Review Request 124200: Introduction of rename variable refactoring

Wed Jul 1 13:06:30 UTC 2015

> On Cze 29, 2015, 4:27 po południu, Milian Wolff wrote:
> > refactoring/kdevrefactorings.cpp, line 103
> > <https://git.reviewboard.kde.org/r/124200/diff/1/?file=381613#file381613line103>
> >
> >     I know you have a FIXME above, but I just wanted to say that this is a _very_ bad idea. You essentially block the whole GUI now whenever a project gets opened. Also, the CMake manager already configures itself as required, so you shouldn't need to worry about this.
> 
> Maciej Poleski wrote:
>     The problem i have (with async) is that project is configured as i requested, but signal (`finished`) is emitted/dispatched *after* i request KDevelop to close (crashing the whole application at shutdown) and not triggering appropriate handler at time. I don't know why.
>     
>     If CMake manager does this as necessary - that's great. There is only one slight problem: i cannot connect to `IProjectBuilder::configured` - Qt says (at runtime), that CMakeBuilder doesn't have this signal (and strangely it looks like indeed there is no such signal (at least it is not listed here: https://projects.kde.org/projects/extragear/kdevelop/kdevelop/repository/revisions/master/entry/projectbuilders/cmakebuilder/cmakebuilder.h nor connected here: https://projects.kde.org/projects/extragear/kdevelop/kdevelop/repository/revisions/master/entry/projectbuilders/cmakebuilder/cmakebuilder.cpp#L82)) even though it is pure virtual in its base class. Could you explain to me what is happening there?
> 
> Milian Wolff wrote:
>     the issue you describe with finished not getting dispatched properly sounds like a bug. Can you create a simplified test case (unrelated to the work in this branch) and show it to us? Or at least show us the code you tried to write here. Then we could debug it together and find a way to fix it. Potentially, your code blocked and did not hand over work to the eventloop, thereby blocking the configure job from executing.
>     
>     And the signals in IProjectBuilder are severly broken, looking at it now. They should a) have the KDevelop:: QID in their arguments for old-style signal/slot connections to work properly (is that what you tried?), and b) most of them never get emitted if I'm not mistaken. So you'll have to fix this first (in separate patches and review requests!).

The issue with `finished` signal was my (very dumb) mistake.

Why KDevelop:: QID is required for old-style signal/slot connection? (and what is KDevelop::QID?). Yes i tried old-style connections.

- Maciej

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/124200/#review81865
-----------------------------------------------------------

On Cze 28, 2015, 12:10 rano, Maciej Poleski wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/124200/
> -----------------------------------------------------------
> 
> (Updated Cze 28, 2015, 12:10 rano)
> 
> 
> Review request for KDevelop.
> 
> 
> Repository: kdev-clang
> 
> 
> Description
> -------
> 
> Introduced rename variable refactoring.
> 
> This patch provides preliminary support for variable rename refactoring.
> Currently only one project is supported.
> Currently it is required to save all opened documents (cache is not ready).
> Regeneration of compile_commands.json needs to be polished. How to get appropriate notification (IProjectBuilder::configured signal)? As a workaround i simply restarted KDevelop to get this regenerated.
> To use this refactoring right click on variable declaration (definition) and select "rename" refactoring.
> 
> 
> Full renaming support will consist of a few parts - one of them is renaming around VarDecl (this patch).
> Quite early, but i wanted to see that this work is going into some direction after month of coding of pure interfaces/details.
> 
> How to split diffs? I noticed that RBTool created diff consisting of all changes ignoring other pending request. I wanted only the last one, because earlier commits are subject of other review.
> 
> 
> Diffs
> -----
> 
>   .reviewboardrc cd1ef2aa08623d813c80b5c5275b2ad0e35ff40e 
>   CMakeLists.txt 875172a8407f4bd9faf330f032a280fa66c2b16f 
>   clangsupport.h 8ed1ec90bbbc41d7c7a94d926e0951c729a6194c 
>   clangsupport.cpp e22c55426a2e839ec11cbe0b2fe1e13721b0583a 
>   cmake/FindClang.cmake 6c9bd6ef0242319122dcc7e6fd6cea8d9f0cbfbb 
>   refactoring/CMakeLists.txt PRE-CREATION 
>   refactoring/documentcache.h PRE-CREATION 
>   refactoring/documentcache.cpp PRE-CREATION 
>   refactoring/interface.h PRE-CREATION 
>   refactoring/interface.cpp PRE-CREATION 
>   refactoring/kdevrefactorings.h PRE-CREATION 
>   refactoring/kdevrefactorings.cpp PRE-CREATION 
>   refactoring/kdevrefactorings_disabled.h PRE-CREATION 
>   refactoring/nooprefactoring.h PRE-CREATION 
>   refactoring/nooprefactoring.cpp PRE-CREATION 
>   refactoring/refactoring.h PRE-CREATION 
>   refactoring/refactoring.cpp PRE-CREATION 
>   refactoring/refactoringinfo.h PRE-CREATION 
>   refactoring/refactoringinfo.cpp PRE-CREATION 
>   refactoring/renamevardeclrefactoring.h PRE-CREATION 
>   refactoring/renamevardeclrefactoring.cpp PRE-CREATION 
>   refactoring/utils.h PRE-CREATION 
>   refactoring/utils.cpp PRE-CREATION 
> 
> Diff: https://git.reviewboard.kde.org/r/124200/diff/
> 
> 
> Testing
> -------
> 
> Manual testing on very small code bases as cache and compile_commands.json regeneration is not fully functional.
> 
> 
> Thanks,
> 
> Maciej Poleski
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20150701/0a555bde/attachment.html>