#pragma once
#include <array>
#include <cstddef>
#include <cstdint>

namespace dmf::v210 {

// SMPTE 75% color bars — 10-bit limited range
//   Y:     64 (black) to 940 (white)
//   Cb/Cr: 64 to 960, 512 = neutral grey
struct Color { uint16_t y, cb, cr; };

constexpr std::array<Color, 7> SMPTE_BARS = {{
    {721, 512, 512},  // white
    {674, 176, 543},  // yellow
    {581, 589, 176},  // cyan
    {534, 253, 207},  // green
    {251, 771, 817},  // magenta
    {204, 435, 848},  // red
    {111, 848, 481},  // blue
}};

// IRE 11-step greyscale bars
// from total black to 100% white
// Y_10bit = 64 + (IRE / 100) × (940 − 64) Rec. 709/Rec. 2020
// Y_10bit = 64 + (IRE × 8.76)
constexpr std::array<Color, 11> IRE_BARS = {{
    {64, 512, 512},  // IRE 0
    {152, 512, 512}, // IRE 10
    {239, 512, 512}, // IRE 20 and so on
    {327, 512, 512},
    {414, 512, 512},
    {502, 512, 512},
    {590, 512, 512},
    {677, 512, 512},
    {765, 512, 512},
    {852, 512, 512},
    {940, 512, 512},
}};

// Pack 6 pixels into 4 x 32-bit V210 words (16 bytes total).
//
// V210 is 4:2:2 — each pair of pixels shares one Cb and one Cr sample.
// The three pairs in a block map to words like this (bits [9:0],[19:10],[29:20]):
//   word 0: Cb(pair0) | Y(px0) | Cr(pair0)
//   word 1: Y(px1)    | Cb(pair1) | Y(px2)
//   word 2: Cr(pair1) | Y(px3)    | Cb(pair2)
//   word 3: Y(px4)    | Cr(pair2) | Y(px5)
inline void pack_block(
    uint8_t* out,
    Color p01, uint16_t y0, uint16_t y1,   // pair 0–1
    Color p23, uint16_t y2, uint16_t y3,   // pair 2–3
    Color p45, uint16_t y4, uint16_t y5)   // pair 4–5
{
    auto* w = reinterpret_cast<uint32_t*>(out);
    w[0] = (p01.cb & 0x3FFu)        | ((y0     & 0x3FFu) << 10) | ((p01.cr & 0x3FFu) << 20);
    w[1] = (y1     & 0x3FFu)        | ((p23.cb & 0x3FFu) << 10) | ((y2     & 0x3FFu) << 20);
    w[2] = (p23.cr & 0x3FFu)        | ((y3     & 0x3FFu) << 10) | ((p45.cb & 0x3FFu) << 20);
    w[3] = (y4     & 0x3FFu)        | ((p45.cr & 0x3FFu) << 10) | ((y5     & 0x3FFu) << 20);
}

// Write one horizontal line of bars.
// `stride` is the line size in bytes as returned by MXL (configInfo.discrete.sliceSizes[0]).
// Bytes beyond the active pixels are already zeroed by the mmap, so no explicit padding needed.
inline void write_bar_line(uint8_t* line, int width, uint32_t /*stride*/)
{
    const int n = static_cast<int>(SMPTE_BARS.size());
    const int blocks = width / 6;  // one V210 block = 6 pixels = 16 bytes

    for (int b = 0; b < blocks; b++) {
        int x = b * 6;
        auto color = [&](int px) -> const Color& {
            return SMPTE_BARS[static_cast<size_t>(px * n / width)];
        };
        const Color& c01 = color(x);
        const Color& c23 = color(x + 2);
        const Color& c45 = color(x + 4);
        pack_block(line + b * 16,
            c01, c01.y, color(x+1).y,
            c23, c23.y, color(x+3).y,
            c45, c45.y, color(x+5).y);
    }
}

// Fill an entire frame buffer with color bars.
inline void fill_frame(uint8_t* buf, int width, int height, uint32_t stride)
{
    for (int y = 0; y < height; y++) {
        write_bar_line(buf + static_cast<ptrdiff_t>(y) * stride, width, stride);
    }
}

} // namespace dmf::v210