// native_bridge/src/RenderFxJob.cpp
// Offline audio-FX render mode (Phase 0+1 of PLAN_MASTERBUS_FX_RACK_VST.md):
// `daw_vst_bridge --render-fx <job.json> --in <input.wav> --out <output.wav>`
// processes an existing WAV through a VST3 effect chain (audio-in/audio-out,
// no MIDI) plus builtin slots (gain/normalize). Also `--scan <dir>` which
// classifies VST3 modules as instrument/effect.
//
// Same JSON/b64/WAV helpers as RenderJob.cpp (sheredom/json.h vendored with
// the VST3 SDK). The VST3 host pattern is the Vst3Instrument.cpp one minus
// MIDI/GUI — component + controller, audio buses, processData with input
// buffers copied directly (setChannelBuffers is unusable: prepare() owns them).
#include "RenderFxJob.h"

// sheredom/json.h (public domain) — vendored copy; SDK copy as fallback.
#if __has_include("sheredom_json.h")
#include "sheredom_json.h"
#elif __has_include("vst3sdk/public.sdk/source/vst/moduleinfo/json.h")
#include "vst3sdk/public.sdk/source/vst/moduleinfo/json.h"
#endif
#ifdef HAVE_VST3SDK
#include "public.sdk/source/vst/hosting/module.h"
#include "public.sdk/source/vst/hosting/hostclasses.h"
#include "public.sdk/source/vst/hosting/processdata.h"
#include "public.sdk/source/vst/hosting/parameterchanges.h"
#include "public.sdk/source/vst/vstpresetfile.h"
#include "public.sdk/source/common/memorystream.h"
#include "pluginterfaces/vst/ivstaudioprocessor.h"
#include "pluginterfaces/vst/ivstcomponent.h"
#include "pluginterfaces/vst/ivsteditcontroller.h"
#include "pluginterfaces/vst/ivstprocesscontext.h"
#include "pluginterfaces/vst/ivstmessage.h"
#include "pluginterfaces/gui/iplugview.h"
#endif

#ifdef _WIN32
#ifndef NOMINMAX
#define NOMINMAX
#endif
#include <windows.h>
#endif
#include "FxRealtimeIPC.h"
#include "BuiltinFxChain.h"

#include <algorithm>
#include <atomic>
#include <cctype>
#include <cmath>
#include <cstdio>
#include <cstdlib>
#include <cstring>
#include <chrono>
#include <filesystem>
#include <fstream>
#include <iostream>
#include <condition_variable>
#include <functional>
#include <iterator>
#include <map>
#include <memory>
#include <mutex>
#include <string>
#include <thread>
#include <vector>

namespace {

// --- sheredom/json.h helpers (same as RenderJob.cpp) ------------------------
const json_object_element_s* member(const json_object_s* o, const char* key) {
    for (const json_object_element_s* e = o ? o->start : nullptr; e; e = e->next)
        if (e->name && e->name->string && std::strcmp(e->name->string, key) == 0)
            return e;
    return nullptr;
}

const json_value_s* memberValue(const json_object_s* o, const char* key) {
    const json_object_element_s* m = member(o, key);
    return m ? m->value : nullptr;
}

std::string memberString(const json_object_s* o, const char* key, const std::string& def) {
    const json_value_s* v = memberValue(o, key);
    if (v && v->type == json_type_string) {
        const auto* s = static_cast<const json_string_s*>(v->payload);
        return std::string(s->string, s->string_size);
    }
    return def;
}

bool memberNumber(const json_object_s* o, const char* key, double& out) {
    const json_value_s* v = memberValue(o, key);
    if (v && v->type == json_type_number) {
        out = std::atof(static_cast<const json_number_s*>(v->payload)->number);
        return true;
    }
    return false;
}

int64_t memberInt(const json_object_s* o, const char* key, int64_t def) {
    double d;
    return memberNumber(o, key, d) ? (int64_t)d : def;
}

bool memberBool(const json_object_s* o, const char* key, bool def) {
    const json_value_s* v = memberValue(o, key);
    if (v && v->type == json_type_true) return true;
    if (v && v->type == json_type_false) return false;
    return def;
}

// --- WAV writer (stdlib only, IEEE Float32 stereo, little-endian) ------------
// Lossless: keeps the full float32 render buffer bit-exact. No PCM truncation
// and no dither here — dither/quantization belongs ONLY to a final export
// stage (lossless-audio-compliance skill, Rules 4 & 7). Format tag 3.
void writeU16(std::ofstream& f, uint16_t v) {
    char b[2] = { (char)(v & 0xFF), (char)((v >> 8) & 0xFF) };
    f.write(b, 2);
}
void writeU32(std::ofstream& f, uint32_t v) {
    char b[4] = { (char)(v & 0xFF), (char)((v >> 8) & 0xFF),
                  (char)((v >> 16) & 0xFF), (char)((v >> 24) & 0xFF) };
    f.write(b, 4);
}
void writeWavHeader(std::ofstream& f, uint32_t sampleRate) {
    f.write("RIFF", 4);
    writeU32(f, 36);
    f.write("WAVE", 4);
    f.write("fmt ", 4);
    writeU32(f, 16);
    writeU16(f, 3);           // IEEE float
    writeU16(f, 2);
    writeU32(f, sampleRate);
    writeU32(f, sampleRate * 8);  // byte rate = sr * 2ch * 4B
    writeU16(f, 8);
    writeU16(f, 32);
    f.write("data", 4);
    writeU32(f, 0);
}
void finishWav(std::ofstream& f, uint64_t dataBytes) {
    f.seekp(4);
    writeU32(f, (uint32_t)(36 + dataBytes));
    f.seekp(40);
    writeU32(f, (uint32_t)dataBytes);
    f.flush();
}
void writeFrames(std::ofstream& f, const float* L, const float* R, uint32_t n) {
    // IEEE float32 little-endian, bit-exact: no clamp, no quantization.
    for (uint32_t i = 0; i < n; ++i) {
        f.write(reinterpret_cast<const char*>(&L[i]), 4);
        f.write(reinterpret_cast<const char*>(&R[i]), 4);
    }
}

// --- WAV reader (stdlib only): PCM 16/24/32-bit + IEEE float32, mono/stereo
// → stereo float32. Returns false on any parse error (job/argument error). ---
struct WavIn {
    uint32_t sampleRate = 0;
    uint16_t channels = 0;
    uint16_t bits = 0;
    std::vector<float> L, R;  // same length
};

uint32_t readU32le(const uint8_t* p) {
    return (uint32_t)p[0] | ((uint32_t)p[1] << 8) | ((uint32_t)p[2] << 16) | ((uint32_t)p[3] << 24);
}
uint16_t readU16le(const uint8_t* p) {
    return (uint16_t)((uint16_t)p[0] | ((uint16_t)p[1] << 8));
}

bool readWav(const std::string& path, WavIn& out) {
    std::ifstream f(path, std::ios::binary);
    if (!f) { std::cerr << "[RenderFx] cannot read input: " << path << std::endl; return false; }
    std::vector<uint8_t> buf((std::istreambuf_iterator<char>(f)), std::istreambuf_iterator<char>());
    if (buf.size() < 44) return false;
    if (std::memcmp(buf.data(), "RIFF", 4) != 0 || std::memcmp(buf.data() + 8, "WAVE", 4) != 0)
        return false;

    uint16_t fmtTag = 0, channels = 0, bits = 0;
    uint32_t sampleRate = 0;
    size_t dataOff = 0, dataLen = 0;
    size_t p = 12;
    bool foundFmt = false, foundData = false;
    while (p + 8 <= buf.size()) {
        const char* id = reinterpret_cast<const char*>(buf.data() + p);
        uint32_t sz = readU32le(buf.data() + p + 4);
        size_t chunkStart = p + 8;
        if (chunkStart + sz > buf.size()) break;  // truncated chunk — stop
        if (std::memcmp(id, "fmt ", 4) == 0 && sz >= 16) {
            fmtTag = readU16le(buf.data() + chunkStart);
            channels = readU16le(buf.data() + chunkStart + 2);
            sampleRate = readU32le(buf.data() + chunkStart + 4);
            bits = readU16le(buf.data() + chunkStart + 14);
            foundFmt = true;
        } else if (std::memcmp(id, "data", 4) == 0) {
            dataOff = chunkStart;
            dataLen = sz;
            foundData = true;
            break;
        }
        p = chunkStart + sz + (sz & 1);
    }
    if (!foundFmt || !foundData) return false;
    // Supported: PCM 16/24/32 and IEEE float 32; mono or stereo.
    const bool isFloat = (fmtTag == 3);
    if (!(fmtTag == 1 || isFloat)) {
        std::cerr << "[RenderFx] unsupported WAV format tag " << fmtTag << " (need PCM/float)" << std::endl;
        return false;
    }
    if (channels < 1 || channels > 2) {
        std::cerr << "[RenderFx] unsupported WAV channel count " << channels << std::endl;
        return false;
    }
    const uint16_t bytes = (bits + 7) / 8;
    if (bytes != 2 && bytes != 3 && bytes != 4) return false;

    const size_t frameBytes = (size_t)channels * bytes;
    const size_t nFrames = dataLen / frameBytes;
    out.sampleRate = sampleRate;
    out.channels = channels;
    out.bits = bits;
    out.L.assign(nFrames, 0.0f);
    out.R.assign(nFrames, 0.0f);
    const uint8_t* d = buf.data() + dataOff;
    for (size_t i = 0; i < nFrames; ++i) {
        float l = 0.f, r = 0.f;
        for (uint16_t c = 0; c < channels; ++c) {
            const uint8_t* s = d + (i * channels + c) * bytes;
            float v = 0.f;
            if (isFloat && bytes == 4) {
                uint32_t u = readU32le(s);
                float fv;
                std::memcpy(&fv, &u, 4);
                v = fv;
            } else if (bytes == 2) {
                v = (float)(int16_t)((uint16_t)s[0] | ((uint16_t)s[1] << 8)) / 32768.f;
            } else if (bytes == 3) {
                int32_t iv = (int32_t)s[0] | ((int32_t)s[1] << 8) | ((int32_t)s[2] << 16);
                if (iv & 0x800000) iv |= ~0xFFFFFF;  // sign extend
                v = (float)iv / 8388608.f;
            } else {  // bytes == 4, PCM int32
                int32_t iv = (int32_t)readU32le(s);
                v = (float)iv / 2147483648.f;
            }
            if (c == 0) l = v; else r = v;
        }
        out.L[i] = l;
        out.R[i] = (channels == 1) ? l : r;
    }
    return true;
}

// --- base64 (RFC 4648) — same helpers as RenderJob.cpp ----------------------
static const char* kBase64Tbl = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
static std::string base64Encode(const uint8_t* data, size_t len) {
    std::string out;
    out.reserve(((len + 2) / 3) * 4);
    for (size_t i = 0; i < len; i += 3) {
        uint32_t n = (uint32_t)data[i] << 16;
        if (i + 1 < len) n |= (uint32_t)data[i + 1] << 8;
        if (i + 2 < len) n |= (uint32_t)data[i + 2];
        out += kBase64Tbl[(n >> 18) & 63];
        out += kBase64Tbl[(n >> 12) & 63];
        out += (i + 1 < len) ? kBase64Tbl[(n >> 6) & 63] : '=';
        out += (i + 2 < len) ? kBase64Tbl[n & 63] : '=';
    }
    return out;
}

