fix(cdn): align repository routing across pull zones

edge/rpkgs-router.test.ts

@@ -117,6 +117,13 @@ Deno.test('rpkgs-router', async (t) => {
       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);

edge/rpkgs-router.ts

@@ -27,6 +27,7 @@ import * as BunnySDK from 'https://esm.sh/@bunny.net/edgescript-sdk@0.12';
 
 const PUBLIC_CDN_ORIGIN = 'https://cran.rpkgs.com';
 const CRAN_ORIGIN = 'https://cran.r-project.org';
+const PUBLIC_CDN_HOSTS = new Set(['cran.rpkgs.com', 'cran.allianceswisspass.devxy.io']);
 
 /** Slots ("<arch>/<os>", comma separated) whose per-minor index is a union. */
 const UNION_SLOTS = new Set(
@@ -47,6 +48,10 @@ const INDEX_FILE_REGEX = /^PACKAGES(\.gz|\.rds)?$/;
 
 const SRC_CONTRIB_REGEX = /^\/src\/contrib\/(.+)$/;
 
+/** A binary archive URL whose upstream source counterpart CRAN can serve. */
+const ARCHIVE_TARBALL_REGEX =
+  /^\/(?:amd64|arm64)\/[a-z0-9._-]+\/latest\/src\/contrib\/Archive\/([^/]+)\/([^/]+\.tar\.gz)$/;
+
 const MACOS_BIN_REGEX =
   /^\/bin\/macosx\/(big-sur-arm64|big-sur-x86_64|monterey-arm64|monterey-x86_64)\/contrib\/([0-9.]+)\/(.+)$/;
 
@@ -85,6 +90,10 @@ function redirectTo(location: string, status = 302): Response {
   });
 }
 
+function publicCdnOrigin(url: URL): string {
+  return PUBLIC_CDN_HOSTS.has(url.hostname) ? url.origin : PUBLIC_CDN_ORIGIN;
+}
+
 function extractRMinor(userAgent: string): string | null {
   for (const regex of R_MINOR_REGEXES) {
     const match = userAgent.match(regex);
@@ -181,15 +190,14 @@ BunnySDK.net.http
     const url = new URL(ctx.request.url);
     const path = normalizePathname(url.pathname);
     const userAgent = ctx.request.headers.get('User-Agent') || '';
+    const publicOrigin = publicCdnOrigin(url);
 
     // macOS clients are served from CRAN's own binary tree.
     const srcContrib = path.match(SRC_CONTRIB_REGEX);
     if (srcContrib && /darwin/.test(userAgent)) {
       const mac = parseMacUserAgent(userAgent);
       if (mac) {
-        return Promise.resolve(
-          redirectTo(`${PUBLIC_CDN_ORIGIN}/bin/macosx/${mac.os}/contrib/${mac.rver}/${srcContrib[1]}`),
-        );
+        return Promise.resolve(redirectTo(`${publicOrigin}/bin/macosx/${mac.os}/contrib/${mac.rver}/${srcContrib[1]}`));
       }
     }
 
@@ -213,7 +221,7 @@ BunnySDK.net.http
       if (target === path) {
         return Promise.resolve(ctx.request);
       }
-      return Promise.resolve(redirectTo(`${PUBLIC_CDN_ORIGIN}${target}`));
+      return Promise.resolve(redirectTo(`${publicOrigin}${target}`));
     }
 
     // The bare `https://cran.rpkgs.com` form, resolved from the User-Agent.
@@ -224,12 +232,27 @@ BunnySDK.net.http
       }
 
       const rest = srcContrib ? srcContrib[1] : '';
-      return Promise.resolve(redirectTo(`${PUBLIC_CDN_ORIGIN}${contribPath(slot, rest, userAgent)}`));
+      return Promise.resolve(redirectTo(`${publicOrigin}${contribPath(slot, rest, userAgent)}`));
     }
 
     return Promise.resolve(ctx.request);
   })
-  .onOriginResponse((ctx) => {
+  .onOriginResponse(async (ctx) => {
+    const path = normalizePathname(new URL(ctx.request.url).pathname);
+    const archive = path.match(ARCHIVE_TARBALL_REGEX);
+
+    // Binary archives can be incomplete when an older build never succeeded.
+    // Preserve renv/remotes version restores by falling back to CRAN's source
+    // package only for an absent archived tarball. A requested version can be
+    // either archived upstream or still current, so probe the archive first.
+    // Other 404s remain visible.
+    if (ctx.response.status === 404 && archive) {
+      const archiveUrl = `${CRAN_ORIGIN}/src/contrib/Archive/${archive[1]}/${archive[2]}`;
+      const archiveResponse = await fetch(archiveUrl, { method: 'HEAD' });
+      const sourceUrl = archiveResponse.ok ? archiveUrl : `${CRAN_ORIGIN}/src/contrib/${archive[2]}`;
+      return redirectTo(sourceUrl);
+    }
+
     ctx.response.headers.append('X-Via', 'MyMiddleware');
     return Promise.resolve(ctx.response);
   });