Review Request: Add possibility to force DeclarationIds to be always direct

[David Nolden]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/100875/#review2065
-----------------------------------------------------------


Normally, when changing any of the *Data* classes, it also means that the value returned in staticItemRepositoryVersion() in itemrepository.cpp needs to be increased, because the disk format has changed. In this case, not increasing it will "only" result in random values of the added boolean though, as the size of the structure stayed equal.

- David


On March 18, 2011, 11:02 a.m., Sven Brauch wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/100875/
> -----------------------------------------------------------
> 
> (Updated March 18, 2011, 11:02 a.m.)
> 
> 
> Review request for KDevelop.
> 
> 
> Summary
> -------
> 
> As suggested by David on the mailing list, this patch adds the possibility to force a DeclarationId to be always "direct", making correct reporting of uses possible for contexts which are imported into the InternalContext of a newly created Declaration, as required for the python language plugin (at least I couldn't find another working solution). It adds a 1-bit boolean to the DeclarationData class, hope that's not a problem.
> 
> 
> Diffs
> -----
> 
>   language/duchain/declaration.h 77abb95 
>   language/duchain/declaration.cpp ce70ff3 
>   language/duchain/declarationdata.h f02a9b1 
> 
> Diff: http://git.reviewboard.kde.org/r/100875/diff
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Sven
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20110320/361f0b58/attachment.html>