// Append-only variant (no return temporary) so it can run inside an SEH
// __try frame (C2712: __try functions cannot have destructor temporaries).
static void base64EncodeTo(const uint8_t* data, size_t len, std::string& out) {
    static const char* tbl = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
    size_t cap = ((len + 2) / 3) * 4;
    out.clear();
    out.reserve(cap);
    for (size_t i = 0; i < len; i += 3) {
        uint32_t n = (uint32_t)data[i] << 16;
        if (i + 1 < len) n |= (uint32_t)data[i + 1] << 8;
        if (i + 2 < len) n |= (uint32_t)data[i + 2];
        out += tbl[(n >> 18) & 63];
        out += tbl[(n >> 12) & 63];
        out += (i + 1 < len) ? tbl[(n >> 6) & 63] : '=';
        out += (i + 2 < len) ? tbl[n & 63] : '=';
    }
}
static int b64val(unsigned char c) {
    if (c >= 'A' && c <= 'Z') return c - 'A';
    if (c >= 'a' && c <= 'z') return c - 'a' + 26;
    if (c >= '0' && c <= '9') return c - '0' + 52;
    if (c == '+') return 62;
    if (c == '/') return 63;
    return -1;
}
static bool base64Decode(const std::string& in, std::vector<uint8_t>& out) {
    out.clear();
    out.reserve((in.size() / 4) * 3);
    uint32_t acc = 0;
    int bits = 0;
    for (unsigned char ch : in) {
        if (ch == '=' || ch == '\n' || ch == '\r' || ch == ' ') continue;
        int v = b64val(ch);
        if (v < 0) return false;
        acc = (acc << 6) | (uint32_t)v;
        bits += 6;
        if (bits >= 8) {
            bits -= 8;
            out.push_back((uint8_t)((acc >> bits) & 0xFF));
        }
    }
    return true;
}

// --- VST3 audio-effect host (offline; no MIDI, no GUI) ----------------------
#ifdef HAVE_VST3SDK
using Steinberg::tresult;
using Steinberg::kResultOk;
using Steinberg::kResultTrue;
using Steinberg::kResultFalse;
using Steinberg::kNoInterface;
using Steinberg::FUnknownPtr;
using Steinberg::IPtr;
using Steinberg::owned;
using Steinberg::FIDString;
using Steinberg::int32;
using Steinberg::uint32;

using Steinberg::Vst::IComponent;
using Steinberg::Vst::IEditController;
using Steinberg::Vst::IAudioProcessor;
using Steinberg::Vst::IComponentHandler;
using Steinberg::Vst::IConnectionPoint;
using Steinberg::Vst::ProcessSetup;
using Steinberg::Vst::ProcessContext;
using Steinberg::Vst::HostProcessData;
using Steinberg::Vst::HostApplication;
using Steinberg::Vst::BusInfo;
using Steinberg::Vst::kAudio;
using Steinberg::Vst::kEvent;
using Steinberg::Vst::kInput;
using Steinberg::Vst::kOutput;
using Steinberg::Vst::kRealtime;
using Steinberg::Vst::kSample32;

class HostComponentHandlerFx : public IComponentHandler {
public:
    tresult queryInterface(const char*, void** v) override { *v = nullptr; return kNoInterface; }
    Steinberg::uint32 addRef() override { return 1; }
    Steinberg::uint32 release() override { return 1; }
    tresult beginEdit(Steinberg::Vst::ParamID) override { return kResultOk; }
    // Forward automation from the plugin's native GUI into the controller so
    // controller-driven meters/UI update with the edited parameter. Null until
    // the controller is attached (set in vst3FxLoadInner).
    Steinberg::Vst::IEditController* controller = nullptr;
    tresult performEdit(Steinberg::Vst::ParamID id, Steinberg::Vst::ParamValue v) override {
        if (controller) controller->setParamNormalized(id, v);
        return kResultOk;
    }
    tresult endEdit(Steinberg::Vst::ParamID) override { return kResultOk; }
    tresult restartComponent(Steinberg::int32) override { return kResultOk; }
};

using Steinberg::IPlugView;
using Steinberg::kPlatformTypeHWND;

// Minimal IPlugFrame so the plugin can resize its editor view.
class HostPlugFrameFx : public Steinberg::IPlugFrame {
public:
    tresult queryInterface(const char*, void** v) override {
        *v = nullptr;
        return kNoInterface;
    }
    Steinberg::uint32 addRef() override { return 1; }
    Steinberg::uint32 release() override { return 1; }
    tresult resizeView(Steinberg::IPlugView* view, Steinberg::ViewRect* newSize) override {
        if (view && newSize) view->onSize(newSize);
        return kResultOk;
    }
};

struct Vst3FxState {
    VST3::Hosting::Module::Ptr module;  // destroyed LAST (owns the factory)
    IPtr<IComponent> component;
    IPtr<IEditController> controller;
    IPtr<HostApplication> hostApp;
    HostProcessData processData;
    ProcessContext processContext;
    HostComponentHandlerFx componentHandler;
    int32 inputChannels = 2;
    int32 outputChannels = 2;
    int32 latencySamples = 0;  // PDC (REPORT_LATENCY, Phase 2.8)
    bool controllerIsComponent = false;
    Steinberg::IPlugFrame* plugFrame = nullptr;  // owned; freed in closeEditor
    IPtr<IPlugView> view;                        // editor view while GUI open
    Steinberg::Vst::ParameterChanges paramChanges;  // live setParam queue -> inputParameterChanges
    std::mutex paramMutex;                          // setParam (loop/worker) vs processAudio
    std::map<std::string, Steinberg::Vst::ParamID> paramByName;  // lower(title/shortTitle) -> id
};

// Class selection for FX: any kVstAudioEffectClass whose subcategory does NOT
// contain "Instrument" (Ozone/ScalerAudio et al.). Pick the first match.
bool pickFxClass(const VST3::Hosting::PluginFactory& factory,
                 VST3::Hosting::ClassInfo& out) {
    auto infos = factory.classInfos();  // temporary vector — copy, never keep pointer
    for (const auto& ci : infos) {
        if (ci.category() != kVstAudioEffectClass) continue;
        if (ci.subCategoriesString().find("Instrument") != std::string::npos) continue;
        out = ci;
        return true;
    }
    return false;
}

// Must live OUTSIDE any __try (C2712: no C++ unwinding inside SEH frames).
// Windows DLL search: LoadLibraryW (VST3 SDK) does NOT search the plugin's own
// directory for dependency DLLs -> ERROR_MOD_NOT_FOUND (0x7E) when a VST3 needs
// helper DLLs next to it (bundle Contents\x86_64-win or sidecar files). Point
// the DLL search path at the plugin directory before Module::create.
#ifdef _WIN32
static void setPluginSearchPath(const std::string& path) {
    if (path.empty()) return;
    std::string dir;
    // Folder bundle (.vst3 dir): deps live in Contents\<arch>\
    // Single file (.vst3/.dll): deps live next to it.
    size_t baseSlash = path.find_last_of("/\\");
    std::string base = baseSlash == std::string::npos ? "" : path.substr(baseSlash + 1);
    std::string parent = baseSlash == std::string::npos ? "" : path.substr(0, baseSlash);
    if (!base.empty() && base.size() > 4 &&
        base.substr(base.size() - 4) == ".vst3") {
        // Could be a folder bundle OR a single .vst3 file. Check folder first.
        DWORD attr = GetFileAttributesA(path.c_str());
        if (attr != INVALID_FILE_ATTRIBUTES && (attr & FILE_ATTRIBUTE_DIRECTORY)) {
            dir = path + "\\Contents\\x86_64-win";
            DWORD dAttr = GetFileAttributesA(dir.c_str());
            if (dAttr == INVALID_FILE_ATTRIBUTES)
                dir = path + "\\Contents";
        } else {
            dir = parent;
        }
    } else {
        dir = parent;
    }
    if (dir.empty()) return;
    // SetDllDirectoryW is process-global, fine here: bridge loads one plugin
    // at a time (render slots serial; realtime chain loads on its own thread).
    std::wstring wdir(dir.begin(), dir.end());
    SetDllDirectoryW(wdir.c_str());
    std::cerr << "[RenderFx] dll search dir: " << dir << std::endl;
}
#endif

bool vst3FxLoadInner(Vst3FxState* s, const std::string& path, double sampleRate,
                     int32 maxBlockSize, std::string& err) {
    using namespace VST3::Hosting;
#ifdef _WIN32
    setPluginSearchPath(path);
#endif
    Module::Ptr module = Module::create(path, err);
    if (!module) { std::cerr << "[RenderFx] STEP module.create FAILED: " << err << std::endl; return false; }
    std::cerr << "[RenderFx] STEP module.create OK" << std::endl;
    const PluginFactory& factory = module->getFactory();
    ClassInfo chosen;
    if (!pickFxClass(factory, chosen)) {
        err = "no audio-effect class in " + path;
        std::cerr << "[RenderFx] STEP pickFxClass FAILED: " << err << std::endl;
        return false;
    }
    std::cerr << "[RenderFx] STEP pickFxClass OK: " << chosen.name() << std::endl;
    IPtr<IComponent> component = factory.createInstance<IComponent>(chosen.ID());
    if (!component) { std::cerr << "[RenderFx] STEP createInstance FAILED" << std::endl; err = "createInstance<IComponent> failed"; return false; }
    std::cerr << "[RenderFx] STEP createInstance OK" << std::endl;
    IPtr<HostApplication> hostApp = owned(new HostApplication());
    FUnknownPtr<Steinberg::IPluginBase> plugBase(component.get());
    if (!plugBase || plugBase->initialize(hostApp) != kResultOk) {
        err = "component initialize failed";
        std::cerr << "[RenderFx] STEP initialize FAILED" << std::endl;
        return false;
    }
    std::cerr << "[RenderFx] STEP initialize OK" << std::endl;
    IPtr<IEditController> controller;
    bool isSingle = false;
    if (component->queryInterface(IEditController::iid, (void**)&controller) == kResultTrue) {
        isSingle = true;
    } else {
        Steinberg::TUID cid = {};
        tresult cidRes = component->getControllerClassId(cid);
        if (cidRes == kResultTrue || cidRes == kResultOk) {
            controller = factory.createInstance<IEditController>(VST3::UID(cid));
            if (controller) {
                FUnknownPtr<Steinberg::IPluginBase> ctrlBase(controller.get());
                if (!ctrlBase || ctrlBase->initialize(hostApp) != kResultOk) controller = nullptr;
            }
        }
    }
    if (!controller) { err = "no edit controller"; std::cerr << "[RenderFx] STEP controller FAILED" << std::endl; return false; }
    std::cerr << "[RenderFx] STEP controller OK" << std::endl;

    s->controllerIsComponent = isSingle;
    controller->setComponentHandler(&s->componentHandler);
    s->componentHandler.controller = controller.get();
    FUnknownPtr<IConnectionPoint> compCP(component);
    FUnknownPtr<IConnectionPoint> ctrlCP(controller);
    if (compCP && ctrlCP) { compCP->connect(ctrlCP); ctrlCP->connect(compCP); }
    std::cerr << "[RenderFx] STEP connection points OK" << std::endl;

    int32 numAudioInputs = component->getBusCount(kAudio, kInput);
    for (int32 i = 0; i < numAudioInputs; ++i) component->activateBus(kAudio, kInput, i, true);
    int32 numAudioOutputs = component->getBusCount(kAudio, kOutput);
    for (int32 i = 0; i < numAudioOutputs; ++i) component->activateBus(kAudio, kOutput, i, true);
    int32 numEventInputs = component->getBusCount(kEvent, kInput);
    for (int32 i = 0; i < numEventInputs; ++i) component->activateBus(kEvent, kInput, i, true);
    int32 numEventOutputs = component->getBusCount(kEvent, kOutput);
    for (int32 i = 0; i < numEventOutputs; ++i) component->activateBus(kEvent, kOutput, i, true);
    std::cerr << "[RenderFx] STEP buses activated (ain=" << numAudioInputs << " aout=" << numAudioOutputs << ")" << std::endl;
    if (numAudioInputs < 1 || numAudioOutputs < 1) {
        err = "plugin has no audio in/out (instrument?)"; return false;
    }

    FUnknownPtr<IAudioProcessor> processor(component);
    if (!processor) { err = "no IAudioProcessor"; return false; }
    ProcessSetup setup{kRealtime, kSample32, maxBlockSize, sampleRate};
    std::cerr << "[RenderFx] STEP setupProcessing begin" << std::endl;
    if (processor->setupProcessing(setup) != kResultOk) { err = "setupProcessing failed"; return false; }
    std::cerr << "[RenderFx] STEP setupProcessing OK" << std::endl;
    std::cerr << "[RenderFx] STEP setActive begin" << std::endl;
    if (component->setActive(true) != kResultOk) { err = "setActive failed"; return false; }
    std::cerr << "[RenderFx] STEP setActive OK" << std::endl;
    processor->setProcessing(true);
    std::cerr << "[RenderFx] STEP setProcessing OK" << std::endl;
    s->latencySamples = processor->getLatencySamples();
    std::cerr << "[RenderFx] STEP getLatencySamples = " << s->latencySamples << std::endl;
    if (!s->processData.prepare(*component, maxBlockSize, kSample32)) {
        err = "processData.prepare failed"; return false;
    }
    std::cerr << "[RenderFx] STEP processData.prepare OK" << std::endl;

