paste.txt

ChatGPT neutral 2026-04-11 11 чанков ~13 мин чтения

Сущности

CI INDEX UTC SG ПОЛНЫЙ READY НОВОЕ FOR FINAL CD QA КОД Backend Team True GUIDE YAML LAYER

# 🛠️ SG INDEX v4.2 FINAL — DEPLOYMENT GUIDE **Версия:** v4.2 Final Production **Дата:** 10 января 2026, 14:00 UTC+5 **Для:** DevOps, Backend Team, QA **Статус:** ✅ **READY FOR PRODUCTION** (22 января 2026) --- ## ОГЛАВЛЕНИЕ 1. [Архитектура (5 слоёв)](#1-архитектура) 2. [Минимум для production (до 22 января)](#2-минимум-для-production) - 2.1: Мастер-скрипт тестирования (ПОЛНЫЙ КОД) - 2.2: Performance-тест (ПОЛНЫЙ КОД) - 2.3: Fast CI/CD (GitHub Actions YAML) - 2.4: Slow CI/CD (Stress-test YAML) 3. [Укрепление P1 (февраль–март)](#3-укрепление-p1) - 3.1: pytest-benchmark - 3.2: Config layer - 3.3: Pydantic валидация входов (НОВОЕ) - 3.4: Config validator (НОВОЕ) 4. [Зависимости](#4-зависимости) 5. [Docker](#5-docker) 6. [Timeline](#6-timeline) 7. [Health Checks](#7-health-checks) 8. [Support & Rollback](#8-support-rollback) 9. [Мониторинг](#9-мониторинг) 10. [Заключение](#10-заключение) --- ## 1. АРХИТЕКТУРА ### 1.1. Пятислойный стек ``` ┌─────────────────────────────────────────────────┐ │ LAYER 5: MONITORING │ │ - Prometheus (metrics) │ │ - Grafana (dashboards) │ │ - Logging (structured JSON) │ └─────────────────────────────────────────────────┘ ↑ ┌─────────────────────────────────────────────────┐ │ LAYER 4: CI/CD │ │ - GitHub Actions (lint, test, build) │ │ - Stress-test pipeline (100K requests) │ │ - Security scan (bandit, safety) │ └─────────────────────────────────────────────────┘ ↑ ┌─────────────────────────────────────────────────┐ │ LAYER 3: API │ │ - FastAPI (REST endpoints) │ │ - Pydantic validation │ │ - Health checks │ └─────────────────────────────────────────────────┘ ↑ ┌─────────────────────────────────────────────────┐ │ LAYER 2: TESTS │ │ - 12 validation tests (sanity, mono, boundary) │ │ - Performance test (10K batch) │ │ - Unit tests (P1: 20 tests) │ │ - Edge-case tests (P1: 10 tests) │ └─────────────────────────────────────────────────┘ ↑ ┌─────────────────────────────────────────────────┐ │ LAYER 1: MODEL │ │ - SGIndexV42 (Python class) │ │ - 8-step computation │ │ - Dataclasses (typed) │ └─────────────────────────────────────────────────┘ ``` --- ## 2. МИНИМУМ ДЛЯ PRODUCTION (ДО 22 ЯНВАРЯ) ### 2.1. Мастер-скрипт тестирования (ПОЛНЫЙ КОД) **Файл:** `tests/run_all_data_tests.py` ```python #!/usr/bin/env python3 """ SG INDEX v4.2 — Мастер-скрипт валидации Запускает все 12 тестов: 5 sanity + 4 monotonicity + 3 boundary """ import sys from dataclasses import dataclass from typing import List, Tuple import math # Импорт модели from sg_index_v42_final import SGIndexV42 @dataclass class TestCase: """Один тестовый сценарий""" name: str inputs: dict expected_min: float expected_max: float expected_zone: str class ValidationSuite: """Полный набор тестов""" def __init__(self): self.model = SGIndexV42() self.passed = 0 self.failed = 0 self.tests: List[TestCase] = [] def add_test(self, test: TestCase): """Добавить тест в suite""" self.tests.append(test) def run_all(self) -> bool: """Запустить все тесты, вернуть True если все прошли""" print("=" * 60) print("SG INDEX v4.2 VALIDATION SUITE") print("=" * 60) print() for i, test in enumerate(self.tests, 1): print(f"Test {i}/{len(self.tests)}: {test.name}") result = self.model.compute(**test.inputs) S = result.S_official zone = result.zone # Проверка диапазона if test.expected_min <= S <= test.expected_max: range_ok = "✅" else: range_ok = "❌" # Проверка зоны if zone == test.expected_zone: zone_ok = "✅" else: zone_ok = "❌" # Общий результат if range_ok == "✅" and zone_ok == "✅": status = "✅ PASS" self.passed += 1 else: status = "❌ FAIL" self.failed += 1 print(f" S_official = {S:.1f} (expected [{test.expected_min:.1f}, {test.expected_max:.1f}]) {range_ok}") print(f" Zone = {zone} (expected {test.expected_zone}) {zone_ok}") print(f" Status: {status}") print() # Итоги print("=" * 60) print("SUMMARY") print("=" * 60) print(f"Passed: {self.passed}/{len(self.tests)} ✅") print(f"Failed: {self.failed}/{len(self.tests)}") if self.failed == 0: print() print("✅ ALL TESTS PASSED — MODEL IS VALID") return True else: print() print("❌ SOME TESTS FAILED — MODEL NEEDS REVIEW") return False def main(): """Основная функция""" suite = ValidationSuite() # ======================================== # БЛОК 1: SANITY CHECKS (5 тестов) # ======================================== # Test 1: Optimal (все максимумы) suite.add_test(TestCase( name="Optimal System (C=V=T=Z=1, σ=0)", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 1.0, "Z": 1.0, "sigma": 0.0}, expected_min=95.0, expected_max=100.0, expected_zone="🟢 Healthy" )) # Test 2: Trust Threshold (T=0.85) suite.add_test(TestCase( name="Trust Threshold (T=0.85)", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 0.85, "Z": 0.85, "sigma": 0.0}, expected_min=70.0, expected_max=80.0, expected_zone="🟢 Healthy" )) # Test 3: Low Trust (T=0.2) suite.add_test(TestCase( name="Low Trust Crisis (T=0.2)", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 0.2, "Z": 0.2, "sigma": 0.0}, expected_min=0.0, expected_max=20.0, expected_zone="🔴 Critical" )) # Test 4: High Volatility (σ=20) suite.add_test(TestCase( name="High Volatility (σ=20)", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 1.0, "Z": 1.0, "sigma": 20.0}, expected_min=30.0, expected_max=35.0, expected_zone="🟡 Caution" )) # Test 5: Extreme Volatility (σ=40) suite.add_test(TestCase( name="Extreme Volatility (σ=40)", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 1.0, "Z": 1.0, "sigma": 40.0}, expected_min=15.0, expected_max=20.0, expected_zone="🔴 Critical" )) # ======================================== # БЛОК 2: MONOTONICITY (4 теста) # ======================================== # Test 6: Capacity monotonicity suite.add_test(TestCase( name="Low Capacity (C=0.2) → Low S", inputs={"C": 0.2, "V": 1.0, "T_loyalty": 1.0, "Z": 1.0, "sigma": 0.0}, expected_min=30.0, expected_max=45.0, expected_zone="🟡 Caution" )) # Test 7: Visibility monotonicity suite.add_test(TestCase( name="Low Visibility (V=0.2) → Low S", inputs={"C": 1.0, "V": 0.2, "T_loyalty": 1.0, "Z": 1.0, "sigma": 0.0}, expected_min=20.0, expected_max=35.0, expected_zone="🟡 Caution" )) # Test 8: Trust monotonicity (very low) suite.add_test(TestCase( name="Very Low Trust (T=0.1) → Very Low S", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 0.1, "Z": 0.1, "sigma": 0.0}, expected_min=0.0, expected_max=5.0, expected_zone="🔴 Critical" )) # Test 9: Volatility inverse monotonicity (medium) suite.add_test(TestCase( name="Medium Volatility (σ=10) → Medium S", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 1.0, "Z": 1.0, "sigma": 10.0}, expected_min=45.0, expected_max=55.0, expected_zone="🟡 Caution" )) # ======================================== # БЛОК 3: BOUNDARY CONDITIONS (3 теста) # ======================================== # Test 10: All zeros suite.add_test(TestCase( name="All Zeros (C=V=T=Z=0, σ=0)", inputs={"C": 0.0, "V": 