#version 430 core
#extension GL_ARB_shader_draw_parameters : require

layout(location = 0) in vec3 aPosition;
layout(location = 1) in vec3 aNormal;
layout(location = 2) in vec2 aTexCoord;

struct InstanceData {
    mat4 transform;
    // Reserved for Phase B.4 follow-up (selection-blink retail-faithful
    // highlight): vec4 highlightColor; — extend stride here, increase the
    // _instanceSsbo upload size in WbDrawDispatcher, add a flat varying out,
    // and consume in mesh_modern.frag.
};

struct BatchData {
    uvec2 textureHandle;   // bindless handle for sampler2DArray
    uint  textureLayer;    // layer index (always 0 for per-instance composites)
    uint  flags;           // reserved — N.5 dispatcher owns all blend state
                           // (glBlendFunc per pass). If a future phase wants
                           // shader-side per-batch additive flag (Decision 2
                           // fallback), encode it here as bit 0.
};

layout(std430, binding = 0) readonly buffer InstanceBuffer {
    InstanceData Instances[];
};

// binding=1 here is the SSBO namespace — distinct from the UBO namespace.
// SceneLighting UBO also uses binding=1 in the fragment shader; GL keeps
// GL_SHADER_STORAGE_BUFFER and GL_UNIFORM_BUFFER binding tables separate.
// Task 10 dispatcher binds:
//   glBindBufferBase(GL_SHADER_STORAGE_BUFFER, 0, instanceSsbo)
//   glBindBufferBase(GL_SHADER_STORAGE_BUFFER, 1, batchSsbo)
// Existing SceneLightingUboBinding handles the UBO side.
layout(std430, binding = 1) readonly buffer BatchBuffer {
    BatchData Batches[];
};

// === Phase U.3: per-cell screen-space clip gate (gl_ClipDistance) =============
// Two SSBOs add the clip mechanism without disturbing binding=0/1 above.
//
// binding=2 — SHARED per-frame clip regions, one CellClip per "slot". Uploaded
// ONCE per frame by ClipFrame.UploadShared (shared across WbDrawDispatcher +
// EnvCellRenderer). Slot 0 is RESERVED = no-clip (count 0 ⇒ every plane passes).
//
// binding=3 — PER-RENDERER per-instance slot index, parallel to the binding=0
// instance buffer and indexed by the IDENTICAL per-instance index
// (gl_BaseInstanceARB + gl_InstanceID). instanceClipSlot[i] selects which
// CellClip region instance i is clipped against. Default all-zeros in U.3 ⇒
// every instance maps to slot 0 ⇒ no clipping ⇒ identical render to pre-U.3.
//
// CellClip std430 layout (144 bytes/slot): a uint count + 3 pad uints (16 bytes)
// then vec4 planes[8] (8 × 16 = 128 bytes). vec4 array stride is 16 under std430.
// ClipFrame on the CPU side lays out the bytes to match exactly (verified by
// ClipFrameLayoutTests). A clip-space vertex is INSIDE iff dot(plane, gl_Position)
// >= 0 for every active plane (see ClipPlaneSet for the plane convention).
struct CellClip {
    uint count;
    uint _p0;
    uint _p1;
    uint _p2;
    vec4 planes[8];
};
layout(std430, binding = 2) readonly buffer ClipRegionBuf {
    CellClip clipRegions[];
};
layout(std430, binding = 3) readonly buffer ClipSlotBuf {
    uint instanceClipSlot[];
};

// Core profile: redeclare gl_PerVertex so writing gl_ClipDistance[] is legal
// alongside gl_Position. The array is sized 8 to match the CellClip plane budget
// and the GL guarantee (GL_MAX_CLIP_DISTANCES >= 8). The host enables
// GL_CLIP_DISTANCE0..7 once at startup; unused planes are set to +1.0 below so
// they pass everything (no clipping) when the slot's count < 8.
out gl_PerVertex {
    vec4  gl_Position;
    float gl_ClipDistance[8];
};

uniform mat4 uViewProjection;

// Phase Post-A.5 (ISSUE #52, 2026-05-10): per-pass offset into Batches[].
// gl_DrawIDARB resets to 0 at the start of each glMultiDrawElementsIndirect
// call, so the transparent pass — which begins later in the indirect buffer
// — was fetching Batches[0..transparentCount) instead of its actual section
// at Batches[opaqueCount..end). The lifestone crystal (a transparent draw)
// ended up reading the FIRST OPAQUE batch's TextureHandle every frame. As
// the camera moved and the opaque front-to-back sort reordered which group
// landed at BatchData[0], the lifestone's apparent texture flickered to
// whatever was first — frequently the player character's body parts.
//
// WbDrawDispatcher.Draw sets this to 0 before the opaque MDI call and to
// _opaqueDrawCount before the transparent MDI call, matching WorldBuilder's
// uDrawIDOffset pattern in BaseObjectRenderManager.cs line 845.
uniform int uDrawIDOffset;

out vec3 vNormal;
out vec2 vTexCoord;
out vec3 vWorldPos;
out flat uvec2 vTextureHandle;
out flat uint  vTextureLayer;

void main() {
    int instanceIndex = gl_BaseInstanceARB + gl_InstanceID;
    mat4 model = Instances[instanceIndex].transform;

    vec4 worldPos = model * vec4(aPosition, 1.0);
    gl_Position = uViewProjection * worldPos;

    // Phase U.3: per-instance clip gate. instanceClipSlot is indexed by the
    // SAME instanceIndex used for the binding=0 transform above, so the slot
    // travels with the instance through the MDI BaseInstance offsets. Slot 0
    // (the U.3 default) has count 0 ⇒ the second loop sets all 8 distances to
    // +1.0 ⇒ nothing is clipped.
    uint _slot = instanceClipSlot[instanceIndex];
    CellClip _c = clipRegions[_slot];
    for (uint i = 0u; i < _c.count; ++i)
        gl_ClipDistance[i] = dot(_c.planes[i], gl_Position);
    for (uint i = _c.count; i < 8u; ++i)
        gl_ClipDistance[i] = 1.0;

    vWorldPos = worldPos.xyz;
    vNormal = normalize(mat3(model) * aNormal);
    vTexCoord = aTexCoord;

    BatchData b = Batches[uDrawIDOffset + gl_DrawIDARB];
    vTextureHandle = b.textureHandle;
    vTextureLayer = b.textureLayer;
}