    BusInfo bi = {};
    if (component->getBusInfo(kAudio, kInput, 0, bi) == kResultOk && bi.channelCount > 0)
        s->inputChannels = bi.channelCount;
    bi = {};
    if (component->getBusInfo(kAudio, kOutput, 0, bi) == kResultOk && bi.channelCount > 0)
        s->outputChannels = bi.channelCount;

    // Live setParam: map lowercased title/shortTitle -> ParamID so the
    // realtime chain can address VST3 params by name (no UI-side tags).
    // Titles are UTF-16 (String128); keep the ASCII subset lowercased.
    for (int32 pi = 0; pi < controller->getParameterCount(); ++pi) {
        Steinberg::Vst::ParameterInfo info = {};
        if (controller->getParameterInfo(pi, info) != kResultOk) continue;
        auto addName = [&](const Steinberg::Vst::TChar* nm) {
            if (!nm) return;
            std::string k;
            for (const Steinberg::Vst::TChar* p = nm; p && *p && k.size() < 64; ++p) {
                if (*p < 0x80) k.push_back((char)std::tolower((unsigned char)*p));
            }
            if (k.empty()) return;
            s->paramByName[k] = info.id;
        };
        addName(info.title);
        addName(info.shortTitle);
    }

    s->module = std::move(module);
    s->component = std::move(component);
    s->controller = std::move(controller);
    s->hostApp = std::move(hostApp);
    s->processContext.sampleRate = sampleRate;
    s->processContext.tempo = 120.0;
    s->processContext.timeSigNumerator = 4;
    s->processContext.timeSigDenominator = 4;
    s->processContext.state = ProcessContext::kPlaying | ProcessContext::kTempoValid |
                              ProcessContext::kTimeSigValid | ProcessContext::kProjectTimeMusicValid;
    std::cout << "[RenderFx] VST3 FX loaded " << path << " (" << chosen.name()
              << ") in=" << s->inputChannels << " out=" << s->outputChannels << std::endl;
    return true;
}
#endif // HAVE_VST3SDK

class Vst3Fx {
public:
    Vst3Fx() = default;
    ~Vst3Fx() {
#ifdef HAVE_VST3SDK
        if (!state_) return;
        auto* s = static_cast<Vst3FxState*>(state_);
        if (s->component) {
            FUnknownPtr<IAudioProcessor> processor(s->component);
            if (processor) processor->setProcessing(false);
            s->component->setActive(false);
            s->component->terminate();
        }
        if (s->controller && !s->controllerIsComponent) s->controller->terminate();
        s->processData.unprepare();
        delete s;
        state_ = nullptr;
#endif
    }
    bool load(const std::string& path, double sampleRate, int32 maxBlockSize) {
#ifndef HAVE_VST3SDK
        (void)path; (void)sampleRate; (void)maxBlockSize;
        return false;
#else
        if (state_) return true;
        auto* s = new Vst3FxState();
        std::string err;
        if (!vst3FxLoadInner(s, path, sampleRate, maxBlockSize, err)) {
            std::cerr << "[RenderFx] VST3 FX load FAILED: " << err << std::endl;
            delete s;
            return false;
        }
        state_ = s;
        return true;
#endif
    }
    // Audio in → audio out, one block. Must be SEH-wrapped by the caller.
    void processAudio(const float* inL, const float* inR,
                      float* outL, float* outR, uint32_t n) {
#ifndef HAVE_VST3SDK
        (void)inL; (void)inR; (void)outL; (void)outR; (void)n;
#else
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s || !s->component || n == 0) return;
        FUnknownPtr<IAudioProcessor> processor(s->component);
        if (!processor) return;
        s->processData.processMode = kRealtime;
        s->processData.numSamples = (int32)n;
        s->processData.inputEvents = nullptr;
        {
            std::lock_guard<std::mutex> lk(s->paramMutex);
            s->processData.inputParameterChanges = &s->paramChanges;
        }
        s->processData.processContext = &s->processContext;
        s->processContext.projectTimeSamples += n;
        s->processContext.projectTimeMusic =
            (double)s->processContext.projectTimeSamples / s->processContext.sampleRate *
            (s->processContext.tempo / 60.0);
        // Copy input into the prepared (owned) input buffers — direct write,
        // setChannelBuffers is unusable (channelBufferOwner=true after prepare).
        if (s->processData.numInputs > 0) {
            const Steinberg::Vst::AudioBusBuffers& ib = s->processData.inputs[0];
            for (int32 c = 0; c < ib.numChannels; ++c) {
                float* dst = ib.channelBuffers32[c];
                if (!dst) continue;
                const float* src = (c == 0) ? inL : (c == 1 ? inR : inL);
                std::memcpy(dst, src, n * sizeof(float));
            }
        }
        // Host buffers must be zeroed before process (plugins skip output
        // leave garbage otherwise — same rule as Vst3Instrument).
        if (s->processData.numOutputs > 0) {
            for (int32 b = 0; b < s->processData.numOutputs; ++b) {
                const Steinberg::Vst::AudioBusBuffers& ob = s->processData.outputs[b];
                for (int32 c = 0; c < ob.numChannels; ++c)
                    if (ob.channelBuffers32[c])
                        std::memset(ob.channelBuffers32[c], 0, n * sizeof(float));
            }
        }
        tresult pr = processor->process(s->processData);
        if (pr >= 0 && s->processData.numOutputs > 0) {
            const Steinberg::Vst::AudioBusBuffers& out = s->processData.outputs[0];
            const float* buf0 = out.numChannels > 0 ? out.channelBuffers32[0] : nullptr;
            const float* buf1 = out.numChannels > 1 ? out.channelBuffers32[1] : nullptr;
            if (buf0) std::memcpy(outL, buf0, n * sizeof(float));
            else std::memset(outL, 0, n * sizeof(float));
            if (buf1) std::memcpy(outR, buf1, n * sizeof(float));
            else if (buf0) std::memcpy(outR, buf0, n * sizeof(float));  // mono → stereo
            else std::memset(outR, 0, n * sizeof(float));
        } else {
            std::memset(outL, 0, n * sizeof(float));
            std::memset(outR, 0, n * sizeof(float));
        }
        {
            // Each block consumes the queue once; clear so the next setParam
            // writes fresh points (adelay reads the LAST point).
            std::lock_guard<std::mutex> lk(s->paramMutex);
            s->paramChanges.clearQueue();
        }
#endif
    }
    // Apply preset: base64 of a raw .vstpreset FILE (VST3 magic + chunk list)
    // OR of the bridge state blob [4B BE compLen][comp][4B BE ctrlLen][ctrl].
    bool applyPreset(const std::string& b64) {
#ifndef HAVE_VST3SDK
        (void)b64;
        return false;
#else
        std::vector<uint8_t> bytes;
        if (!base64Decode(b64, bytes) || bytes.empty()) return false;
        std::vector<uint8_t> comp, ctrl;
        if (bytes.size() >= 4 && std::memcmp(bytes.data(), "VST3", 4) == 0) {
            // Raw .vstpreset file: parse with the SDK PresetFile.
            Steinberg::MemoryStream stream;
            if (stream.write(bytes.data(), (int32)bytes.size(), nullptr) != Steinberg::kResultOk) return false;
            stream.seek(0, Steinberg::IBStream::kIBSeekSet, nullptr);
            Steinberg::Vst::PresetFile pf(&stream);
            if (!pf.readChunkList()) return false;
            auto readChunk = [&](Steinberg::Vst::ChunkType which, std::vector<uint8_t>& outc) -> bool {
                const Steinberg::Vst::PresetFile::Entry* e = pf.getEntry(which);
                if (!e) return true;
                bool seeked = (which == Steinberg::Vst::kComponentState)
                                  ? pf.seekToComponentState() : pf.seekToControllerState();
                if (!seeked) return false;
                outc.resize((size_t)e->size);
                if (e->size == 0) return true;
                int32 got = 0;
                return stream.read(outc.data(), (int32)e->size, &got) == Steinberg::kResultOk &&
                       got == e->size;
            };
            if (!readChunk(Steinberg::Vst::kComponentState, comp) ||
                !readChunk(Steinberg::Vst::kControllerState, ctrl))
                return false;
        } else {
            // Bridge state blob.
            auto read32 = [&bytes](size_t off) -> uint32_t {
                return ((uint32_t)bytes[off] << 24) | ((uint32_t)bytes[off + 1] << 16) |
                       ((uint32_t)bytes[off + 2] << 8) | (uint32_t)bytes[off + 3];
            };
            if (bytes.size() < 8) return false;
            size_t off = 0;
            uint32_t clen = read32(off); off += 4;
            if (off + clen > bytes.size()) return false;
            comp.assign(bytes.begin() + off, bytes.begin() + off + clen);
            off += clen;
            if (off + 4 > bytes.size()) return false;
            uint32_t klen = read32(off); off += 4;
            if (off + klen > bytes.size()) return false;
            ctrl.assign(bytes.begin() + off, bytes.begin() + off + klen);
        }
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s) return false;
        if (!comp.empty()) {
            Steinberg::MemoryStream stream;
            stream.write(comp.data(), (int32)comp.size(), nullptr);
            stream.seek(0, Steinberg::IBStream::kIBSeekSet, nullptr);
            if (!s->component || s->component->setState(&stream) != kResultOk) return false;
        }
        if (!ctrl.empty()) {
            Steinberg::MemoryStream stream;
            stream.write(ctrl.data(), (int32)ctrl.size(), nullptr);
            stream.seek(0, Steinberg::IBStream::kIBSeekSet, nullptr);
            if (!s->controller || s->controller->setState(&stream) != kResultOk) return false;
        }
        std::cout << "[RenderFx] preset applied component=" << comp.size()
                  << " controller=" << ctrl.size() << std::endl;
        return true;
#endif
    }
    // Capture the plugin's CURRENT state (component + edit controller) as the
    // bridge blob [4B BE compLen][comp][4B BE ctrlLen][ctrl] base64 — same
    // format applyPreset accepts. Lets the GUI bridge report knob tweaks so a
    // saved project can restore them. Must run on the thread owning the plugin.
    void captureState(std::string& out) {
#ifndef HAVE_VST3SDK
        (void)out;
        return;
#else
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s) return;
        std::vector<uint8_t> comp, ctrl;
        {
            Steinberg::MemoryStream stream;
            if (s->component && s->component->getState(&stream) == kResultOk && stream.getSize() > 0)
                comp.assign(stream.getData(), stream.getData() + stream.getSize());
        }
        if (!s->controllerIsComponent) {
            Steinberg::MemoryStream stream;
            if (s->controller && s->controller->getState(&stream) == kResultOk && stream.getSize() > 0)
                ctrl.assign(stream.getData(), stream.getData() + stream.getSize());
        }
        if (comp.empty() && ctrl.empty()) return;
        std::vector<uint8_t> blob;
        auto append32 = [&blob](uint32_t v) {
            blob.push_back((uint8_t)(v >> 24));
            blob.push_back((uint8_t)(v >> 16));
            blob.push_back((uint8_t)(v >> 8));
            blob.push_back((uint8_t)v);
        };
        append32((uint32_t)comp.size());
        blob.insert(blob.end(), comp.begin(), comp.end());
        append32((uint32_t)ctrl.size());
        blob.insert(blob.end(), ctrl.begin(), ctrl.end());
        base64EncodeTo(blob.data(), blob.size(), out);
#endif
    }
    // Live param change (realtime chain, NO restart): address by lowercase
    // title/shortTitle or numeric ParamID; value normalized 0..1. Delivered
    // via setParamNormalized (controller/UI state) + inputParameterChanges
    // queue (split plugins read the queue in process() — same rule as
    // Vst3Instrument). ponytail: beginEdit/endEdit automation gestures when a
    // plugin ignores setParamNormalized (fallback: restart with new preset).
    bool setParam(const std::string& key, double value) {
#ifndef HAVE_VST3SDK
        (void)key; (void)value;
        return false;
#else
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s || !s->controller) return false;
        Steinberg::Vst::ParamID tag = 0;
        bool found = false;
        try {  // numeric ParamID passes through
            size_t consumed = 0;
            unsigned long num = std::stoul(key, &consumed);
            if (consumed == key.size()) { tag = (Steinberg::Vst::ParamID)num; found = true; }
        } catch (...) {}
        if (!found) {
            std::string k = key;
            std::transform(k.begin(), k.end(), k.begin(),
                           [](unsigned char c) { return (char)std::tolower(c); });
            auto it = s->paramByName.find(k);
            if (it == s->paramByName.end()) {
                std::cerr << "[RealtimeFx] vst3 setParam unknown key: " << key << std::endl;
                return false;
            }
            tag = it->second;
        }
        Steinberg::Vst::ParamValue v =
            (Steinberg::Vst::ParamValue)std::max(0.0, std::min(1.0, value));
        {
            std::lock_guard<std::mutex> lk(s->paramMutex);
            s->controller->setParamNormalized(tag, v);
            int32 idx = 0;
            if (Steinberg::Vst::IParamValueQueue* q =
                    s->paramChanges.addParameterData(tag, idx))
                q->addPoint(0, v, idx);
        }
        std::cout << "[RealtimeFx] vst3 setParam '" << key << "' -> tag " << tag
                  << " = " << v << std::endl;
        return true;
#endif
    }
    // Native editor: attach the plugin's IPlugView to a host window (must run
    // on the thread that will pump its messages — same rule as Vst3Instrument).
    bool openEditor(void* parentWindowHandle) {
#ifndef HAVE_VST3SDK
        (void)parentWindowHandle;
        return false;
#else
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s || !s->controller || !parentWindowHandle) return false;
        if (s->plugFrame) return true;  // already attached
        s->plugFrame = new HostPlugFrameFx();
        IPlugView* rawView = nullptr;
        s->controller->queryInterface(IPlugView::iid, (void**)&rawView);
        FUnknownPtr<IPlugView> view(rawView);
        if (!view) {
            // Some plugins expose IPlugView on the component instead.
            IPlugView* rawViewC = nullptr;
            s->component->queryInterface(IPlugView::iid, (void**)&rawViewC);
            view = FUnknownPtr<IPlugView>(rawViewC);
        }
        if (!view) {
            // Official editorhost.cpp pattern: IEditController::createView.
            // JUCE-based plugins (Scaler2) expose the editor only this way.
            view = owned(s->controller->createView(Steinberg::Vst::ViewType::kEditor));
        }
        if (!view) {
            delete s->plugFrame;
            s->plugFrame = nullptr;
            return false;
        }
        view->setFrame(s->plugFrame);
        view->isPlatformTypeSupported(kPlatformTypeHWND);
        if (view->attached(parentWindowHandle, kPlatformTypeHWND) != kResultOk) {
            view->setFrame(nullptr);
            delete s->plugFrame;
            s->plugFrame = nullptr;
            return false;
        }
        s->view = view;
#ifdef _WIN32
        HWND hwnd = (HWND)parentWindowHandle;
        Steinberg::ViewRect rect;
        if (view->getSize(&rect) == kResultOk) {
            int w = rect.right - rect.left;
            int h = rect.bottom - rect.top;
            RECT r = { 0, 0, w, h };
            // Size to the window's own style: borderless (WS_POPUP) needs no
            // caption/border adjustment.
            AdjustWindowRect(&r, GetWindowLongPtrA(hwnd, GWL_STYLE), FALSE);
            SetWindowPos(hwnd, nullptr, 0, 0, r.right - r.left, r.bottom - r.top,
                         SWP_NOMOVE | SWP_NOZORDER | SWP_NOACTIVATE);
            view->onSize(&rect);  // some Skia/OpenGL editors skip first paint
        }
#endif
        return true;
#endif
    }
    // Detach editor. Must run on the SAME thread that attached (the GUI message
    // loop has exited). removed() is intentionally NOT called — Vst3Instrument
    // documents a deadlock when the pump is not running.
    void closeEditor() {
#ifndef HAVE_VST3SDK
        return;
#else
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s) return;
        if (s->view) {
            s->view->setFrame(nullptr);
            s->view = nullptr;  // IPtr releases the view
        }
        if (s->plugFrame) {
            delete s->plugFrame;
            s->plugFrame = nullptr;
        }
#endif
    }
    // Resize the editor view to w x h (VST3 host resize flow: checkSizeConstraint
    // lets the plugin clamp to a supported size, host resizes the window, then
    // onSize tells the plugin -- it moves/resizes its own child window). Runs on
    // the GUI pump thread (WM_FXGUI_RESIZE) -- plugin calls must not come from
    // the HTTP worker thread.
    bool resizeEditor(void* parentWindowHandle, int w, int h) {
#ifndef HAVE_VST3SDK
        (void)parentWindowHandle; (void)w; (void)h;
        return false;
#else
        auto* s = static_cast<Vst3FxState*>(state_);
        if (!s || !s->view || w <= 0 || h <= 0) return false;
        Steinberg::ViewRect vr = {0, 0, w, h};
        s->view->checkSizeConstraint(&vr);  // plugin clamps to a supported size
        const int cw = vr.right - vr.left;
        const int ch = vr.bottom - vr.top;
        if (cw <= 0 || ch <= 0) return false;
#ifdef _WIN32
        RECT r = {0, 0, cw, ch};
        AdjustWindowRectEx(&r,
                           (DWORD)GetWindowLongPtrA((HWND)parentWindowHandle, GWL_STYLE),
                           FALSE,
                           (DWORD)GetWindowLongPtrA((HWND)parentWindowHandle, GWL_EXSTYLE));
        SetWindowPos((HWND)parentWindowHandle, nullptr, 0, 0,
                     r.right - r.left, r.bottom - r.top,
                     SWP_NOMOVE | SWP_NOZORDER | SWP_NOACTIVATE);
#endif
        s->view->onSize(&vr);
        return true;
#endif
    }
    bool loaded() const { return state_ != nullptr; }
    // Plugin-reported latency samples (PDC) — 0 nếu plugin không báo.
    int32 latencySamples() const {
#ifndef HAVE_VST3SDK
        return 0;
#else
        return state_ ? static_cast<Vst3FxState*>(state_)->latencySamples : 0;
#endif
    }

