#!/usr/bin/env bun

/**
 * Example 2: YouTube synthetic dataset workflow
 * 1. Create project → 2. Generate synthetic data from YouTube → 3. Create snapshot → 4. Get recommendations → 5. Run finetuning
 */

const API_BASE_URL = process.env.API_BASE_URL || 'https://localhost';
const API_KEY = process.env.API_KEY;
const YOUTUBE_URL = 'https://www.youtube.com/watch?v=51y4KatMBFI';

const USER_INSTRUCTIONS =
	'Generate detailed question-answer pairs about the key concepts, main topics, important details, and technical information presented in the video. Focus on extracting specific facts, explanations, definitions, examples, and insights that are directly mentioned or demonstrated. Create questions that test understanding of the core subject matter and provide comprehensive answers based on the video content.';

// Disable TLS verification for local development with self-signed certs
process.env.NODE_TLS_REJECT_UNAUTHORIZED = '0';

if (!API_KEY) {
	console.error('Error: API_KEY environment variable is required');
	console.error('Please create a .env file based on .env.example');
	process.exit(1);
}

async function api(endpoint: string, options: any = {}): Promise<any> {
	const res = await fetch(`${API_BASE_URL}${endpoint}`, {
		...options,
		headers: { Authorization: `Bearer ${API_KEY}`, ...options.headers },
	});
	if (!res.ok) {
		const err: any = await res.json().catch(() => ({}));
		const errorMsg = typeof err.error === 'string' ? err.error : JSON.stringify(err);
		throw new Error(`${res.status}: ${errorMsg}`);
	}
	return res.json();
}

function sleep(ms: number) {
	return new Promise((r) => setTimeout(r, ms));
}

async function main() {
	console.log('\n=== YouTube Synthetic Workflow ===\n');

	// 1. Create project
	console.log('1. Creating project...');
	const { project_id } = await api('/api/v1/public/projects/create', {
		method: 'POST',
		headers: { 'Content-Type': 'application/json' },
		body: JSON.stringify({ name: 'YouTube AI Project', goal: 'Train from video content' }),
	});
	console.log(`   ✓ Project: ${project_id}\n`);

	// 2. Generate synthetic dataset
	console.log('2. Generating synthetic dataset from YouTube...');
	console.log(`   URL: ${YOUTUBE_URL}`);
	const formData = new FormData();
	formData.append('project_id', project_id);
	formData.append('name', 'YouTube Dataset');
	formData.append('youtube_urls[0]', YOUTUBE_URL);
	formData.append('pairs_to_generate', '50');
	formData.append('pair_type', 'qa');
	formData.append('user_instructions', USER_INSTRUCTIONS);

	const { dataset_id } = await api('/api/v1/public/datasets/create-synthetic', {
		method: 'POST',
		body: formData,
	});
	console.log(`   ✓ Dataset: ${dataset_id}`);

	// Wait for dataset (can take several minutes)
	console.log('   Waiting for generation (may take 5-10 minutes)...');
	let dataset;
	let checks = 0;
	do {
		await sleep(5000);
		dataset = await api(`/api/v1/public/datasets/${dataset_id}`);
		if (checks++ % 6 === 0) {
			console.log(`   Status: ${dataset.status}, ${dataset.datapoints_count} datapoints`);
		}
	} while (dataset.status === 'processing');
	console.log(`   ✓ Ready: ${dataset.datapoints_count} datapoints\n`);

	// 3. Create snapshot
	console.log('3. Creating snapshot...');
	const { snapshot_id } = await api('/api/v1/public/snapshots/create', {
		method: 'POST',
		headers: { 'Content-Type': 'application/json' },
		body: JSON.stringify({ dataset_id, split_percentage: 80 }),
	});
	console.log(`   ✓ Snapshot: ${snapshot_id}\n`);

	// 4. Generate recommendations
	console.log('4. Generating recommendations...');
	await api('/api/v1/public/recommendations/generate', {
		method: 'POST',
		headers: { 'Content-Type': 'application/json' },
		body: JSON.stringify({ snapshot_id, reasoning: false }),
	});

	let recs;
	do {
		await sleep(5000);
		recs = await api(`/api/v1/public/recommendations/${snapshot_id}?reasoning=false`);
	} while (recs.status === 'processing');

	console.log(`   ✓ Recommended models:`);
	const recommendedCount = recs.recommended_models.filter((m: any) => m.recommended).length;
	console.log(`   Total models: ${recs.recommended_models.length}, Recommended: ${recommendedCount}`);
	recs.recommended_models.forEach((m: any) => {
		if (m.recommended) console.log(`     - ${m.baseModelId}`);
	});
	console.log();

	// 5. Create finetuning job
	console.log('5. Creating finetuning job...');
	const experiments = recs.recommended_models
		.filter((m: any) => m.recommended)
		.map((m: any) => ({
			base_model_id: m.baseModelId,
			batch_size: m.lora_hyperparameters.batchSize,
			learning_rate_multiplier: m.lora_hyperparameters.learningRateMultiplier,
			n_epochs: m.lora_hyperparameters.nEpochs,
			lora: true,
		}));

	if (experiments.length === 0) {
		console.error('\n✗ Error: No recommended models found. Cannot create finetuning job.');
		process.exit(1);
	}

	const { job_id } = await api('/api/v1/public/finetuning/create', {
		method: 'POST',
		headers: { 'Content-Type': 'application/json' },
		body: JSON.stringify({ snapshot_id, name: 'YouTube Model', reasoning: false, experiments }),
	});
	console.log(`   ✓ Job: ${job_id}\n`);

	// 6. Monitor (5 minutes max)
	console.log('6. Monitoring job...');
	for (let i = 0; i < 30; i++) {
		await sleep(10000);
		const job = await api(`/api/v1/public/finetuning/${job_id}`);
		console.log(`   Status: ${job.status}`);
		job.experiments.forEach((e: any) => {
			console.log(`     - Exp #${e.experiment_number}: ${e.status} ${e.model_id || ''}`);
		});
		if (job.status !== 'processing') break;
	}

	console.log('\n✓ Done!\n');
}

main().catch((err) => {
	console.error('\n✗ Error:', err.message);
	process.exit(1);
});