Review Request 128880: [OS X] use a different tab bar widget for tabbed documents

[Aleix Pol Gonzalez]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/128880/#review99117
-----------------------------------------------------------




sublime/container.cpp (line 61)
<https://git.reviewboard.kde.org/r/128880/#comment66727>

    the static declaration can be placed here. Also it should probably be a QPointer.



sublime/container.cpp (line 65)
<https://git.reviewboard.kde.org/r/128880/#comment66728>

    I'm not super fond of using fusion randomly here. Why does it work for Kate and not for us?


- Aleix Pol Gonzalez


On Sept. 12, 2016, 1:20 a.m., René J.V. Bertin wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/128880/
> -----------------------------------------------------------
> 
> (Updated Sept. 12, 2016, 1:20 a.m.)
> 
> 
> Review request for KDE Software on Mac OS X and KDevelop.
> 
> 
> Bugs: 363473
>     http://bugs.kde.org/show_bug.cgi?id=363473
> 
> 
> Repository: kdevplatform
> 
> 
> Description
> -------
> 
> This is a potential fix for an issue raised on BKO: https://bugs.kde.org/show_bug.cgi?id=363473
> 
> It's also the most complete/implementation:
> - applies only when the Macintosh widget style is being used
> - if so, creates a QStyle object for the Fusion widget style
> - when successful, sets the `Sublime::ContainerTabBar` to use that style
> 
> This solves all issues stemming from Qt's use of a "native" widget that is intended only for use in dialogs and not in tabbed document interfaces.
> 
> In my testing, the `ContainerTabBar` ctor is called only rarely, apparently only when changing views (e.g. code -> patch review and back again, or code -> debug). If that observation is correct, use of a global `qTabBarStyle` variable is justified (but more elegant solutions might exist). This observation also justifies (IMHO) the check for the active application style rather than using an `#ifdef Q_OS_OSX` or even applying the fix across all platforms and application styles. That is certainly a possibility that doesn't lead to any shocking style mismatches in my eyes. It does cause some loss of compactness when using my QtCurve settings, which is why I added the style check; a small cost as a gesture to users of a highly configurable style.
> 
> There is still some weirdness behind the tabs which looks like a misaligned well or frame. I'd love to get that right too.
> 
> 
> Diffs
> -----
> 
>   sublime/container.cpp b04f6c3 
> 
> Diff: https://git.reviewboard.kde.org/r/128880/diff/
> 
> 
> Testing
> -------
> 
> See https://bugsfiles.kde.org/attachment.cgi?id=99160 (unpatched) and the last series of screenshots attached to the ticket on BKO. They show the fix applied to various styles on OS X.
> 
> 
> Thanks,
> 
> René J.V. Bertin
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20160911/3d5da105/attachment.html>