private:
    void* state_ = nullptr;
};

// One FX slot from the job. builtin id: "gain" | "normalize" (legacy) hoặc 8
// DSP (PLAN_DAW_A Phase 2) — id ∈ eq|eqpro|imager|maximizer|compressor|limiter
// |exciter|rebalance → bfx (BuiltinFxChain).
struct FxSlot {
    bool vst = false;        // true = VST3, false = builtin
    std::string path;        // vst only
    std::string presetB64;   // vst only
    bool bypass = false;
    std::string builtinId;   // builtin only
    double db = 0.0;         // gain param
    double peak = 0.95;      // normalize param
    std::unique_ptr<Vst3Fx> fx;
    std::unique_ptr<BuiltinFx> bfx;  // 8 DSP builtin (Phase 2)
};

}  // namespace

// SEH-guarded single-slot process call. Kept OUT of run_render_fx_job's frame
// (C2712: __try cannot live in a function whose C++ objects need unwinding).
// This frame holds only references/raw pointers — nothing to unwind.
static bool runFxSlotSafe(const FxSlot& sl, const float* inL, const float* inR,
                          float* outL, float* outR, uint32_t n, uint32_t* crashCode) {
#ifdef _WIN32
    __try {
        sl.fx->processAudio(inL, inR, outL, outR, n);
        return true;
    } __except (*crashCode = (uint32_t)GetExceptionCode(), EXCEPTION_EXECUTE_HANDLER) {
        return false;
    }
#else
    sl.fx->processAudio(inL, inR, outL, outR, n);
    return true;
#endif
}

// SEH-guarded builtin DSP slot (in-place). Frame chỉ raw pointers.
static bool runBuiltinSlotSafe(BuiltinFx* fx, float* L, float* R, uint32_t n,
                               uint32_t* crashCode) {
#ifdef _WIN32
    __try {
        fx->process(L, R, n);
        return true;
    } __except (*crashCode = (uint32_t)GetExceptionCode(), EXCEPTION_EXECUTE_HANDLER) {
        return false;
    }
#else
    fx->process(L, R, n);
    return true;
#endif
}

int run_render_fx_job(const std::string& jobPath, const std::string& inPath,
                      const std::string& outPath) {
#ifdef _WIN32
    CoInitializeEx(nullptr, COINIT_MULTITHREADED);
    struct ComGuard { ~ComGuard() { CoUninitialize(); } } comGuard;
#endif

    int rc = 1;
    std::ofstream wav;
    auto fail = [&](const std::string& msg) {
        std::cerr << "[RenderFx] " << msg << std::endl;
        if (wav.is_open()) wav.close();
        std::remove(outPath.c_str());
    };

    // 1. Read input WAV.
    WavIn in;
    if (!readWav(inPath, in)) { fail("cannot read input WAV: " + inPath); return rc; }
    if (in.L.empty()) { fail("input WAV is empty"); return rc; }

    // 2. Parse job.
    std::ifstream jf(jobPath, std::ios::binary);
    if (!jf) { fail("cannot read job file: " + jobPath); return rc; }
    std::string data((std::istreambuf_iterator<char>(jf)), std::istreambuf_iterator<char>());
    json_parse_result_s pres = {};
    json_value_s* root = json_parse_ex(data.data(), data.size(), json_parse_flags_default,
                                       nullptr, nullptr, &pres);
    if (!root) { fail("job JSON parse error (code " + std::to_string(pres.error) + ")"); return rc; }
    struct RootHolder { json_value_s* p = nullptr; ~RootHolder() { if (p) std::free(p); } } rootH;
    rootH.p = root;
    if (root->type != json_type_object) { fail("job root must be a JSON object"); return rc; }
    const json_object_s* job = static_cast<const json_object_s*>(root->payload);

    double srD = (double)in.sampleRate;
    memberNumber(job, "sample_rate", srD);
    if (srD < 8000.0 || srD > 192000.0) { fail("bad sample_rate: " + std::to_string(srD)); return rc; }
    const uint32_t sampleRate = (uint32_t)srD;
    if (in.sampleRate != sampleRate) {
        fail("input WAV sr (" + std::to_string(in.sampleRate) + ") != job sample_rate ("
             + std::to_string(sampleRate) + ")"); return rc;
    }
    int64_t block = memberInt(job, "block_size", 512);
    block = std::max<int64_t>(32, std::min<int64_t>(block, 4096));
    const uint32_t chunk = (uint32_t)block;

    // 3. Build slot list.
    std::vector<FxSlot> slots;
    const json_value_s* chainV = memberValue(job, "fx_chain");
    if (chainV && chainV->type == json_type_array) {
        const json_array_s* chain = static_cast<const json_array_s*>(chainV->payload);
        for (const json_array_element_s* el = chain->start; el; el = el->next) {
            const json_value_s* v = el->value;
            if (!v || v->type != json_type_object) continue;
            const json_object_s* o = static_cast<const json_object_s*>(v->payload);
            const std::string type = memberString(o, "type", "");
            FxSlot sl;
            sl.bypass = memberBool(o, "bypass", false);
            if (type == "vst3" || type == "vst") {
                sl.vst = true;
                sl.path = memberString(o, "path", "");
                sl.presetB64 = memberString(o, "preset_b64", "");
            } else if (type == "builtin") {
                sl.vst = false;
                sl.builtinId = memberString(o, "id", "");
                const json_object_s* po = nullptr;
                const json_value_s* pv = memberValue(o, "params");
                if (pv && pv->type == json_type_object)
                    po = static_cast<const json_object_s*>(pv->payload);
                double d = sl.db; if (po) memberNumber(po, "db", d); sl.db = d;
                double pk = sl.peak; if (po) memberNumber(po, "peak", pk); sl.peak = pk;
                if (sl.builtinId != "gain" && sl.builtinId != "normalize") {
                    // 8 DSP (PLAN_DAW_A Phase 2) — khớp Python _apply_builtin_fx_chain.
                    sl.bfx = createBuiltinFx(sl.builtinId, po, srD);
                    if (!sl.bfx) {
                        fail("unknown builtin id: " + sl.builtinId); return rc;
                    }
                }
            } else {
                fail("unknown fx_chain slot type: " + type); return rc;
            }
            slots.push_back(std::move(sl));
        }
    }
    if (slots.empty()) { fail("fx_chain is empty — nothing to render"); return rc; }

