[core] For binary image data use uint8_t, not char

1 files changed, 4 insertions, 4 deletions

diff --git a/include/mbgl/util/image.hpp b/include/mbgl/util/image.hpp
index f58c2c0989..fc4078501e 100644
--- a/include/mbgl/util/image.hpp
+++ b/include/mbgl/util/image.hpp
@@ -7,14 +7,14 @@
 namespace mbgl {
 namespace util {
 
-std::string compress_png(int width, int height, const void *rgba);
+std::string compress_png(size_t width, size_t height, const uint8_t* rgba);
 
 
 class Image {
 public:
-    explicit Image(const std::string &img);
+    explicit Image(const std::string& img);
 
-    inline const char *getData() const { return img.get(); }
+    inline const uint8_t* getData() const { return img.get(); }
     inline uint32_t getWidth() const { return width; }
     inline uint32_t getHeight() const { return height; }
     inline operator bool() const { return img && width && height; }
@@ -24,7 +24,7 @@ private:
     uint32_t width = 0, height = 0;
 
     // the raw image data
-    std::unique_ptr<char[]> img;
+    std::unique_ptr<uint8_t[]> img;
 };