| Re: [cdt-dev] visibility of typedefs/type aliases |
|
> > I agree that typedefs require special treatment. Will you be willing to
> > give it a shot? > I've been shooting already ;-) And that's why I was asking. The question is > whether the code stated below is required, or if it is superfluous. > My first approach was to remove the check altogether, and in fact the typedefs > started to work. But I don't know what might break by this change. First I > thought it might be forward declarations but then I realized that those cannot > be subject of this check. Do you have any hints what this check is good for > (comments in the code would be a boon ;-)? I don't know about this piece of code in particular, but a good way to assess whether a change breaks something is to run relevant test suites and see if the change causes any failures. For changes to the parser/indexer/semantics code, I typically run DOMParserTestSuite and IndexTests. Another idea is to run the codan tests (I don't recall the name of the test suite at the moment). Regards, Nate |