    // 4. Load VST3 slots (bypassed slots stay unloaded → passthrough).
    for (auto& sl : slots) {
        if (!sl.vst || sl.bypass) continue;
        if (sl.path.empty()) { fail("vst3 slot missing path"); return rc; }
        auto fx = std::make_unique<Vst3Fx>();
        if (!fx->load(sl.path, sampleRate, (int32)chunk)) { rc = 2; fail("VST3 FX load failed: " + sl.path); return rc; }
        if (!sl.presetB64.empty() && !fx->applyPreset(sl.presetB64)) {
            fail("preset import FAILED for " + sl.path); return rc;
        }
        sl.fx = std::move(fx);
    }

    // 4b. Per-slot latency (PDC, Phase 3): vst3 = getLatencySamples() (0 nếu
    // chưa load/bypass), builtin = 0. 1 dòng stdout để native_render parse →
    // render_engine align track (cùng thứ tự mảng slots).
    std::cout << "[RenderFx] FX_LATENCIES";
    for (const auto& sl : slots) {
        uint32_t lat = 0;
        if (sl.vst && !sl.bypass && sl.fx) lat = (uint32_t)sl.fx->latencySamples();
        std::cout << ' ' << lat;
    }
    std::cout << std::endl;

    // 5. Process chain block-wise. Double-buffered: cur ← slot → nxt, swap.
    std::vector<float> L0(chunk), R0(chunk), L1(chunk), R1(chunk);
    std::vector<float> curL(chunk), curR(chunk);
    bool normalizePending = false;
    double normalizePeak = 0.95;
    uint32_t crashCode = 0;
    for (const auto& sl : slots) {
        if (sl.bypass) continue;
        if (sl.vst && !sl.fx) continue;  // defensive: bypassed/unloaded
        if (!sl.vst && sl.builtinId == "normalize") {
            normalizePending = true;
            normalizePeak = std::max(0.01, std::min(1.0, sl.peak));
            continue;  // global post-pass
        }
        const bool isVst = sl.vst;
        const double gainLin = sl.vst ? 1.0 : std::pow(10.0, sl.db / 20.0);
        for (uint64_t pos = 0; pos < in.L.size(); pos += chunk) {
            const uint32_t n = (uint32_t)std::min<uint64_t>(chunk, in.L.size() - pos);
            std::memcpy(curL.data(), in.L.data() + pos, n * sizeof(float));
            std::memcpy(curR.data(), in.R.data() + pos, n * sizeof(float));
            if (isVst) {
                bool ok = runFxSlotSafe(sl, curL.data(), curR.data(), L0.data(), R0.data(),
                                        n, &crashCode);
                if (!ok) { rc = 3; fail("VST3 FX crashed inside processAudio (SEH code=0x"
                                        + std::to_string(crashCode) + ")"); return rc; }
                std::memcpy(in.L.data() + pos, L0.data(), n * sizeof(float));
                std::memcpy(in.R.data() + pos, R0.data(), n * sizeof(float));
            } else if (sl.bfx) {
                // Builtin DSP (8 loại) — in-place, xen kẽ VST3 đúng thứ tự UI.
                bool ok = runBuiltinSlotSafe(sl.bfx.get(), in.L.data() + pos,
                                             in.R.data() + pos, n, &crashCode);
                if (!ok) {
                    rc = 3; fail("builtin DSP crashed (SEH code=0x"
                                 + std::to_string(crashCode) + ")"); return rc;
                }
            } else {
                for (uint32_t i = 0; i < n; ++i) {
                    in.L[pos + i] = curL[i] * (float)gainLin;
                    in.R[pos + i] = curR[i] * (float)gainLin;
                }
            }
        }
        if (isVst) std::cout << "[RenderFx] slot processed: " << sl.path << std::endl;
        else std::cout << "[RenderFx] slot processed: builtin " << sl.builtinId << std::endl;
    }

    // 6. Global normalize pass (only reduces when peak exceeds target — mirrors
    //    render_project's clip guard; the final mastering slot by convention).
    if (normalizePending) {
        float peak = 0.f;
        for (float v : in.L) peak = std::max(peak, std::fabs(v));
        for (float v : in.R) peak = std::max(peak, std::fabs(v));
        if (peak > (float)normalizePeak && peak > 1e-9f) {
            const float k = (float)normalizePeak / peak;
            for (float& v : in.L) v *= k;
            for (float& v : in.R) v *= k;
            std::cout << "[RenderFx] normalize: peak " << peak << " → " << normalizePeak << std::endl;
        } else {
            std::cout << "[RenderFx] normalize: peak " << peak << " ≤ " << normalizePeak << " (no-op)" << std::endl;
        }
    }

    // 7. Write output WAV.
    wav.open(outPath, std::ios::binary);
    if (!wav) { fail("cannot open output: " + outPath); return rc; }
    writeWavHeader(wav, sampleRate);
    uint64_t dataBytes = 0;
    for (uint64_t pos = 0; pos < in.L.size(); pos += chunk) {
        const uint32_t n = (uint32_t)std::min<uint64_t>(chunk, in.L.size() - pos);
        writeFrames(wav, in.L.data() + pos, in.R.data() + pos, n);
        dataBytes += (uint64_t)n * 8;  // stereo float32: 2ch * 4B
    }
    finishWav(wav, dataBytes);
    wav.close();
    std::cout << "[RenderFx] wrote " << outPath << " (" << in.L.size() << " frames)" << std::endl;
    rc = 0;
    return rc;
}

// --- RealtimeFxChain (T2.4) ---------------------------------------------------
// Live VST3 FX on the bridge master mix. setChain() enqueues JSON; a worker
// thread (own COM init) parses + loads plugins then swaps the chain under the
// mutex. The audio thread copies the shared_ptr (cheap, uncontended) and runs
// the plugins in-place — Vst3Fx::processAudio copies input to owned buffers
// before writing output, so out==in is safe. Old chains are retired on the
// worker so plugin destructors never run on the audio thread.
namespace {

// SEH frame holds only raw pointers / trivials (C2712 — no unwinding locals).
static void realtimeRunChain(Vst3Fx** vst, BuiltinFx** bfx, const bool* bypass,
                             uint32_t count, float* inL, float* inR, uint32_t n) {
#ifdef _WIN32
    __try {
        for (uint32_t i = 0; i < count; ++i) {
            if (bypass && bypass[i]) continue;
            if (vst[i]) vst[i]->processAudio(inL, inR, inL, inR, n);  // in-place safe
            else if (bfx[i]) bfx[i]->process(inL, inR, n);           // builtin in-place
        }
    } __except (EXCEPTION_EXECUTE_HANDLER) {
        std::cerr << "[RealtimeFx] chain crashed in process — slot bypassed" << std::endl;
    }
#else
    (void)vst; (void)bfx; (void)bypass; (void)count; (void)inL; (void)inR; (void)n;
#endif
}

}  // namespace

struct RealtimeFxChain::Impl {
    // VST3 (fx) hoặc builtin DSP (bfx) — 1 trong 2; xen kẽ đúng thứ tự UI.
    struct Entry {
        std::shared_ptr<Vst3Fx> fx;
        std::unique_ptr<BuiltinFx> bfx;
        bool bypass = false;
    };
    struct Chain { std::vector<Entry> entries; };

    // SET_PARAM chờ chain swap (setParam tới trước setChain): worker áp sau
    // khi swap — không mất param race lúc khởi động.
    struct PendingParam { int slot; std::string key; double value; };
    std::mutex pmutex_;                    // guards pending_
    std::vector<PendingParam> pending_;
    std::atomic<int> paramQueued_{0};      // wake worker: param chờ apply

    std::mutex mutex_;               // guards chain_ / retired_
    std::shared_ptr<Chain> chain_;
    std::shared_ptr<Chain> retired_; // dtor deferred to the worker thread
    std::atomic<uint64_t> gen_{0};   // chain generation (Phase 2.8)

    std::mutex qmutex_;              // guards q_ / quit_
    std::condition_variable qcv_;
    std::vector<std::string> q_;
    bool quit_ = false;

    std::thread worker_;
    double sampleRate_ = 44100.0;
    int32_t blockSize_ = 256;

    // VST3 instance cache: giữ plugin đã load qua các lần buildChain để toggle
    // bypass không phải load lại từ disk (tránh delay dry->wet). Key = path + #occ.
    std::map<std::string, std::shared_ptr<Vst3Fx>> cache_;
    std::map<std::string, std::string> lastPreset_;  // preset b64 da ap (tranh ap lai)
    std::map<std::string, int> occ_;                 // so lan xuat hien cua path trong chain

    std::shared_ptr<Chain> buildChain(const std::string& json);
    void workerLoop();
};

RealtimeFxChain::RealtimeFxChain() : impl_(std::make_unique<Impl>()) {
    impl_->worker_ = std::thread([this]() { impl_->workerLoop(); });
}

RealtimeFxChain::~RealtimeFxChain() { shutdown(); }

void RealtimeFxChain::shutdown() {
    if (!impl_) return;
    {
        std::lock_guard<std::mutex> lk(impl_->qmutex_);
        if (impl_->quit_) return;
        impl_->quit_ = true;
    }
    impl_->qcv_.notify_all();
    if (impl_->worker_.joinable()) impl_->worker_.join();
    {
        std::lock_guard<std::mutex> lk(impl_->mutex_);
        impl_->chain_.reset();
        impl_->retired_.reset();
    }
}

void RealtimeFxChain::setChain(const std::string& json, double sampleRate, int32_t blockSize) {
    if (!impl_) return;
    {
        std::lock_guard<std::mutex> lk(impl_->qmutex_);
        if (sampleRate > 0) impl_->sampleRate_ = sampleRate;
        if (blockSize > 0) impl_->blockSize_ = blockSize;
        impl_->q_.push_back(json);
    }
    impl_->qcv_.notify_all();
}

void RealtimeFxChain::process(float* inL, float* inR, uint32_t n) {
    if (!impl_ || n == 0) return;
    std::shared_ptr<Impl::Chain> c;
    {
        std::lock_guard<std::mutex> lk(impl_->mutex_);
        c = impl_->chain_;
    }
    if (!c || c->entries.empty()) return;
    constexpr uint32_t kMaxSlots = 16;  // v1 cap
    Vst3Fx* vst[kMaxSlots];
    BuiltinFx* bfx[kMaxSlots];
    bool bypass[kMaxSlots];
    uint32_t count = 0;
    for (const auto& e : c->entries) {
        if (count >= kMaxSlots) break;
        vst[count] = e.fx.get();
        bfx[count] = e.bfx.get();
        bypass[count] = e.bypass;
        ++count;
    }
    realtimeRunChain(vst, bfx, bypass, count, inL, inR, n);
}

void RealtimeFxChain::setParam(int slot, const std::string& key, double value) {
    // SET_PARAM (SHM control ring, Phase 2.8): KHÔNG gọi VST API trên audio
    // thread (chain.setParam chạy từ RealtimeFxLoop drain ctrl ring) —
    // controller->setParamNormalized là COM call cross-thread với GUI pump,
    // block ms→chục ms → out ring cạn → worklet underrun → crackle. Mọi
    // param defer vào pending_, worker thread áp (builtin nhanh, VST3 an
    // toàn) — audio thread chỉ push queue + notify.
    if (!impl_) return;
    {
        std::lock_guard<std::mutex> lk(impl_->pmutex_);
        impl_->pending_.push_back({slot, key, value});
    }
    impl_->paramQueued_.fetch_add(1, std::memory_order_release);
    impl_->qcv_.notify_all();
}