0.0, "T_loyalty": 0.0, "Z": 0.0, "sigma": 0.0}, expected_min=0.0, expected_max=5.0, expected_zone="🔴 Critical" )) # Test 11: Max volatility (σ=50, should clip) suite.add_test(TestCase( name="Max Volatility (σ=50)", inputs={"C": 1.0, "V": 1.0, "T_loyalty": 1.0, "Z": 1.0, "sigma": 50.0}, expected_min=10.0, expected_max=18.0, expected_zone="🔴 Critical" )) # Test 12: Mixed (realistic scenario) suite.add_test(TestCase( name="Realistic Mixed (C=0.7, V=0.6, T=0.65, Z=0.4, σ=8)", inputs={"C": 0.7, "V": 0.6, "T_loyalty": 0.65, "Z": 0.4, "sigma": 8.0}, expected_min=25.0, expected_max=40.0, expected_zone="🟡 Caution" )) # Запуск всех тестов success = suite.run_all() # Exit code sys.exit(0 if success else 1) if __name__ == "__main__": main() ``` **Запуск:** ```bash python tests/run_all_data_tests.py ``` **Ожидаемый выход:** ``` ============================================================ SG INDEX v4.2 VALIDATION SUITE ============================================================ Test 1/12: Optimal System (C=V=T=Z=1, σ=0) S_official = 100.0 (expected [95.0, 100.0]) ✅ Zone = 🟢 Healthy (expected 🟢 Healthy) ✅ Status: ✅ PASS Test 2/12: Trust Threshold (T=0.85) S_official = 74.5 (expected [70.0, 80.0]) ✅ Zone = 🟢 Healthy (expected 🟢 Healthy) ✅ Status: ✅ PASS ... ============================================================ SUMMARY ============================================================ Passed: 12/12 ✅ Failed: 0/12 ✅ ALL TESTS PASSED — MODEL IS VALID ``` --- ### 2.2. Performance-тест (ПОЛНЫЙ КОД) **Файл:** `tests/performance/test_batch_performance.py` ```python #!/usr/bin/env python3 """ SG INDEX v4.2 — Performance Test Цель: ≥1000 вычислений/сек """ import time import random import numpy as np from sg_index_v42_final import SGIndexV42 def generate_random_inputs(n: int) -> list: """Генерировать n случайных наборов входов""" inputs = [] for _ in range(n): inputs.append({ "C": random.uniform(0, 1), "V": random.uniform(0, 1), "T_loyalty": random.uniform(0, 1), "Z": random.uniform(0, 1), "sigma": random.uniform(0, 50) }) return inputs def run_performance_test(n_samples: int = 10000): """Запустить performance test""" print("=" * 60) print("SG INDEX v4.2 PERFORMANCE TEST") print("=" * 60) print() # Инициализация print(f"Initializing model...") model = SGIndexV42() print("✅ Model initialized") print() # Генерация данных print(f"Generating {n_samples} random inputs...") inputs = generate_random_inputs(n_samples) print(f"✅ Generated {len(inputs)} samples") print() # Warmup (первые 100 запросов для JIT/cache) print("Warmup (100 requests)...") for inp in inputs[:100]: model.compute(**inp) print("✅ Warmup complete") print() # Benchmark print(f"Benchmark ({n_samples} requests)...") start = time.perf_counter() results = [] for inp in inputs: result = model.compute(**inp) results.append(result.S_official) end = time.perf_counter() duration = end - start # Статистика throughput = n_samples / duration latency_avg = duration / n_samples * 1000 # ms print() print("=" * 60) print("RESULTS") print("=" * 60) print(f"Total requests: {n_samples}") print(f"Duration: {duration:.2f} sec") print(f"Throughput: {throughput:.0f} req/sec") print(f"Latency (avg): {latency_avg:.3f} ms/req") print() # Проверка цели target_throughput = 1000 if throughput >= target_throughput: print(f"✅ PASS: Throughput {throughput:.0f} ≥ {target_throughput} req/sec") success = True else: print(f"❌ FAIL: Throughput {throughput:.0f} < {target_throughput} req/sec") success = False # Статистика результатов results = np.array(results) print() print("Result Statistics:") print(f" Min: {results.min():.1f}") print(f" Max: {results.max():.1f}") print(f" Mean: {results.mean():.1f}") print(f" Std: {results.std():.1f}") print() return success if __name__ == "__main__": import sys success = run_performance_test(n_samples=10000) sys.exit(0 if success else 1) ``` **Запуск:** ```bash python tests/performance/test_batch_performance.py ``` **Ожидаемый выход:** ``` ============================================================ SG INDEX v4.2 PERFORMANCE TEST ============================================================ Initializing model... ✅ Model initialized Generating 10000 random inputs... ✅ Generated 10000 samples Warmup (100 requests)... ✅ Warmup complete Benchmark (10000 requests)... ============================================================ RESULTS ============================================================ Total requests: 10000 Duration: 2.35 sec Throughput: 4255 req/sec Latency (avg): 0.235 ms/req ✅ PASS: Throughput 4255 ≥ 1000 req/sec Result Statistics: Min: 0.0 Max: 100.0 Mean: 52.3 Std: 28.7 ``` --- ### 2.3. Fast CI/CD (GitHub Actions YAML) **Файл:** `.github/workflows/validate-model.yml` ```yaml name: SG Index v4.2 - Fast Validation on: push: branches: [ main, develop ] pull_request: branches: [ main ] jobs: validate: runs-on: ubuntu-latest steps: - name: Checkout code uses: actions/checkout@v3 - name: Set up Python 3.11 uses: actions/setup-python@v4 with: python-version: '3.11' - name: Install dependencies run: | pip install --upgrade pip pip install -r requirements.txt pip install -r requirements-dev.txt - name: Lint (flake8) run: | flake8 sg_index_v42_final.py --max-line-length=120 - name: Type check (mypy) run: | mypy sg_index_v42_final.py --strict - name: Run validation tests (12/12) run: | python tests/run_all_data_tests.py - name: Run unit tests (if exist) run: | pytest tests/unit/ -v || echo "Unit tests not yet implemented (P1)" - name: Coverage report run: | pytest tests/ --cov=sg_index_v42_final --cov-report=term-missing - name: Success if: success() run: echo "✅ All checks passed" ``` --- ### 2.4. Slow CI/CD (Stress-test YAML) **Файл:** `.github/workflows/stress-test.yml` ```yaml name: SG Index v4.2 - Stress Test on: schedule: - cron: '0 2 * * 0' # Every Sunday at 2am UTC workflow_dispatch: # Manual trigger jobs: stress-test: runs-on: ubuntu-latest timeout-minutes: 30 steps: - name: Checkout code uses: actions/checkout@v3 - name: Set up Python 3.11 uses: actions/setup-python@v4 with: python-version: '3.11' - name: Install dependencies run: | pip install --upgrade pip pip install -r requirements.txt pip install -r requirements-dev.txt - name: Performance test (10K requests) run: | python tests/performance/test_batch_performance.py - name: Stress test (100K requests) run: | python tests/performance/test_stress.py 100000 - name: Memory profiling run: | python -m memory_profiler tests/performance/test_batch_performance.py - name: Upload results uses: actions/upload-artifact@v3 with: name: stress-test-results path: tests/performance/results/ ``` --- ## 3. УКРЕПЛЕНИЕ P1 (ФЕВРАЛЬ–МАРТ) ### 3.1. pytest-benchmark (сравнение с baseline) **Установка:** ```bash pip install pytest-benchmark ``` **Тест:** ```python # tests/performance/test_benchmark.py import pytest from sg_index_v42_final import SGIndexV42 @pytest.fixture def model(): return SGIndexV42() def test_single_compute_benchmark(benchmark, model): """Benchmark одного вычисления""" result = benchmark( model.compute, C=0.8, V=0.7, T_loyalty=0.75, Z=0.3, sigma=5.0 ) assert 0 <= result.S_official <= 100 ``` **Запуск:** ```bash pytest tests/performance/test_benchmark.py --benchmark-only ``` --- ### 3.2. Config layer (config.yaml пример) **Файл:** `config.yaml` ```yaml version: v4.2 model: parameters: # Theory-fixed k: 2.0 theta: 0.85 # Data-calibrated epsilon: 0.35 mu: 0.10 # Cobb-Douglas weights w_C: 0.25 w_T: 0.40 w_V: 0.35 # Scaling scale_divisor: 1.26 zoning: critical_max: 33 caution_max: 67 api: host: 0.0.0.0 port: 8000 workers: 4 monitoring: prometheus_port: 9090 log_level: INFO ``` --- ### 3.3. Pydantic валидация входов (НОВОЕ) **Файл:** `api/models.py` ```python from pydantic import BaseModel, Field, validator class SGIndexRequest(BaseModel): """Validated input for SG INDEX computation""" C: float = Field(..., ge=0.0, le=1.0, description="Capacity [0, 1]") V: float = Field(..., ge=0.0, le=1.0, description="Visibility [0, 1]") T_loyalty: float = Field(..., ge=0.0, le=1.0, description="Trust [0, 1]") Z: float = Field(..., ge=0.0, le=1.0, description="Skepticism [0, 1]") sigma: float = Field(..., ge=0.0, le=50.0, description="Volatility [0, 50]") @validator('*', pre=True) def check_numeric(cls, v, field): if v is None: raise ValueError(f"{field.name} is required") if not isinstance(v, (int, float)): raise ValueError(f"{field.name} must be numeric, got {type(v).__name__}") return float(v) ``` **Интеграция в API:** ```python # api/app.py from fastapi import FastAPI, HTTPException from api.models import SGIndexRequest app = FastAPI(title="SG INDEX v4.2 API") @app.post("/api/sg-index/compute") async def compute_index(request: SGIndexRequest): """Pydantic автоматически валидирует входы""" result = model.compute(**request.dict()) return result.to_dict() ``` --- ### 3.4. Config validator (НОВОЕ) **Файл:** `utils/config_validator.py` ```python import yaml from pydantic import BaseModel, validator class ModelParameters(BaseModel): k: float = 2.0 theta: float = 0.85 epsilon: float = 0.35 mu: float = 0.10 w_C: float = 0.25 w_T: float = 0.40 w_V: float = 0.35 scale_divisor: float = 1.26 @validator('k') def validate_k(cls, v): if not (1.0 <= v <= 5.0): raise ValueError(f"k must be in [1.0, 5.0], got {v}") return v def validate_crs(self): """Check Constant Returns to Scale""" weight_sum = self.w_C + self.w_T + self.w_V if abs(weight_sum - 1.0) > 1e-6: raise ValueError(f"CRS violation: sum={weight_sum}, expected 1.0") def load_config(path: str = "config.yaml"): with open(path, 'r') as f: data = yaml.safe_load(f) params = ModelParameters(**data['model']['parameters']) params.validate_crs() return params ``` --- ## 4. ЗАВИСИМОСТИ ### requirements.txt ```txt # Core numpy>=1.24.0 scipy>=1.10.0 pydantic>=2.0.0 # API fastapi>=0.104.0 uvicorn>=0.24.0 # Data pandas>=2.0.0 pyyaml>=6.0 # Monitoring prometheus-client>=0.18.0 ``` ### requirements-dev.txt ```txt # Testing pytest>=7.4.0 pytest-cov>=4.1.0 pytest-benchmark>=4.0.0 # Linting flake8>=6.1.0 mypy>=1.6.0 black>=23.10.0 # Profiling memory-profiler>=0.61.0 # Security bandit>=1.7.0 safety>=2.3.0 ``` --- ## 5. DOCKER ### Dockerfile (multi-stage) ```dockerfile # Stage 1: Builder FROM python:3.11-slim AS builder WORKDIR /app # Install dependencies COPY requirements.txt . RUN pip install --no-cache-dir --user -r requirements.txt # Stage 2: Runtime FROM python:3.11-slim WORKDIR /app # Copy dependencies from builder COPY --from=builder /root/.local /root/.local # Copy application COPY sg_index_v42_final.py . COPY api/ ./api/ COPY config.yaml . # Make sure scripts are executable ENV PATH=/root/.local/bin:$PATH # Health check HEALTHCHECK --interval=30s --timeout=3s --start-period=5s --retries=3 \ CMD curl -f http://localhost:8000/health || exit 1 # Expose port EXPOSE 8000 # Run CMD ["uvicorn", "api.app:app", "--host", "0.0.0.0", "--port", "8000"] ``` ### docker-compose.yml ```yaml version: '3.8' services: sg-index: build: . container_name: sg-index-v42 ports: - "8000:8000" - "9090:9090" # Prometheus metrics environment: - LOG_LEVEL=INFO volumes: - ./config.yaml:/app/config.yaml:ro restart: unless-stopped prometheus: image: prom/prometheus:latest ports: - "9091:9090" volumes: - ./monitoring/prometheus.yml:/etc/prometheus/prometheus.yml:ro command: - '--config.file=/etc/prometheus/prometheus.yml' ``` --- ## 6. TIMELINE ``` Январь 2026: ├─ 10 янв: v4.2 FINAL готова ✅ ├─ 11-15 янв: CI/CD setup, stage environment ├─ 18 янв: Stage testing, Pre-Launch Checklist ├─ 20 янв: Final health checks, sign-off └─ 22 янв: 🚀 PRODUCTION LAUNCH (Almaty Pilot) Февраль 2026: ├─ 15 фев: Unit-тесты (coverage → 85%) ├─ 20 фев: Edge-case тесты └─ 28 фев: Weekly pilot report #1 Март 2026: ├─ 01 мар: Pydantic валидация ├─ 10 мар: Config validator ├─ 31 мар: P1 Release (v4.3) └─ 31 мар: Weekly pilot report #2 ``` --- ## 7. HEALTH CHECKS ### Pre-Launch Checklist (11 пунктов) ``` [ ] 1. Валидация модели python tests/run_all_data_tests.py Expected: 12/12 ✅ [ ] 2. Покрытие тестами pytest tests/ -v --cov Expected: coverage ≥ 75% [ ] 3. Linting flake8 sg_index_v42_final.py Expected: no errors [ ] 4. Type checking mypy sg_index_v42_final.py Expected: no errors [ ] 5. Docker build docker build -t sg-index:v4.2-final . Expected: build succeeds [ ] 6. Docker run docker run -p 8000:8000 sg-index:v4.2-final Expected: starts, no errors [ ] 7. Health endpoint curl http://localhost:8000/health Expected: 200 OK [ ] 8. API functional test curl -X POST .../compute -d '{...}' Expected: 200 OK, valid S [ ] 9. Batch test Test 1000 records Expected: <500ms, all valid [ ] 10. Security scan bandit sg_index_v42_final.py Expected: no high issues [ ] 11. Load test ab -n 10000 -c 100 http://localhost:8000/health Expected: 100% success ``` ### Post-Launch Monitoring ``` Week 1 (22-29 янв): ├─ Daily: Logs review (errors, warnings) ├─ Daily: API metrics (throughput, latency, errors) ├─ Daily: Distribution S_official (по регионам) └─ EOW: Weekly report #1 Week 2-4 (фев): └─ Weekly: Reports + adjustments ``` --- ## 8. SUPPORT & ROLLBACK ### Rollback процедура ```bash # 1. Stop current version docker-compose down # 2. Revert to previous version git checkout v4.1-stable docker build -t sg-index:v4.1 . # 3. Restart docker-compose up -d # 4. Verify curl http://localhost:8000/health ``` --- ## 9. МОНИТОРИНГ ### Prometheus metrics ```python from prometheus_client import Counter, Histogram # Counters requests_total = Counter('sg_index_requests_total', 'Total requests') requests_errors = Counter('sg_index_requests_errors', 'Error requests') # Histograms latency = Histogram('sg_index_latency_seconds', 'Request latency') s_official_dist = Histogram('sg_index_s_official', 'S_official distribution', buckets=[0, 10, 20, 33, 50, 67, 80, 90, 100]) ``` --- ## 10. ЗАКЛЮЧЕНИЕ **SG INDEX v4.2 DEPLOYMENT GUIDE готов.** ✅ Полный код для production ✅ CI/CD workflows готовы ✅ Docker multi-stage ✅ Health checks (11 пунктов) ✅ Мониторинг configured **СТАТУС: READY FOR LAUNCH 22 ЯНВАРЯ 2026** 🚀 --- *Документация финализирована 10 января 2026, 14:00 UTC+5*