import { Database } from "jsr:@db/sqlite";
import {
  assert,
  assertEquals,
} from "https://deno.land/std@0.224.0/testing/asserts.ts";

/**
 * Proof of Concept: Embedded Vector Database (Gen 2)
 *
 * Demonstrates the concept of fuzzy semantic retrieval using an actual
 * SQLite database. While we are not loading a C extension like `sqlite-vec`
 * directly here to keep the PoC universally executable without native build
 * dependencies, we simulate it via SQL and User Defined Functions (UDF)
 * provided by Deno's `jsr:@db/sqlite`.
 */

// A simple mock for cosine similarity of 1D arrays
function cosineSimilarity(vecA: number[], vecB: number[]): number {
  let dotProduct = 0;
  let normA = 0;
  let normB = 0;
  for (let i = 0; i < vecA.length; i++) {
    dotProduct += vecA[i] * vecB[i];
    normA += vecA[i] ** 2;
    normB += vecB[i] ** 2;
  }
  if (normA === 0 || normB === 0) return 0;
  return dotProduct / (Math.sqrt(normA) * Math.sqrt(normB));
}

if (import.meta.main) {
  console.log("Running Embedded Vector Database PoC (Gen 2) tests...");

  try {
    const db = new Database(":memory:");

    // Create a user-defined function in SQLite to perform vector similarity!
    db.function("vec_distance", (aStr: string, bStr: string) => {
      const vecA = JSON.parse(aStr) as number[];
      const vecB = JSON.parse(bStr) as number[];
      return cosineSimilarity(vecA, vecB);
    });

    db.exec(`
      CREATE TABLE documents (
        id TEXT PRIMARY KEY,
        text TEXT,
        vector TEXT
      );
    `);

    const insert = db.prepare(
      "INSERT INTO documents (id, text, vector) VALUES (?, ?, ?)",
    );

    insert.run(
      "docs-1",
      "How to run the server",
      JSON.stringify([0.8, 0.1, 0.1, 0.0]),
    );
    insert.run(
      "docs-2",
      "Database connection logic",
      JSON.stringify([0.1, 0.9, 0.2, 0.1]),
    );
    insert.run(
      "telemetry-1",
      "Server latency spikes",
      JSON.stringify([0.2, 0.1, 0.9, 0.3]),
    );
    insert.finalize();

    // Query representing "I have a slow server issue"
    const queryVectorStr = JSON.stringify([0.3, 0.0, 0.9, 0.2]);

    console.log("Querying Vector DB...");

    const results = db.prepare(`
      SELECT id, text, vec_distance(vector, ?) as score
      FROM documents
      ORDER BY score DESC
    `).all(queryVectorStr) as { id: string; text: string; score: number }[];

    console.log(
      "Top result:",
      results[0].text,
      `(Score: ${results[0].score.toFixed(2)})`,
    );

    assert(
      results[0].score > 0.8,
      "The telemetry doc should be the highest match",
    );
    assertEquals(results[0].id, "telemetry-1");

    console.log(
      "✅ Embedded Vector DB PoC (Gen 2) successful: Real SQLite fuzzy semantic match via UDF.",
    );

    db.close();
  } catch (err) {
    console.error("❌ Embedded Vector DB PoC (Gen 2) failed:", err);
    Deno.exit(1);
  }
}