I was not remembering correctly and my last comment was off. I am not using a create() method. So the editorial correction is
s/and have a create() method instead of ordinal() that generates the object//g On Sun, Jun 25, 2017 at 6:23 PM, Vassilis Virvilis <vasv...@gmail.com> wrote: > Hm... > > For me this d3 ordinal thing like implementing the () operator in c++. Is > it a function with members and methods or is it an object with the ability > to be called as a function? > > Anyway if you and Roberto say it is not a bug then it's not a bug. > > When I revisit my D3 bindings I will explore your proposals - but AFAIK > there is no easy solution that resembles the original javascript API. > Currently I prefer to copy paste among the inheritance pyramid (there are 2 > or 3 subtypes) and have a create() method instead of ordinal() that > generates the object. > > If there is no objection I will post another stumbling block I encountered > in my jsinterop journey. > > > On Sun, Jun 25, 2017 at 5:35 PM, Thomas Broyer <t.bro...@gmail.com> wrote: > >> >> >> On Sunday, June 25, 2017 at 1:24:40 PM UTC+2, Vassilis Virvilis wrote: >>> >>> Ηι, >>> >>> thanks for answering >>> >>> d3.ordinal does not exist as you said. However d3.scale.ordinal exists >>> and that part works great. >>> >> >> But it is not a type, so you cannot use 'instanceof' on it. >> >> >>> in d3 ordinal is both a function ordinal() which returns an ordinal >>> instance and an object with more methods such as range() which is the crux >>> of my issue. >>> >>> The problem there is that >>> >>> - public native <I, O> O range(I[] values); *// this doesn't >>> work* >>> - // public native <I> Ordinal range(I[] values); *// this works* >>> >>> I would like to use the first definition because it allows reusing some >>> code between different scales. Otherwise I have to copy-paste the range() >>> method multiple times instead of put it in a base class. >>> >> >> In Java, generics are more or less syntactic sugar for using the erased >> type and casting values. So when using the first one, GWT will, when used >> with O==Ordinal, generate a cast from Object to Ordinal, which IIUC is >> what's causing the error, because your JsType is not a type that can be >> checked (a cast is translated in JS to a type-check using 'instanceof'). >> >> >>> The problem is on the exceptions generated by the attempting casting. >>> >>> As you said this may not be a bug but not for the reason you mentioned. >>> >>> So the question remains: Should I report it as an issue? >>> >> >> I would say no. Your @JsType is wrong. You should either use an interface >> with @JsType(isNative=true, name="?", namespace=JsPackage.GLOBAL), or if >> you want to use fields use a class with @JsType(isNative=true, >> name="Object", namespace=JsPackage.GLOBAL) >> >> -- >> 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 google-web-toolkit-contributors+unsubscr...@googlegroups.com. >> To view this discussion on the web visit https://groups.google.com/d/ms >> gid/google-web-toolkit-contributors/6206fb29-d3c1-48de-ab04- >> 0610cbb3c2bd%40googlegroups.com >> <https://groups.google.com/d/msgid/google-web-toolkit-contributors/6206fb29-d3c1-48de-ab04-0610cbb3c2bd%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > > > -- > Vassilis Virvilis > -- Vassilis Virvilis -- 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 google-web-toolkit-contributors+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/google-web-toolkit-contributors/CAKbOjEz_uocutZGGp%3DsvDbUSf1eW9gnziGsCtVOkvd1iGRT%2Bjg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
