build-cran-binaries/edge/rpkgs-router.test.ts

/**
 * Routing matrix for `edge/rpkgs-router.ts`.
 *
 * The script is exercised through the SDK's local server rather than by
 * importing its internals, so what is tested is the artifact that gets
 * deployed. Requests that the script passes through are proxied to the real
 * origin, which keeps the "no redirect" cases honest: they assert that the
 * client reached the flat slot, not merely that no `Location` was set.
 *
 * Run with `just edge-test`.
 */
import { assertEquals } from 'jsr:@std/assert@1';

const SCRIPT = new URL('./rpkgs-router.ts', import.meta.url).pathname;
const BASE = 'http://127.0.0.1:8080';
const UNION_SLOTS = 'amd64/alpine324';

const UA_R45_MUSL = 'R (4.5.3 x86_64-pc-linux-musl x86_64 linux-musl)';
const UA_R46_MUSL = 'R (4.6.0 x86_64-pc-linux-musl x86_64 linux-musl)';
const UA_R45_ALPINE = 'R/4.5.3 R (4.5.3 x86_64-pc-linux-musl x86_64 linux-musl) Alpine Linux 3.24';
const UA_R45_DARWIN = 'R (4.5.1 aarch64-apple-darwin20 aarch64 darwin20)';
const UA_CURL = 'curl/8.0.1';

const SLOT = '/amd64/alpine324/latest/src/contrib';
const OTHER_SLOT = '/amd64/noble/latest/src/contrib';

interface Probe {
  status: number;
  location: string | null;
  cacheControl: string | null;
}

async function probe(path: string, userAgent: string): Promise<Probe> {
  const res = await fetch(BASE + path, {
    headers: { 'User-Agent': userAgent },
    redirect: 'manual',
  });
  await res.body?.cancel();
  return {
    status: res.status,
    location: res.headers.get('location'),
    cacheControl: res.headers.get('cache-control'),
  };
}

/** Kill tolerantly: the child has already exited if the script failed to load. */
async function stopServer(child: Deno.ChildProcess): Promise<void> {
  try {
    child.kill();
  } catch {
    // already gone
  }
  await child.status;
}

async function startServer(): Promise<Deno.ChildProcess> {
  const child = new Deno.Command(Deno.execPath(), {
    args: ['run', '-A', SCRIPT],
    env: { UNION_SLOTS },
    stdout: 'null',
    stderr: 'inherit',
  }).spawn();

  for (let attempt = 0; attempt < 150; attempt++) {
    try {
      const res = await fetch(`${BASE}/`, {
        headers: { 'User-Agent': UA_CURL },
        redirect: 'manual',
      });
      await res.body?.cancel();
      return child;
    } catch {
      await new Promise((resolve) => setTimeout(resolve, 200));
    }
  }

  await stopServer(child);
  throw new Error('edge script did not start listening on ' + BASE);
}

Deno.test('rpkgs-router', async (t) => {
  const server = await startServer();

  try {
    await t.step("routes an index request to the client's R minor", async () => {
      const res = await probe(`${SLOT}/PACKAGES.gz`, UA_R45_MUSL);
      assertEquals(res.status, 302);
      assertEquals(res.location, `https://cran.rpkgs.com${SLOT}/4.5/PACKAGES.gz`);
    });

    await t.step('routes R 4.6 to its own slot', async () => {
      const res = await probe(`${SLOT}/PACKAGES.gz`, UA_R46_MUSL);
      assertEquals(res.location, `https://cran.rpkgs.com${SLOT}/4.6/PACKAGES.gz`);
    });

    await t.step('routes PACKAGES and PACKAGES.rds too', async () => {
      for (const file of ['PACKAGES', 'PACKAGES.rds']) {
        const res = await probe(`${SLOT}/${file}`, UA_R45_MUSL);
        assertEquals(res.location, `https://cran.rpkgs.com${SLOT}/4.5/${file}`, `expected ${file} to be routed`);
      }
    });

    await t.step('marks the redirect uncacheable', async () => {
      const res = await probe(`${SLOT}/PACKAGES.gz`, UA_R45_MUSL);
      assertEquals(res.cacheControl, 'no-store');
    });

    await t.step('leaves a slot outside UNION_SLOTS alone', async () => {
      const res = await probe(`${OTHER_SLOT}/PACKAGES.gz`, UA_R45_MUSL);
      assertEquals(res.location, null);
      assertEquals(res.status, 200);
    });

    await t.step('never routes a tarball', async () => {
      const res = await probe(`${SLOT}/jsonlite_2.0.0.tar.gz`, UA_R45_MUSL);
      assertEquals(res.location, null);
      assertEquals(res.status, 200);
    });

    await t.step('serves an archived binary when it exists', async () => {
      const path = `${SLOT}/Archive/xml2/xml2_1.5.2.tar.gz`;
      const res = await probe(path, UA_R45_MUSL);
      assertEquals(res.status, 200);
      assertEquals(res.location, null);
    });

    await t.step('does not redirect a path already under a minor', async () => {
      const res = await probe(`${SLOT}/4.5/PACKAGES.gz`, UA_R45_MUSL);
      assertEquals(res.location, null);
      assertEquals(res.status, 200);
    });

    await t.step('leaves a client without an R version alone', async () => {
      const res = await probe(`${SLOT}/PACKAGES.gz`, UA_CURL);
      assertEquals(res.location, null);
      assertEquals(res.status, 200);
    });

    await t.step('resolves the bare root to slot and minor', async () => {
      const res = await probe('/src/contrib/PACKAGES.gz', UA_R45_ALPINE);
      assertEquals(res.location, `https://cran.rpkgs.com${SLOT}/4.5/PACKAGES.gz`);
    });

    await t.step('sends an unidentifiable distro to CRAN', async () => {
      const res = await probe('/src/contrib/PACKAGES.gz', UA_R45_MUSL);
      assertEquals(res.location, 'https://cran.r-project.org/src/contrib/PACKAGES.gz');
    });

    await t.step('keeps the macOS rewrite', async () => {
      const res = await probe('/src/contrib/foo_1.0.tar.gz', UA_R45_DARWIN);
      assertEquals(res.location, 'https://cran.rpkgs.com/bin/macosx/big-sur-arm64/contrib/4.5/foo_1.0.tar.gz');
    });

    await t.step('keeps the macOS binary passthrough to CRAN', async () => {
      const path = '/bin/macosx/big-sur-arm64/contrib/4.5/foo_1.0.tar.gz';
      const res = await probe(path, UA_R45_DARWIN);
      assertEquals(res.location, `https://cran.r-project.org${path}`);
    });

    await t.step('collapses duplicate slashes before matching', async () => {
      const res = await probe(`/amd64/alpine324//latest/src/contrib//PACKAGES.gz`, UA_R45_MUSL);
      assertEquals(res.location, `https://cran.rpkgs.com${SLOT}/4.5/PACKAGES.gz`);
    });
  } finally {
    await stopServer(server);
  }
});