This bug report was migrated from our old Bugzilla tracker.

Reported in version: 2.0.4
Reported for operating system, platform: Linux, x86_64

Comments on the original bug report:

On 2018-06-20 06:48:54 +0000, Adam Bryant wrote:

In this sample code:

#include
#include <SDL2/SDL.h>
#include

int main() {
SDL_Window *window = NULL;
SDL_Renderer *renderer = NULL;
SDL_CreateWindowAndRenderer(1, 1, 0, &window, &renderer);

SDL_Texture *texture = SDL_CreateTexture(renderer, SDL_PIXELFORMAT_RGBA8888, SDL_TEXTUREACCESS_STATIC, 1, 1);
SDL_SetRenderTarget(renderer, texture);
SDL_Texture *renderTexture = SDL_GetRenderTarget(renderer);

if (renderTexture != texture) {
    throw std::runtime_error("What?");
}
std::this_thread::sleep_for(std::chrono::seconds(1));
return 0;

}

the exception is thrown.

Is SDL_GetRenderTarget returning a different address expected behavior?

On 2018-06-22 17:45:29 +0000, Sam Lantinga wrote:

Yes, this is expected if you're using a texture format that isn't supported by the renderer.

Try using SDL_PIXELFORMAT_ARGB8888 instead if you don't want texture conversion.

On 2018-06-22 21:17:25 +0000, Adam Bryant wrote:

(In reply to Sam Lantinga from comment # 1)

Yes, this is expected if you're using a texture format that isn't supported
by the renderer.

Try using SDL_PIXELFORMAT_ARGB8888 instead if you don't want texture
conversion.

Thanks, it works now.
I changed my code around to select a supported format.

On 2019-06-10 18:44:46 +0000, Martin Vee wrote:

This behaviour should be described in the wiki, as it's a bit unexpected to have a "getter" returning a different value than the "setter".

Is it possible to query a texture->native pointer to get it's parent texture pointer?