email protected]> wrote: > SDM expects that classes/interfaces get generated as a whole either all of > it or none. As you know, class literals are not really part of the class > but they are field in a separate synthetic class). Even SDM (with all > optimizations off, no pruning at UnifyAST) if a reference to the class > literal is not seen when generating the JS for the class then its class > literal field wont be part of the JS for the class. > > For SDM it could only happen for interfaces (and maybe primitive types) > because GwtAstBuilder will synthezise a getClass for each class containing > reference to its class literals. > > The problem is further obscured because we (re)generate an epilogue each > time and it contains all the class literals that are referenced from the > code that is being recompiled whose classes where not part of the current > compile. This serves a purpose in NON SDM where if only the class literal > for a class is referenced, the class can be pruned. > > The only glitch remaining is that in SDM there might be duplicated class > literals (but it should not affect the correctness). Is this an issue for > Closure? > > On Tue, Jun 9, 2015 at 9:48 AM, 'Ray Cromwell' via GWT Contributors < > [email protected]> wrote: > >> Chris, >> My change actually fixed a bug which may have obscured a problem. There >> were duplicate class literals being generated. So you'd get >> InterfaceFoo.class twice in the output, and it may be in SDM you'd be >> covered, but in regular compiled mode you'd get the duplicates which were >> causing closure compiler issues. >> >> Roberto, Maybe this is an issue with >> ControlFlowRecorder/RapidTypeAnalyzer and interface literals? >> >> >> On Tue, Jun 9, 2015 at 9:43 AM, 'Chris DiGiano' via GWT Contributors < >> [email protected]> wrote: >> >>> @Ray, rolling back your change indeed fixed the problem ( >>> https://gwt-review.googlesource.com/#/c/12311/). How do you recommend >>> we proceed? Anything I can do to help? >>> >>> @Roberto, thanks for your suggestion, but I had already tried restarting >>> the code server and clearing the cache. This did not fix things. >>> >>> Chris >>> >>> On Mon, Jun 8, 2015 at 6:55 PM 'Roberto Lublinerman' via GWT >>> Contributors <[email protected]> wrote: >>> >>>> This might be due to the way we handle class literals. Class literals >>>> for interfaces if not referenced during the initial compile might cause >>>> that error. The error should go aways if you restart SDM. >>>> >>>> The offending sequence is >>>> >>>> 0) Suppose initially you have (interface A, class B and class C) and >>>> start SDM. >>>> 1) Now edit class B to add a reference to A.class and recompile with >>>> SDM (do not touch A at this time nor anything that makes it recompile) . >>>> This should work fine because the literals that are referenced but whose >>>> classes where not part of this compile will be generated in the epilogue. >>>> 2) Now edit class C in any way that does not make B or A recompile. >>>> Here you will have an error like the one you described. >>>> >>>> If you restart SDM at this point it should work. >>>> >>>> I'll get a fix for this tomorrow. >>>> >>>> >>>> >>>> >>>> On Mon, Jun 8, 2015 at 3:23 PM, 'Ray Cromwell' via GWT Contributors < >>>> [email protected]> wrote: >>>> >>>>> Ooops, wrong pointer (for external users) This one >>>>> https://gwt-review.googlesource.com/#/c/12311/ >>>>> >>>>> >>>>> On Mon, Jun 8, 2015 at 3:14 PM, Ray Cromwell <[email protected]> >>>>> wrote: >>>>> >>>>>> Try rolling back this CL and see if it fixes it ( >>>>>> https://critique.corp.google.com/#review/92873682/depot/google3/third_party/java_src/gwt/svn/trunk/dev/core/src/com/google/gwt/dev/jjs/impl/GenerateJavaScriptAST.java >>>>>> ) >>>>>> >>>>>> >>>>>> On Mon, Jun 8, 2015 at 3:04 PM, Chris DiGiano <[email protected]> >>>>>> wrote: >>>>>> >>>>>>> I'm having trouble referencing a JsType interface when running under >>>>>>> superdevmode. I'm trying to pass the JsType as a class reference to a >>>>>>> method that uses the class to coerce the results into the expected type: >>>>>>> >>>>>>> myelement.getCustomStampedElement("dialog", >>>>>>> PolymerDialog.class).open(); >>>>>>> >>>>>>> Unfortunately, in superdevmode (and only in SDM) I get this error: >>>>>>> >>>>>>> Uncaught ReferenceError: >>>>>>> Lcom_google_ccc_groups_rosters_frontend_polymer_jstype_PolymerDialog_2_classLit_0_g$ >>>>>>> is not defined >>>>>>> >>>>>>> PolymerDialog is declared like this: >>>>>>> >>>>>>> @JsType >>>>>>> public interface PolymerDialog { >>>>>>> void open(); >>>>>>> } >>>>>>> >>>>>>> and getCustomStampedElement is defined like this: >>>>>>> >>>>>>> /** >>>>>>> * Returns the element specified in this custom element's >>>>>>> template that matches the given id >>>>>>> * with the result coerced to a non-vanilla element type. >>>>>>> Equivalent to Polymer's $ function. >>>>>>> * >>>>>>> * @see >>>>>>> * <a href=" >>>>>>> https://www.polymer-project.org/1.0/docs/devguide/local-dom.html#node-finding >>>>>>> "> >>>>>>> * Polymer automatic node finding</a> >>>>>>> */ >>>>>>> <T> T getCustomStampedElement(String id, Class<T> type); >>>>>>> >>>>>>> It appears that the PolymerDialog JsType is being left out of the >>>>>>> list of class literals available in superdevmode. Is this a bug? Any >>>>>>> ideas >>>>>>> for workarounds? >>>>>>> >>>>>>> Chris DiGiano >>>>>>> >>>>>>> -- >>>>>>> You received this message because you are subscribed to the Google >>>>>>> Groups "GWT Contributors" group. >>>>>>> To unsubscribe from this group and stop receiving emails from it, >>>>>>> send an email to >>>>>>> [email protected]. >>>>>>> To view this discussion on the web visit >>>>>>> https://groups.google.com/d/msgid/google-web-toolkit-contributors/9826758e-1b0f-4797-8902-738a1fade037%40googlegroups.com >>>>>>> <https://groups.google.com/d/msgid/google-web-toolkit-contributors/9826758e-1b0f-4797-8902-738a1fade037%40googlegroups.com?utm_medium=email&utm_source=footer> >>>>>>> . >>>>>>> For more options, visit https://groups.google.com/d/optout. >>>>>>> >>>>>> >>>>>> >>>>> -- >>>>> You received this message because you are subscribed to the Google >>>>> Groups "GWT Contributors" group. >>>>> To unsubscribe from this group and stop receiving emails from it, send >>>>> an email to >>>>> [email protected]. >>>>> To view this discussion on the web visit >>>>> https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAPVRV7fZhv0AkiPYqfM2WWaW8zFvdgk-BYEucwRP8k4ewXEp6w%40mail.gmail.com >>>>> <https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAPVRV7fZhv0AkiPYqfM2WWaW8zFvdgk-BYEucwRP8k4ewXEp6w%40mail.gmail.com?utm_medium=email&utm_source=footer> >>>>> . >>>>> >>>>> For more options, visit https://groups.google.com/d/optout. >>>>> >>>> >>>> -- >>>> You received this message because you are subscribed to the Google >>>> Groups "GWT Contributors" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to >>>> [email protected]. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAC7T7gk_U85%2B6z1QK9XpAKEXbBJgviBfZ1j0kgE8_pYZsV6Y3g%40mail.gmail.com >>>> <https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAC7T7gk_U85%2B6z1QK9XpAKEXbBJgviBfZ1j0kgE8_pYZsV6Y3g%40mail.gmail.com?utm_medium=email&utm_source=footer> >>>> . >>>> For more options, visit https://groups.google.com/d/optout. >>>> >>> -- >>> You received this message because you are subscribed to the Google >>> Groups "GWT Contributors" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected] >>> . >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAHbZCdNJ6mBv%2BRwso0z_OSNF7531b9rSOgLE3zJ%3DvB7FxkFD0A%40mail.gmail.com >>> <https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAHbZCdNJ6mBv%2BRwso0z_OSNF7531b9rSOgLE3zJ%3DvB7FxkFD0A%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> -- >> You received this message because you are subscribed to the Google Groups >> "GWT Contributors" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAPVRV7cU1HQb6Ns2U967CxuomGHvxxGt6YtmJxaOea6n8X_poQ%40mail.gmail.com >> <https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAPVRV7cU1HQb6Ns2U967CxuomGHvxxGt6YtmJxaOea6n8X_poQ%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > -- > You received this message because you are subscribed to the Google Groups > "GWT Contributors" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAC7T7gmQh2YnD7nUcJ8UtEn7eAaWik1Ya%2BmwDEQNz-De%2Bm4UVQ%40mail.gmail.com > <https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAC7T7gmQh2YnD7nUcJ8UtEn7eAaWik1Ya%2BmwDEQNz-De%2Bm4UVQ%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "GWT Contributors" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAPVRV7ePAvWoXju-%3DOpm9F9%2B-yGWFRm9B_zMcHLzrjeES%3D0xFQ%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
