FYI,
calligra-2.7.92 fails to build on arm platform (where qreal != double).
I think it comes from this commit:
https://projects.kde.org/projects/calligra/repository/revisions/3b54859a0514fc215cb9c46fc17acd13d2f700be
Please be careful mixing double and qreal types. Thanks.
I'll try to help work on a patch soon (attached is what I have so far,
untested).
-- Rex
diff -up calligra-2.7.92/krita/image/kis_convolution_painter.cc.qreal calligra-2.7.92/krita/image/kis_convolution_painter.cc
diff -up calligra-2.7.92/krita/image/kis_convolution_worker_spatial.h.qreal calligra-2.7.92/krita/image/kis_convolution_worker_spatial.h
--- calligra-2.7.92/krita/image/kis_convolution_worker_spatial.h.qreal 2014-02-09 04:18:42.000000000 -0600
+++ calligra-2.7.92/krita/image/kis_convolution_worker_spatial.h 2014-02-10 11:21:05.258576197 -0600
@@ -109,11 +109,11 @@ public:
this->m_progress->setProgress(0);
// Iterate over all pixels in our rect, create a cache of pixels around the current pixel and convolve them.
- m_pixelPtrCache = new double*[m_cacheSize];
- m_pixelPtrCacheCopy = new double*[m_cacheSize];
+ m_pixelPtrCache = new qreal*[m_cacheSize];
+ m_pixelPtrCacheCopy = new qreal*[m_cacheSize];
for (quint32 c = 0; c < m_cacheSize; ++c) {
- m_pixelPtrCache[c] = new double[channelCount];
- m_pixelPtrCacheCopy[c] = new double[channelCount];
+ m_pixelPtrCache[c] = new qreal[channelCount];
+ m_pixelPtrCacheCopy[c] = new qreal[channelCount];
}
// decide caching strategy
_______________________________________________
calligra-devel mailing list
calligra-devel@kde.org
https://mail.kde.org/mailman/listinfo/calligra-devel