uint64_t RealtimeFxChain::chainGen() {
    return impl_ ? impl_->gen_.load(std::memory_order_acquire) : 0;
}

std::vector<uint32_t> RealtimeFxChain::entryLatencies() {
    std::vector<uint32_t> out;
    if (!impl_) return out;
    std::shared_ptr<Impl::Chain> c;
    { std::lock_guard<std::mutex> lk(impl_->mutex_); c = impl_->chain_; }
    if (!c) return out;
    out.reserve(c->entries.size());
    for (const auto& e : c->entries)
        out.push_back((!e.bypass && e.fx) ? (uint32_t)e.fx->latencySamples() : 0u);
    return out;
}

std::shared_ptr<RealtimeFxChain::Impl::Chain> RealtimeFxChain::Impl::buildChain(const std::string& json) {
    auto chain = std::make_shared<Chain>();
    occ_.clear();  // key theo vị trí slot trong CHAIN NÀY — reset mỗi build
    json_parse_result_s pres = {};
    json_value_s* root = json_parse_ex(json.data(), json.size(), json_parse_flags_default,
                                       nullptr, nullptr, &pres);
    if (!root || root->type != json_type_array) {
        if (root) std::free(root);
        std::cerr << "[RealtimeFx] chain JSON parse error" << std::endl;
        return chain;  // empty chain = passthrough
    }
    const json_array_s* arr = static_cast<const json_array_s*>(root->payload);
    for (const json_array_element_s* e = arr->start; e; e = e->next) {
        if (!e->value || e->value->type != json_type_object) continue;
        const json_object_s* o = static_cast<const json_object_s*>(e->value->payload);
        const std::string type = memberString(o, "type", "");
        if (type == "builtin") {
            // Builtin DSP (Phase 2): id ∈ 8 loại — xen kẽ VST3 đúng thứ tự.
            const std::string id = memberString(o, "id", "");
            if (id.empty()) continue;
            const json_object_s* po = nullptr;
            const json_value_s* pv = memberValue(o, "params");
            if (pv && pv->type == json_type_object)
                po = static_cast<const json_object_s*>(pv->payload);
            Entry ent;
            ent.bypass = memberBool(o, "bypass", false);
            ent.bfx = createBuiltinFx(id, po, sampleRate_);
            if (!ent.bfx) {
                std::cerr << "[RealtimeFx] unknown builtin id — slot skipped: " << id << std::endl;
                continue;
            }
            chain->entries.push_back(std::move(ent));
            continue;
        }
        const std::string path = memberString(o, "path", "");
        if (path.empty()) continue;
        Entry ent;
        ent.bypass = memberBool(o, "bypass", false);
        const std::string preset = memberString(o, "preset_b64", "");
        // Key ổn định theo vị trí slot: đếm MỌI lần path xuất hiện (kể cả
        // bypassed) — toggle bypass không đổi key → không reload nhầm instance.
        const std::string ckey = path + "#" + std::to_string(occ_[path]++);
        if (!ent.bypass) {
            auto it = cache_.find(ckey);
            if (it != cache_.end()) {
                ent.fx = it->second;  // đã load: reuse ngay, khong reload
                const std::string last = lastPreset_[ckey];
                if (!preset.empty() && preset != last) {
                    if (ent.fx->applyPreset(preset))
                        std::cerr << "[RealtimeFx] preset applied (cached): " << path << std::endl;
                    else
                        std::cerr << "[RealtimeFx] preset apply FAILED (default kept): " << path << std::endl;
                    lastPreset_[ckey] = preset;
                }
                if (preset.empty()) lastPreset_[ckey] = "";
            }
            else {
                auto fx = std::make_shared<Vst3Fx>();
                if (fx->load(path, sampleRate_, blockSize_)) {
                    if (!preset.empty()) {
                        if (fx->applyPreset(preset))
                            std::cerr << "[RealtimeFx] preset applied: " << path << std::endl;
                        else
                            std::cerr << "[RealtimeFx] preset apply FAILED (default kept): " << path << std::endl;
                    }
                    ent.fx = fx;
                    cache_[ckey] = fx;
                    lastPreset_[ckey] = preset;
                }
                else {
                    std::cerr << "[RealtimeFx] load failed — slot bypassed: " << path << std::endl;
                    ent.bypass = true;
                    occ_[path]--;  // khong dem lan that bai
                }
            }
        }
        chain->entries.push_back(std::move(ent));
    }
    std::free(root);
    std::cout << "[RealtimeFx] chain applied: " << chain->entries.size() << " slot(s)" << std::endl;
    return chain;
}

void RealtimeFxChain::Impl::workerLoop() {
#ifdef _WIN32
    CoInitializeEx(nullptr, COINIT_MULTITHREADED);
#endif
    for (;;) {
        std::string job;
        {
            std::unique_lock<std::mutex> lk(qmutex_);
            qcv_.wait(lk, [&]() {
                return quit_ || !q_.empty() || paramQueued_.load(std::memory_order_acquire) > 0;
            });
            if (quit_) break;
            if (!q_.empty()) {
                job = std::move(q_.front());
                q_.erase(q_.begin());
            }
        }
        if (!job.empty()) {
            std::shared_ptr<Chain> next = buildChain(job);  // slow: plugin load
            {
                std::lock_guard<std::mutex> lk(mutex_);
                retired_ = std::move(chain_);  // old chain destroyed on THIS thread
                chain_ = std::move(next);
            }
            gen_.fetch_add(1, std::memory_order_release);  // loop báo lại latency
        }
        {
            // Áp param chờ trên WORKER thread (audio thread không bao giờ gọi
            // VST API): queue trước swap áp vào chain mới; queue trong lúc
            // chain tồn tại áp ngay, không cần đợi setChain.
            std::vector<Impl::PendingParam> toApply;
            {
                std::lock_guard<std::mutex> lk(pmutex_);
                if (!pending_.empty()) {
                    toApply.swap(pending_);
                    paramQueued_.store(0, std::memory_order_release);
                }
            }
            if (!toApply.empty()) {
                std::shared_ptr<Chain> c;
                { std::lock_guard<std::mutex> lk(mutex_); c = chain_; }
                if (c) {
                    for (const auto& p : toApply) {
                        if (p.slot < 0 || (size_t)p.slot >= c->entries.size()) continue;
                        auto& e = c->entries[(size_t)p.slot];
                        if (e.bfx) e.bfx->setParam(p.key, p.value);
                        else if (e.fx) e.fx->setParam(p.key, p.value);
                    }
                } else {
                    // Chain chưa tồn tại — stash lại, áp sau swap.
                    std::lock_guard<std::mutex> lk(pmutex_);
                    pending_.insert(pending_.begin(), toApply.begin(), toApply.end());
                    paramQueued_.fetch_add((int)toApply.size(), std::memory_order_release);
                }
            }
        }
    }
#ifdef _WIN32
    CoUninitialize();
#endif
}

// --- --open-fx-gui <job.json>: native editor window for one VST3 FX ----------
#ifdef _WIN32
// Worker (HTTP) thread -> pump thread editor-resize request. Declared
// outside the anonymous namespace: the definition sits after the namespace.
static void fxGuiResizeSafe(void* fx, HWND hwnd, int w, int h);
static void fxGuiCaptureSafe(void* fx, std::string& out);

namespace {

const char* kFxGuiWndClass = "SonicForge_FX_GUI_Class";

struct FxGuiCtx {
    bool closing = false;
};
// Parent watchdog: engine spawn fx-gui voi SF_PARENT_PID=<engine pid>. Neu
// engine chet (crash/quit) ma bridge fx-gui khong tu thoat -> cua so topmost
// van lo lung, khong tat duoc. Watchdog thread: parent mat -> WM_CLOSE.
static uint32_t fxGuiParentPid() {
    const char* e = std::getenv("SF_PARENT_PID");
    return e ? (uint32_t)std::atoi(e) : 0;
}
static bool fxGuiParentAlive(uint32_t pid) {
    if (pid == 0) return true;
    HANDLE h = OpenProcess(PROCESS_QUERY_LIMITED_INFORMATION, FALSE, pid);
    if (!h) return false;
    CloseHandle(h);
    return true;
}
static void fxGuiStartWatchdog(HWND hwnd, uint32_t parentPid) {
    if (parentPid == 0) return;
    std::thread([hwnd, parentPid]() {
        while (IsWindow(hwnd)) {
            Sleep(2000);
            if (!fxGuiParentAlive(parentPid)) {
                PostMessageA(hwnd, WM_CLOSE, 0, 0);  // pump exits -> process returns
                return;
            }
        }
    }).detach();
}

// Worker (HTTP) thread -> pump thread editor-resize request. The plugin's
// checkSizeConstraint/onSize must run on the thread that owns the editor.
#define WM_FXGUI_RESIZE (WM_APP + 0x51)

// --- preset capture (GUI state -> project save) ------------------------
// getState must run on the pump thread that owns the plugin, so the HTTP
// worker posts WM_FXGUI_CAPTURE and waits (up to timeoutMs) for the pump
// thread to park the base64 blob. Empty outB64 = timeout or no state.
#define WM_FXGUI_CAPTURE (WM_APP + 0x52)
static std::mutex& gCapMutex() { static std::mutex m; return m; }
static std::condition_variable& gCapCv() { static std::condition_variable cv; return cv; }
static std::string& gCapB64() { static std::string s; return s; }
static bool& gCapDone() { static bool b = false; return b; }

LRESULT CALLBACK FxGuiWndProc(HWND hwnd, UINT msg, WPARAM wp, LPARAM lp) {
    FxGuiCtx* ctx = (FxGuiCtx*)GetWindowLongPtrA(hwnd, GWLP_USERDATA);
    switch (msg) {
        case WM_CLOSE:
            DestroyWindow(hwnd);
            return 0;
        case WM_DESTROY:
            if (ctx) ctx->closing = true;
            return 0;
        case WM_FXGUI_RESIZE:
            fxGuiResizeSafe((void*)GetPropA(hwnd, "FXGUI_FX"), hwnd, (int)wp, (int)lp);
            return 0;
        case WM_FXGUI_CAPTURE: {
            std::string b64;
            fxGuiCaptureSafe((void*)GetPropA(hwnd, "FXGUI_FX"), b64);
            {
                std::lock_guard<std::mutex> lk(gCapMutex());
                gCapB64() = std::move(b64);
                gCapDone() = true;
            }
            gCapCv().notify_one();
            return 0;
        }
        default:
            return DefWindowProcA(hwnd, msg, wp, lp);
    }
}

}  // namespace

// SEH frames hold only raw pointers / trivials (C2712 — no unwinding locals).
static bool fxGuiAttachSafe(void* fx, HWND hwnd) {
#ifdef HAVE_VST3SDK
    __try { return static_cast<Vst3Fx*>(fx)->openEditor(hwnd); }
    __except (EXCEPTION_EXECUTE_HANDLER) { return false; }
#else
    (void)fx; (void)hwnd;
    return false;
#endif
}
static void fxGuiDetachSafe(void* fx) {
#ifdef HAVE_VST3SDK
    __try { static_cast<Vst3Fx*>(fx)->closeEditor(); }
    __except (EXCEPTION_EXECUTE_HANDLER) {}
#else
    (void)fx;
#endif
}
static void fxGuiResizeSafe(void* fx, HWND hwnd, int w, int h) {
#ifdef HAVE_VST3SDK
    __try { static_cast<Vst3Fx*>(fx)->resizeEditor(hwnd, w, h); }
    __except (EXCEPTION_EXECUTE_HANDLER) {}
#else
    (void)fx; (void)hwnd; (void)w; (void)h;
#endif
}
static void fxGuiCaptureSafe(void* fx, std::string& out) {
#ifdef HAVE_VST3SDK
    __try { static_cast<Vst3Fx*>(fx)->captureState(out); }
    __except (EXCEPTION_EXECUTE_HANDLER) {}
#else
    (void)fx; (void)out;
#endif
}
// Worker (HTTP) thread asks the GUI pump thread to resize the plugin view.
void fxGuiRequestResize(HWND hwnd, int w, int h) {
    PostMessageA(hwnd, WM_FXGUI_RESIZE, (WPARAM)w, (LPARAM)h);
}

