/**
 * earth-now.co
 * Author:  Simon-Pierre Boucher
 * Contact: contact@spboucher.ai
 * File:    apps/ingest/test/runners.test.ts
 * Purpose: End-to-end run functions (fetch → archive → parse → summary) with fixture-backed fakes, no network
 */

import { existsSync, mkdtempSync, readdirSync, readFileSync } from "node:fs";
import { tmpdir } from "node:os";
import { join } from "node:path";
import { afterEach, beforeEach, describe, expect, it } from "vitest";
import type { FetchLike } from "../src/fetch-like";
import { KNOWN_SOURCE_IDS, runSource } from "../src/runners";

const fixture = (name: string): string =>
  readFileSync(new URL(`./fixtures/${name}`, import.meta.url), "utf8");

let previousRawDir: string | undefined;

beforeEach(() => {
  previousRawDir = process.env.RAW_ARCHIVE_DIR;
  process.env.RAW_ARCHIVE_DIR = mkdtempSync(join(tmpdir(), "earth-now-runners-"));
});

afterEach(() => {
  if (previousRawDir === undefined) delete process.env.RAW_ARCHIVE_DIR;
  else process.env.RAW_ARCHIVE_DIR = previousRawDir;
});

describe("runSource", () => {
  it("knows exactly the declared sources", () => {
    expect([...KNOWN_SOURCE_IDS]).toEqual(["usgs_fdsn", "noaa_gml_mlo"]);
  });

  it("rejects unknown source ids listing the known ones", async () => {
    await expect(runSource("nope")).rejects.toThrow(/Unknown source 'nope'.*usgs_fdsn/);
  });

  it("usgs_fdsn: fetches, archives the raw payload and summarizes one observation", async () => {
    const body = fixture("usgs_count_sample.json");
    const fetchImpl: FetchLike = async () => ({ ok: true, status: 200, text: async () => body });

    const summary = await runSource("usgs_fdsn", fetchImpl);
    expect(summary.sourceId).toBe("usgs_fdsn");
    expect(summary.observationCount).toBe(1);
    expect(summary.first?.value).toBe(132);
    expect(existsSync(summary.rawPath)).toBe(true);
    expect(readFileSync(summary.rawPath, "utf8")).toBe(body);
    expect(existsSync(`${summary.rawPath}.sha256`)).toBe(true);
  });

  it("noaa_gml_mlo: fetches, archives and summarizes the monthly series", async () => {
    const body = fixture("co2_mm_mlo_sample.txt");
    const fetchImpl: FetchLike = async () => ({ ok: true, status: 200, text: async () => body });

    const summary = await runSource("noaa_gml_mlo", fetchImpl);
    expect(summary.observationCount).toBe(15);
    expect(summary.first?.time).toBe("2024-01-15T00:00:00.000Z");
    expect(summary.last?.value).toBe(429.64);
    expect(readFileSync(summary.rawPath, "utf8")).toBe(body);
  });

  it("noaa_gml_mlo: still archives the raw payload even when parsing fails (truncated file)", async () => {
    const body = fixture("co2_mm_mlo_truncated.txt");
    const fetchImpl: FetchLike = async () => ({ ok: true, status: 200, text: async () => body });

    await expect(runSource("noaa_gml_mlo", fetchImpl)).rejects.toThrow(/truncated/);
    // The raw bytes were archived before parsing — nothing ingested is ever lost.
    const archivedDir = join(process.env.RAW_ARCHIVE_DIR ?? "", "noaa_gml_mlo");
    expect(existsSync(archivedDir)).toBe(true);
    expect(readdirSync(archivedDir).some((f) => f.endsWith(".raw"))).toBe(true);
  });
});