First of all, thanks you guys who working of DXR. It is my major tool for studying the Gecko code base and it helps a lot. I would like to have DXR lists all the implementation of a virtual function. This is quite handy when you are studying an interface which has more than one level of inheritance hierarchy.
Best Regards, Shih-Chiang Chien Mozilla Taiwan On Oct 26, 2013, at 3:25 AM, Zack Weinberg <za...@panix.com> wrote: > On 2013-10-25 11:10 AM, Ehsan Akhgari wrote: >> On 2013-10-25 10:45 AM, Erik Rose wrote: >>> ... would it be better, from your point of view, to >>> index the generated files or to magically turn up the IDL line >>> "attribute short foo" when you search for "function:GetFoo" or >>> "function:SetFoo"? (I'm not sure both are feasible; I just want to get >>> an early read.) >> >> The latter is better in my opinion! > > If I can have only one, I agree the latter is better -- but I want to mention > that sometimes I've needed to see what the I(P)DL compiler has produced from > a particular definition file in order to fix a bug. For instance, a few > years ago -- this may have been fixed since -- there were many gaps in the > documentation of what C++ types corresponded to IDL types and vice versa, so > one had to look at all three of the IDL, the generated .h, and the > hand-written .cpp that expected to implement an interface to understand what > was *really* going on. > > So, if you can make DXR do both, that would be better. (With a clear > indication that certain files are generated, and where the real source is, if > at all possible.) > > zw > _______________________________________________ > dev-platform mailing list > dev-platform@lists.mozilla.org > https://lists.mozilla.org/listinfo/dev-platform _______________________________________________ dev-platform mailing list dev-platform@lists.mozilla.org https://lists.mozilla.org/listinfo/dev-platform
