source: code/branches/tutorial/Tutorial/Media/materials/programs/DepthShadowmapNormalMapReceiverVp.glsl @ 31

attribute vec3 tangent;

uniform mat4 world;
uniform mat4 worldViewProj;
uniform mat4 texViewProj;
uniform vec4 lightPosition; // object space
uniform vec4 shadowDepthRange;

varying vec3 tangentLightDir;


void main()
{
        gl_Position = ftransform();
        
        vec4 worldPos = world * gl_Vertex;

        // Get object space light direction 
    vec3 lightDir = normalize(lightPosition.xyz -  (gl_Vertex.xyz * lightPosition.w));

        // calculate shadow map coords
        gl_TexCoord[0] = texViewProj * worldPos;
#if LINEAR_RANGE
        // adjust by fixed depth bias, rescale into range
        gl_TexCoord[0].z = (gl_TexCoord[0].z - shadowDepthRange.x) * shadowDepthRange.w;
#endif

        // pass the main uvs straight through unchanged 
        gl_TexCoord[1] = gl_MultiTexCoord0;

        // Calculate the binormal (NB we assume both normal and tangent are 
        // already normalised) 
        vec3 binormal = cross(gl_Normal, tangent); 

        // Form a rotation matrix out of the vectors 
        mat3 rotation = mat3(tangent, binormal, gl_Normal); 
    
        // Transform the light vector according to this matrix 
        tangentLightDir = normalize(rotation * lightDir); 

}