Review Request 127380: More conservative approach to saving some disk accesses

[David Faure]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/127380/#review93756
-----------------------------------------------------------




autotests/kiconloader_benchmark.cpp (line 82)
<https://git.reviewboard.kde.org/r/127380/#comment63909>

    Try weird sizes if you want to see a possible behaviour difference between MatchExact and MatchBest :-)
    
    But even then you won't be able to detect if KIconLoader took a bigger icon and scaled it down, or took a small icon and scaled it up.
    
    Antonio Larrosa might be able to help remind us about the reasoning and corner cases of this code.


- David Faure


On March 15, 2016, 1:52 a.m., Aleix Pol Gonzalez wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/127380/
> -----------------------------------------------------------
> 
> (Updated March 15, 2016, 1:52 a.m.)
> 
> 
> Review request for KDE Frameworks and Christoph Feck.
> 
> 
> Repository: kiconthemes
> 
> 
> Description
> -------
> 
> My previous approach to KIconThemes felt like a dead end, I decided I'll take a more conservative approach. Here's a first step.
> 
> 
> Diffs
> -----
> 
>   autotests/CMakeLists.txt 61e81f6 
>   autotests/kiconloader_benchmark.cpp PRE-CREATION 
>   src/kiconloader.cpp 75ab482 
>   src/kicontheme.cpp 0996054 
> 
> Diff: https://git.reviewboard.kde.org/r/127380/diff/
> 
> 
> Testing
> -------
> 
> Attempts to call `KIconThemeDir::iconPath` on the benchmark: 4318 after/5040 before= 17% less QFile::exists
> 
> 
> Thanks,
> 
> Aleix Pol Gonzalez
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-frameworks-devel/attachments/20160320/d4c34d40/attachment-0001.html>