RGBA_blend

#define UNPREMULTIPLY_ALPHA(x, y) ((((x) << 16) - ((x) << 9) + (x)) / ((((x) + (y)) << 8) - ((x) + (y)) - (y) * (x)))
#define FAST_DIV255(x) ((((x) + 128) * 257) >> 16)

int RtmpPublisher_putVideoData(RtmpPublisher* rp, uint8_t* data, int dataSize) {

//blend RGBA
    if (rp->watermarkEnable)
    {
        if( rp->overlayX > rp->dst_width || rp->overlayY > rp->dst_height ||
           rp->mPngRGBA->width > rp->dst_width || rp->mPngRGBA->height > rp->dst_height)
        {
        }else{
            //this mirror
            int startline = rp->dst_width * rp->overlayY * 4;
            int startinline = rp->overlayX * 4;
            int overlayWidth = rp->mPngRGBA->width < rp->dst_width ? rp->mPngRGBA->width : rp->dst_width;
            int overlaystartline = 0;
            int i = 0;
            int j = 0;
            for (j = 0; j < rp->mPngRGBA->height; j++) {

                for (i = 0; i < overlayWidth * 4; i += 4) {
                    // R = R1 * a + R2 * (1-a);
                    int alpha = UNPREMULTIPLY_ALPHA(rp->mPngRGBA->pixels[overlaystartline + i + 3], *(data + startline + startinline + i + 3));
                    int oneminusalpha = 255 - alpha;
                    
                    *(data + startline + startinline + i + 0) = FAST_DIV255( ((rp->mPngRGBA->pixels[overlaystartline + i + 0] * alpha) + (oneminusalpha * *(data + startline + startinline + i + 0))) );

                    *(data + startline + startinline + i + 1) = FAST_DIV255(  ((rp->mPngRGBA->pixels[overlaystartline + i + 1] * alpha) + (oneminusalpha * *(data + startline + startinline + i + 1))) );

                    *(data + startline + startinline + i + 2) =FAST_DIV255(  ((rp->mPngRGBA->pixels[overlaystartline + i + 2] * alpha) + (oneminusalpha * *(data + startline + startinline + i + 2))) );
                }
                
                startline += rp->dst_width * 4;
                overlaystartline += rp->mPngRGBA->width * 4;
            }
        }
        //rp->mPngRGBA->pixels, rp->mPngRGBA->width, rp->mPngRGBA->height;
    }
}

//--------------------------
 if (rp->mirrorEnable != rp->logomirrorEnable) {
    rp->logomirrorEnable = rp->mirrorEnable;
    int k = 0;
    int line = 0;
    
    for (line = 0; line < rp->mPngRGBA->height * rp->mPngRGBA->width * 4; ) {
        int last = rp->mPngRGBA->width * 4 - 4 ;
        for (k = 0; k< last; k += 4, last -= 4) {
            unsigned char r = rp->mPngRGBA->pixels[line + last + 0];
            unsigned char g = rp->mPngRGBA->pixels[line + last + 1];
            unsigned char b = rp->mPngRGBA->pixels[line + last + 2];
            unsigned char a = rp->mPngRGBA->pixels[line + last + 3];
            
            unsigned char r1 = rp->mPngRGBA->pixels[line + k + 0];
            unsigned char g1 = rp->mPngRGBA->pixels[line + k + 1];
            unsigned char b1 = rp->mPngRGBA->pixels[line + k + 2];
            unsigned char a1 = rp->mPngRGBA->pixels[line + k + 3];
            
            rp->mPngRGBA->pixels[line + last + 0] = r1;
            rp->mPngRGBA->pixels[line + last + 1] = g1;
            rp->mPngRGBA->pixels[line + last + 2] = b1;
            rp->mPngRGBA->pixels[line + last + 3] = a1;
            
            rp->mPngRGBA->pixels[line + k + 0] = r;
            rp->mPngRGBA->pixels[line + k + 1] = g;
            rp->mPngRGBA->pixels[line + k + 2] = b;
            rp->mPngRGBA->pixels[line + k + 3] = a;
        }
        line += rp->mPngRGBA->width * 4;
    }
}