Review Request: Offer create private/protected members when possible in missingdeclarationaction

[Jonas Jacobi]

> On Dec. 12, 2011, 9:10 p.m., Aleix Pol Gonzalez wrote:
> > Hi, sorry that it took us so long to answer.
> > 
> > I'm not proficient in this code but: isn't it too much code if we're just adding different visibility types?
> > Maybe you can find a cleaner solution?
> > 
> > Anyway, Milian or David will know better than I would.

mhh it's like 20 lines of code and you have to determine what visibility types are applicable, so i don't see it getting much better. although i have to admit that i was (am) absolutely new to the current kdevelop/duchain when i wrote this, so maybe i missed something.


- Jonas


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/102625/#review8899
-----------------------------------------------------------


On Sept. 17, 2011, 8:28 a.m., Jonas Jacobi wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/102625/
> -----------------------------------------------------------
> 
> (Updated Sept. 17, 2011, 8:28 a.m.)
> 
> 
> Review request for KDevelop.
> 
> 
> Description
> -------
> 
> Currently only public member variables can be declared in some contexts.
> This patch checks whether private/protected members would be also accessible from the current context (disregarding friend declarations atm)
> and if so, offer the possibility to create the members.
> e.g.
> 
> class A {
> public:
>   A(int x) : mUndeclared(x) {} // <- declare mUndeclared as private/protected/public member
> };
> 
> class B : public A {
> public:
>   void foo() {
>     A a(2);
>     a.undeclared = 5; // <- declare undeclared as protected/public
>   }
> };
> 
> class X {
> public:
>   void bar() {
>     B b;
>     b.undeclared = 5; // <- declare undeclared as public
>   }
> };
> 
> 
> Diffs
> -----
> 
>   languages/cpp/cppduchain/missingdeclarationassistant.cpp bafc159 
> 
> Diff: http://git.reviewboard.kde.org/r/102625/diff/diff
> 
> 
> Testing
> -------
> 
> I'm currently using it
> 
> 
> Thanks,
> 
> Jonas Jacobi
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20111212/fc71d00f/attachment.html>