// Capture the GUI plugin state on the pump thread; blocks up to timeoutMs.
void fxGuiRequestCapture(void* hwndV, std::string& outB64, int timeoutMs) {
    HWND hwnd = (HWND)hwndV;
    {
        std::lock_guard<std::mutex> lk(gCapMutex());
        gCapDone() = false;
        gCapB64().clear();
    }
    PostMessageA(hwnd, WM_FXGUI_CAPTURE, 0, 0);
    {
        std::unique_lock<std::mutex> lk(gCapMutex());
        if (!gCapCv().wait_for(lk, std::chrono::milliseconds(timeoutMs), [] { return gCapDone(); }))
            return;  // timeout - outB64 stays empty
        outB64 = gCapB64();
    }
}


// --- SHM audio feeder (GUI meters) -----------------------------------------
// The GUI process PEEKS the SAME input ring the realtime session uses (never
// consumes -> no contention with RealtimeFxLoop) and runs fx->processAudio()
// on the GUI plugin instance so its own meters/UI move with real audio.
// Meters show the PRE-chain mix (input ring peek); per-plugin position in the
// chain is approximated (ponytail: feed from a per-plugin output ring if exact
// post-FX metering is ever needed).
struct FxGuiShmCtl {
    std::mutex m;
    std::string name;  // "" = no live session to feed
};
static std::shared_ptr<FxGuiShmCtl>& fxGuiShmCtlRef() {
    static std::shared_ptr<FxGuiShmCtl> p;
    return p;
}
// Called from FxGuiServer /shm route (HTTP worker thread) when the frontend
// reports which realtime session is live (or "" after it stopped/restarted).
void fxGuiSetShm(const std::string& name) {
    auto p = fxGuiShmCtlRef();
    if (!p) return;
    std::lock_guard<std::mutex> lk(p->m);
    p->name = name;
}

struct FxGuiShmView { void* map = nullptr; void* view = nullptr; };
static FxGuiShmView* fxGuiOpenShm(const std::string& name, size_t size) {
    int wlen = MultiByteToWideChar(CP_UTF8, 0, name.c_str(), -1, nullptr, 0);
    std::wstring wname(wlen, L'\0');
    MultiByteToWideChar(CP_UTF8, 0, name.c_str(), -1, &wname[0], wlen);
    HANDLE map = OpenFileMappingW(FILE_MAP_ALL_ACCESS, FALSE, wname.c_str());
    if (!map) return nullptr;
    void* view = MapViewOfFile(map, FILE_MAP_ALL_ACCESS, 0, 0, size);
    if (!view) { CloseHandle(map); return nullptr; }
    FxGuiShmView* v = new FxGuiShmView();
    v->map = map;
    v->view = view;
    return v;
}
static void fxGuiCloseShm(FxGuiShmView* v) {
    if (!v) return;
    if (v->view) UnmapViewOfFile(v->view);
    if (v->map) CloseHandle((HANDLE)v->map);
    delete v;
}

// SEH frame holds only raw pointers / trivials (C2712 - no unwinding locals).
static void fxGuiFeedAudioSafe(Vst3Fx* fx, const float* inL, const float* inR,
                               float* outL, float* outR, uint32_t n) {
#ifdef _WIN32
    __try { fx->processAudio(inL, inR, outL, outR, n); }
    __except (EXCEPTION_EXECUTE_HANDLER) {
        std::cerr << "[FxGui] plugin crashed in audio feed" << std::endl;
    }
#else
    (void)fx; (void)inL; (void)inR; (void)outL; (void)outR; (void)n;
#endif
}

static void fxGuiFeedLoop(Vst3Fx* fx, HWND hwnd, std::shared_ptr<FxGuiShmCtl> ctl) {
    FxGuiShmView* v = nullptr;
    FxRealtimeIPC* ipc = nullptr;
    std::string curName;
    uint64_t lastWrite = 0;  // feeder-local: newest inWrite already peeked
    auto nextTick = std::chrono::steady_clock::now();
    while (IsWindow(hwnd)) {
        std::string name;
        {
            std::lock_guard<std::mutex> lk(ctl->m);
            name = ctl->name;
        }
        if (name != curName) {
            if (v) { fxGuiCloseShm(v); v = nullptr; ipc = nullptr; }
            curName = name;
            lastWrite = 0;  // session (re)start -> re-anchor
        }
        if (!v && !name.empty()) {  // retry open each pass: mapping may appear late
            v = fxGuiOpenShm(name, sizeof(FxRealtimeIPC));
            ipc = v ? static_cast<FxRealtimeIPC*>(v->view) : nullptr;
        }
        if (ipc && ipc->h.magic == FXRT_MAGIC &&
            ipc->h.inWrite > lastWrite) {
            if (ipc->h.inSlots == FXRT_IN_SLOTS && ipc->h.blockSize == FXRT_BLOCK) {
                const uint32_t slot = (ipc->h.inWrite - 1) & (ipc->h.inSlots - 1);
                float L[FXRT_BLOCK], R[FXRT_BLOCK], oL[FXRT_BLOCK], oR[FXRT_BLOCK];
                std::memcpy(L, ipc->inL[slot], FXRT_BLOCK * sizeof(float));
                std::memcpy(R, ipc->inR[slot], FXRT_BLOCK * sizeof(float));
                fxGuiFeedAudioSafe(fx, L, R, oL, oR, FXRT_BLOCK);
                lastWrite = ipc->h.inWrite;
            }
        }
        // Pace the feeder to realtime (one FXRT_BLOCK per its audio duration,
        // 44.1k fallback) instead of Sleep(2) busy-spin (~3x realtime CPU).
        // Less CPU in the GUI process = less contention with the pump thread,
        // which was stalling /input (dead clicks) and /preset (crackle reloads).
        const uint32_t sr = (ipc && ipc->h.sampleRate > 0) ? ipc->h.sampleRate : 44100u;
        nextTick += std::chrono::microseconds(1000000ULL * FXRT_BLOCK / sr);
        const auto now = std::chrono::steady_clock::now();
        if (now > nextTick + std::chrono::milliseconds(50))
            nextTick = now;  // fell behind (plugin hiccup): resync, don't burst
        std::this_thread::sleep_until(nextTick);
    }
    if (v) fxGuiCloseShm(v);
}

// Split feeder setup: fxGuiShmSetName runs EARLY (before the server loop) so
// /shm pushes from the frontend are not lost, while the detached thread starts
// only AFTER load+attach succeeded — a feeder calling processAudio() mid-load
// can crash/corrupt a heavy plugin and break the editor attach.
// ponytail: feeder feeds the raw live pre-chain mix; add a separate audio path
// if GUI processing of the full insert chain is wanted later.
static void fxGuiShmSetName(const std::string& initialShm) {
    auto p = fxGuiShmCtlRef();
    if (!p) { p = std::make_shared<FxGuiShmCtl>(); fxGuiShmCtlRef() = p; }
    std::lock_guard<std::mutex> lk(p->m);
    p->name = initialShm;
}
static void fxGuiStartFeederThread(Vst3Fx* fx, HWND hwnd) {
    auto p = fxGuiShmCtlRef();
    if (!p) { p = std::make_shared<FxGuiShmCtl>(); fxGuiShmCtlRef() = p; }
    std::thread([fx, hwnd, p]() { fxGuiFeedLoop(fx, hwnd, p); }).detach();
}

// --- split setup for server mode (startup speed) -----------------------------
// 1. Parse job { path, name } — fast. Returns 0 ok, 1 job/argument error.
static int fxGuiParseJob(const std::string& jobPath, std::string& path, std::string& name, std::string& shm, std::string& presetB64) {
    std::ifstream jf(jobPath, std::ios::binary);
    if (!jf) { std::cerr << "[FxGui] cannot read job file: " << jobPath << std::endl; return 1; }
    std::string data((std::istreambuf_iterator<char>(jf)), std::istreambuf_iterator<char>());
    json_parse_result_s pres = {};
    json_value_s* root = json_parse_ex(data.data(), data.size(), json_parse_flags_default,
                                       nullptr, nullptr, &pres);
    if (!root) { std::cerr << "[FxGui] job JSON parse error (code " << std::to_string(pres.error) << ")" << std::endl; return 1; }
    struct RootHolder { json_value_s* p = nullptr; ~RootHolder() { if (p) std::free(p); } } rootH;
    rootH.p = root;
    if (root->type != json_type_object) { std::cerr << "[FxGui] job root must be a JSON object" << std::endl; return 1; }
    const json_object_s* job = static_cast<const json_object_s*>(root->payload);
    path = memberString(job, "path", "");
    name = memberString(job, "name", "");
    shm = memberString(job, "shm", "");
    presetB64 = memberString(job, "preset_b64", "");
    if (path.empty()) { std::cerr << "[FxGui] job missing path" << std::endl; return 1; }
    return 0;
}

// 2. Create the host window — fast. Embedded mode parks it far off-screen
//    (never steals focus; still visible to GDI so BitBlt captures real pixels).
//    WS_EX_TOOLWINDOW: an off-screen visible window would otherwise sit in the
//    taskbar; clicking its button "restores" it to -20000,-20000 and it looks
//    minimized/unrestorable. Tool window has no taskbar button. WS_MINIMIZEBOX
//    is dropped too (a minimized tool window is hard to get back while parked).
//    Returns 0 ok, 1 window creation error.
static int fxGuiCreateWindow(const std::string& name, bool offscreen,
                             LONG_PTR userData, HWND* hwndOut) {
    WNDCLASSA wc = {};
    wc.lpfnWndProc = FxGuiWndProc;
    wc.hInstance = GetModuleHandleA(nullptr);
    wc.lpszClassName = kFxGuiWndClass;
    RegisterClassA(&wc);
    const int x = offscreen ? -20000 : CW_USEDEFAULT;
    const int y = offscreen ? -20000 : CW_USEDEFAULT;
    const DWORD exStyle = offscreen ? (WS_EX_TOOLWINDOW | WS_EX_NOACTIVATE) : 0;
    // Borderless (WS_POPUP): GetWindowRect == client rect, so anchored
    // placement over the embed panel is exact. A captioned/bordered window
    // (WS_OVERLAPPEDWINDOW) put the client area ~31px lower/right of the
    // requested rect -> cursor jumped on hover and clicks landed on the webview.
    const DWORD style = (offscreen ? WS_POPUP : WS_OVERLAPPEDWINDOW)
                        | WS_VISIBLE | WS_CLIPCHILDREN | WS_CLIPSIBLINGS;
    HWND hwnd = CreateWindowExA(exStyle, kFxGuiWndClass,
                                (name.empty() ? "VST FX" : name).c_str(),
                                style,
                                x, y, 800, 600,
                                nullptr, nullptr, GetModuleHandleA(nullptr), nullptr);
    if (!hwnd) { std::cerr << "[FxGui] create window failed" << std::endl; return 1; }
    SetWindowLongPtrA(hwnd, GWLP_USERDATA, userData);
    *hwndOut = hwnd;
    return 0;
}

// 3. Load the plugin + attach the editor (SEH-guarded — plugin GUI code may
//    crash). Slow — runs on the pump thread in server mode, after HTTP is up.
//    Returns 0 ok, 2 plugin load or editor attach failure.
static int fxGuiLoadAttach(Vst3Fx* fx, const std::string& path, const std::string& name, HWND hwnd, const std::string& presetB64) {
    // Fixed 44.1k/512 — GUI only, no audio processed here.
    if (!fx->load(path, 44100.0, 512)) { std::cerr << "[FxGui] plugin load failed: " << path << std::endl; return 2; }
    // Khôi phục preset GUI capture cuối (hide/show mất settings bug): áp
    // preset NGAY SAU load — trước attach editor để GUI hiện đúng cài đặt.
    if (!presetB64.empty() && !fx->applyPreset(presetB64))
        std::cerr << "[FxGui] preset apply FAILED (default kept): " << path << std::endl;
    else if (!presetB64.empty())
        std::cerr << "[FxGui] preset applied: " << path << std::endl;
    if (!fxGuiAttachSafe(fx, hwnd)) {
        std::cerr << "[FxGui] editor attach failed for " << path << std::endl;
        DestroyWindow(hwnd);
        return 2;
    }
    std::cout << "[FxGui] editor open: " << (name.empty() ? path : name) << std::endl;
    return 0;
}

// Combined setup (legacy open mode): parse + load + window + attach, in order.
static int fxGuiSetup(const std::string& jobPath, Vst3Fx* fx, HWND* hwndOut,
                      bool offscreen, LONG_PTR userData, std::string& shmOut) {
    std::string path, name, presetB64;
    int rc = fxGuiParseJob(jobPath, path, name, shmOut, presetB64);
    if (rc) return rc;
    if (!fx->load(path, 44100.0, 512)) { std::cerr << "[FxGui] plugin load failed: " << path << std::endl; return 2; }
    rc = fxGuiCreateWindow(name, offscreen, userData, hwndOut);
    if (rc) return rc;
    return fxGuiLoadAttach(fx, path, name, *hwndOut, presetB64);
}

int run_open_fx_gui(const std::string& jobPath) {
    // The editor window and its message pump live on this (STA) thread for the
    // whole GUI session — VST3 editors need the COM apartment + pump together.
    CoInitializeEx(nullptr, COINIT_APARTMENTTHREADED);
    struct ComGuard { ~ComGuard() { CoUninitialize(); } } comGuard;

    Vst3Fx fx;
    HWND hwnd = nullptr;
    FxGuiCtx ctx;
    std::string shm;
    const int rc = fxGuiSetup(jobPath, &fx, &hwnd, /*offscreen=*/false, (LONG_PTR)&ctx, shm);
    if (rc) return rc;
    fxGuiStartWatchdog(hwnd, fxGuiParentPid());
    fxGuiShmSetName(shm);
    fxGuiStartFeederThread(&fx, hwnd);

    // 5. Pump messages until the window is destroyed.
    MSG msg;
    while (!ctx.closing) {
        BOOL r = GetMessageA(&msg, nullptr, 0, 0);
        if (r <= 0) break;
        TranslateMessage(&msg);
        DispatchMessageA(&msg);
    }

    // 6. Detach + exit.
    fxGuiDetachSafe(&fx);
    std::cout << "[FxGui] editor closed" << std::endl;
    return 0;
}

// Embedded GUI server mode: same setup but off-screen, then hand the message
// pump + HTTP capture/input to FxGuiServer.cpp (see PLAN_MASTERBUS_VST_GUI_EMBED.md).
extern int fxGuiServerLoop(HWND hwnd, const std::function<void()>& onStarted);

int run_fx_gui_server(const std::string& jobPath) {
    CoInitializeEx(nullptr, COINIT_APARTMENTTHREADED);
    struct ComGuard { ~ComGuard() { CoUninitialize(); } } comGuard;

    Vst3Fx fx;
    HWND hwnd = nullptr;
    // Fast path: parse job + create the host window, then start the HTTP
    // server — SF_FXGUI_PORT prints immediately. The slow VST load + editor
    // attach is deferred to the pump thread (onStarted), so plugins.py stops
    // blocking on a multi-second plugin load (startup speed issue).
    std::string path, name, shm, presetB64;
    int rc = fxGuiParseJob(jobPath, path, name, shm, presetB64);
    if (rc) return rc;
    rc = fxGuiCreateWindow(name, /*offscreen=*/true, 0, &hwnd);
    if (rc) return rc;
    fxGuiStartWatchdog(hwnd, fxGuiParentPid());
    SetPropA(hwnd, "FXGUI_FX", &fx);  // for WM_FXGUI_RESIZE on the pump thread
    fxGuiShmSetName(shm);  // EARLY — frontend may /shm push before load finishes

    int loadRc = 0;
    const int srv = fxGuiServerLoop(hwnd, [&]() {
        loadRc = fxGuiLoadAttach(&fx, path, name, hwnd, presetB64);
        if (loadRc == 0) fxGuiStartFeederThread(&fx, hwnd);  // after load+attach only
        if (loadRc != 0) PostMessageA(hwnd, WM_CLOSE, 0, 0);  // exit; port already printed
    });
    RemovePropA(hwnd, "FXGUI_FX");
    fxGuiDetachSafe(&fx);
    std::cout << "[FxGuiServer] closed" << std::endl;
    return loadRc ? loadRc : srv;
}

#else  // !_WIN32

int run_open_fx_gui(const std::string& jobPath) {
    (void)jobPath;
    std::cerr << "[FxGui] GUI mode chỉ hỗ trợ Windows" << std::endl;
    return 1;
}

int run_fx_gui_server(const std::string& jobPath) {
    (void)jobPath;
    std::cerr << "[FxGui] GUI mode chỉ hỗ trợ Windows" << std::endl;
    return 1;
}

#endif  // _WIN32

// --- --scan: classify VST3 modules as instrument/effect ----------------------
namespace {

std::string jsonEscape(const std::string& s) {
    std::string o;
    o.reserve(s.size() + 8);
    for (char ch : s) {
        switch (ch) {
            case '"': o += "\\\""; break;
            case '\\': o += "\\\\"; break;
            case '\n': o += "\\n"; break;
            case '\r': o += "\\r"; break;
            case '\t': o += "\\t"; break;
            default: o += ch;
        }
    }
    return o;
}

#ifdef HAVE_VST3SDK
struct ScanEntry {
    std::string path;
    std::string name;
    bool isFx = false;
    bool isInstrument = false;
};

// C++ locals live here (no __try in this frame) — a crashing module loader
// is caught by safeScanModule's SEH and skipped, not fatal to the whole scan.
bool scanModuleInner(const std::string& path, std::vector<ScanEntry>& out) {
    using namespace VST3::Hosting;
#ifdef _WIN32
    setPluginSearchPath(path);  // deps cạnh plugin (0x7E fix) — scan cũng load module
#endif
    std::string err;
    Module::Ptr module = Module::create(path, err);
    if (!module) { std::cerr << "[Scan] skip (module load failed): " << path << " — " << err << std::endl; return false; }
    const PluginFactory& factory = module->getFactory();
    auto infos = factory.classInfos();
    int added = 0;
    for (const auto& ci : infos) {
        if (ci.category() != kVstAudioEffectClass) continue;
        const bool isInstr = ci.subCategoriesString().find("Instrument") != std::string::npos;
        ScanEntry e;
        e.path = path;
        e.name = ci.name();
        e.isFx = !isInstr;
        e.isInstrument = isInstr;
        out.push_back(std::move(e));
        ++added;
    }
    if (added == 0)
        std::cerr << "[Scan] " << path << ": no audio classes" << std::endl;
    return added > 0;
}

// Per-module scan context for the timeout worker thread.
struct ScanCtx {
    const std::string* path;
    std::vector<ScanEntry>* out;
    bool ok = false;
};
static DWORD WINAPI scanWorker(LPVOID p) {
    auto* ctx = static_cast<ScanCtx*>(p);
    __try { ctx->ok = scanModuleInner(*ctx->path, *ctx->out); }
    __except (EXCEPTION_EXECUTE_HANDLER) {
        std::cerr << "[Scan] skip (crashed module): " << *ctx->path << std::endl;
        ctx->ok = false;
    }
    return 0;
}

bool safeScanModule(const std::string& path, std::vector<ScanEntry>& out) {
#ifdef _WIN32
    // Run the VST3 loader on a dedicated thread so a HANGING module (e.g.
    // Ozone 11 Clarity.vst3 — Module::create blocks forever inside the SDK)
    // cannot stall the whole --scan. SEH only catches crashes, not hangs, so
    // a watchdog timeout + TerminateThread is required. Safe here: --scan is
    // a one-shot process; the worker only touches the VST3 SDK loader and the
    // process exits right after the scan completes.
    ScanCtx ctx{&path, &out};
    HANDLE hThread = CreateThread(nullptr, 0, scanWorker, &ctx, 0, nullptr);
    if (!hThread) { std::cerr << "[Scan] skip (thread fail): " << path << std::endl; return false; }
    const DWORD timeoutMs = 4000;  // 4s/module — module hợp lệ load 1-3s (đo
    // thực tế); module treo thì treo vĩnh viễn nên 4s đủ phân biệt. Trước đây
    // 12s × ~15 module Ozone 11 treo = 180s+ vượt timeout Python → mất CẢ dir
    // → MASTERING PANEL rỗng.
    DWORD wait = WaitForSingleObject(hThread, timeoutMs);
    if (wait == WAIT_TIMEOUT) {
        TerminateThread(hThread, 0);
        CloseHandle(hThread);
        std::cerr << "[Scan] skip (timeout 12s): " << path << std::endl;
        return false;
    }
    CloseHandle(hThread);
    return ctx.ok;
#else
    return scanModuleInner(path, out);
#endif
}
#endif // HAVE_VST3SDK

}  // namespace

int run_scan_dir(const std::string& dirPath) {
#ifdef _WIN32
    CoInitializeEx(nullptr, COINIT_MULTITHREADED);
    struct ComGuard { ~ComGuard() { CoUninitialize(); } } comGuard;
#endif
    std::vector<std::string> vst3Paths;   // folders or files ending .vst3
    std::vector<std::string> vst2Paths;   // .dll / .so (cannot introspect)

    namespace fs = std::filesystem;
    auto isExt = [](const std::string& p, const char* ext) {
        const size_t n = std::strlen(ext);
        return p.size() >= n && p.compare(p.size() - n, n, ext) == 0;
    };
    try {
        fs::recursive_directory_iterator it(dirPath), end;
        for (; it != end; ++it) {
            const std::string p = it->path().string();
            std::string low = p;
            for (auto& c : low) c = (char)std::tolower((unsigned char)c);
            if (it->is_directory()) {
                if (isExt(low, ".vst3")) {
                    vst3Paths.push_back(p);
                    it.disable_recursion_pending();  // bundle: don't walk Contents/
                }
            } else if (it->is_regular_file()) {
                if (isExt(low, ".vst3")) vst3Paths.push_back(p);
                else if (isExt(low, ".dll") || isExt(low, ".so")) vst2Paths.push_back(p);
            }
        }
    } catch (...) { std::cerr << "[Scan] walk failed: " << dirPath << std::endl; }

    std::cout << "{\"plugins\":[";
    bool first = true;
    auto emit = [&](const std::string& path, const std::string& name,
                    bool isFx, bool isInstr, const char* type) {
        if (!first) std::cout << ",";
        first = false;
        std::cout << "{\"path\":\"" << jsonEscape(path)
                  << "\",\"name\":\"" << jsonEscape(name)
                  << "\",\"is_fx\":" << (isFx ? "true" : "false")
                  << ",\"is_instrument\":" << (isInstr ? "true" : "false")
                  << ",\"type\":\"" << type << "\"}";
    };
#ifdef HAVE_VST3SDK
    for (const auto& p : vst3Paths) {
        std::vector<ScanEntry> entries;
        if (!safeScanModule(p, entries)) {
            // Module unreadable/crashed/hung (e.g. Ozone 11 Clarity): emit
            // entry UNKNOWN (is_fx=false, is_instrument=false) — trước đây
            // emit is_instrument=true làm VST FX lọt vào danh sách instrument
            // (nút Synth). Backend loại entry unknown khỏi vst_instruments:
            // không xác định được loại → không xếp bừa vào instrument.
            emit(p, p.substr(p.find_last_of("/\\") + 1), false, false, "VST3");
            continue;
        }
        for (const auto& e : entries) emit(e.path, e.name, e.isFx, e.isInstrument, "VST3");
    }
#else
    for (const auto& p : vst3Paths) {
        std::string name = p.substr(p.find_last_of("/\\") + 1);
        emit(p, name, false, true, "VST3");
    }
#endif
    for (const auto& p : vst2Paths) {
        std::string name = p.substr(p.find_last_of("/\\") + 1);
        emit(p, name, false, true, "VST2");
    }
    std::cout << "]}" << std::endl;
    return 0;
}