Feat: Expire static images after 90 days like R2, and link top images to direct-img in ntfy

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

README.md

@@ -305,6 +305,7 @@ Every request runs the Worker, which counts toward the free plan's 100,000 reque
 - The top 19,900 with at least `MIN_VIEWS` (20) are promoted, up to `MAX_BYTES` (4 GB) in total. The free plan allows 20,000 static files per deploy
 - Only JPEG, PNG, GIF and WebP are promoted. Static files get their type from their extension, which these have none of, so `public/_headers` declares them all `image/jpeg`, and browsers detect these four formats from the bytes anyway
 - Images are cached between runs (Actions cache), so only newly promoted ones download from R2
+- An image is dropped once it's 90 days old, the same as the R2 lifecycle rule, so a popular query (e.g. `current us president`) is still searched again instead of staying static forever. It's promoted again the next day if it's still popular
 
 Static files ignore query strings, so the zone's rules (`scripts/rules.js`) fold the site and `?i=` into the path first. The visitor's URL doesn't change:
 

scripts/notify.js

@@ -1,6 +1,6 @@
 // Sends the deploy's outcome to ntfy: what got promoted, how close the account is to the free plan's limits, and what
 // to do when something needs attention. Runs last in .github/workflows/deploy.yml, even when an earlier step failed.
-import { readFile, appendFile } from "node:fs/promises";
+import { readFile } from "node:fs/promises";
 
 const env = process.env;
 const r = JSON.parse(await readFile("promoted/report.json", "utf8").catch(() => "{}"));
@@ -31,21 +31,17 @@ if (r.workers) {
 }
 if (r.workersError) lines.push(`**Worker requests:** couldn't read (${esc(r.workersError)})`);
 
-// The top images, with thumbnails, go on the run's summary page, which the message links to. free/2/orange+cat →
-// https://free.direct-img.link/orange+cat?i=2 (a static file, so viewing the page doesn't cost Worker requests).
-if (r.top?.length && env.GITHUB_STEP_SUMMARY) {
-  const zone = env.ZONE || "direct-img.link";
-  const rows = r.top.map(({ id, views }) => {
-    const [f, i, name] = id.split("/"), url = `https://${f === "web" ? "" : `${f}.`}${zone}/${name}${i > 1 ? `?i=${i}` : ""}`;
-    return `| <img src="${url}" height="64"> | [${esc(name.replaceAll("+", " "))}](${url}) | ${f === "web" ? "main" : f} | ${i} | ${fmt(views)} |`;
-  });
-  await appendFile(env.GITHUB_STEP_SUMMARY, `## Top static images (views in the last ${r.days} days)\n\n| | Query | Site | i | Views |\n|---|---|---|---|---|\n${rows.join("\n")}\n`);
-}
+// web/2/orange+cat → [orange cat](https://direct-img.link/orange+cat?i=2), so each opens on its site
+const zone = env.ZONE || "direct-img.link";
+const link = id => {
+  const [f, i, name] = id.split("/");
+  return `[${esc(name.replaceAll("+", " "))}](https://${f === "web" ? "" : `${f}.`}${zone}/${name}${i > 1 ? `?i=${i}` : ""})${i > 1 ? ` i=${i}` : ""}${f === "web" ? "" : ` (${f})`}`;
+};
 
 const failed = env.STATUS !== "success";
 const body = [
   lines.join("  \n"),
-  r.top?.length && env.RUN_URL && `[See top images](${env.RUN_URL})`,
+  r.top?.length && `**Top images**\n${r.top.map((t, n) => `${n + 1}. ${link(t.id)} — ${fmt(t.views)}`).join("\n")}`,
   actions.length && `**⚠ Needs attention**\n${actions.map(a => `- ${a}`).join("\n")}`,
 ].filter(Boolean).join("\n\n") || "No report was written.";
 console.log(body);

scripts/promote.js

@@ -4,7 +4,7 @@
 // promoted/report.json for scripts/notify.js. promoted/ is kept between runs (Actions cache), so only new images download.
 import { mkdir, readFile, writeFile, link, rm, readdir } from "node:fs/promises";
 import { dirname } from "node:path";
-import { MAX_INDEX, siteOf, folder, normalizeQuery, cacheKey, r2Key } from "../src/_utils/keys.js";
+import { TTL_SECONDS, MAX_INDEX, siteOf, folder, normalizeQuery, cacheKey, r2Key } from "../src/_utils/keys.js";
 import { parseFreeSource } from "../src/_utils/free.js";
 import { parseAiSource } from "../src/_utils/ai.js";
 
@@ -83,20 +83,24 @@ async function workerRequests(day) {
   return { total: Object.values(byScript).reduce((t, r) => t + r, 0), byScript };
 }
 
-// The image from the last run's cache, or from R2. Skips a failed search's empty marker and formats browsers can't detect.
+// The image from the last run's cache, or from R2. Skips a failed search's empty marker, formats browsers can't detect,
+// and images past TTL_SECONDS, so a popular query gets searched again like any other instead of staying static forever.
+const expired = uploaded => Date.now() - uploaded >= TTL_SECONDS * 1000;
 async function fetchImage(img, prev) {
-  const file = `${CACHE}/files/${img.id}`;
-  if (prev[img.id] && await readFile(file).then(() => true, () => false)) return { ...prev[img.id], cached: true };
+  const file = `${CACHE}/files/${img.id}`, p = prev[img.id];
+  if (p?.uploaded && await readFile(file).then(() => true, () => false)) return expired(p.uploaded) ? { skip: "expired" } : { ...p, cached: true };
   const res = await api(`/accounts/${env.CLOUDFLARE_ACCOUNT_ID}/r2/buckets/${BUCKET}/objects/${img.r2}`);
   if (res.status === 404) return { skip: "missing" };
   if (!res.ok) return { skip: "failed", error: `${img.id}: HTTP ${res.status}` };
   const type = res.headers.get("content-type")?.split(";")[0].trim().toLowerCase(), buf = Buffer.from(await res.arrayBuffer());
+  const uploaded = Date.parse(res.headers.get("last-modified"));
   if (!buf.length) return { skip: "missing" };
+  if (expired(uploaded)) return { skip: "expired" };
   if (!TYPES.includes(type)) return { skip: "type" };
   if (buf.length > 25 * 2 ** 20) return { skip: "size" };
   await mkdir(dirname(file), { recursive: true });
   await writeFile(file, buf);
-  return { bytes: buf.length, type };
+  return { bytes: buf.length, type, uploaded };
 }
 
 const report = { ok: false, minViews: MIN_VIEWS, days: DAYS, maxFiles: MAX_FILES, maxBytes: MAX_BYTES, skipped: {} };
@@ -140,7 +144,7 @@ for (let k = 0; k < ranked.length; k += 4) {
     } else if (n >= MAX_FILES) report.notFitted = (report.notFitted || 0) + 1;
     else if (bytes + g.bytes > MAX_BYTES) report.skipped.budget = (report.skipped.budget || 0) + 1;
     else {
-      chosen[img.id] = { views: img.views, bytes: g.bytes, type: g.type, r2: img.r2 };
+      chosen[img.id] = { views: img.views, bytes: g.bytes, type: g.type, r2: img.r2, uploaded: g.uploaded };
       bytes += g.bytes;
       n++;
     }
@@ -167,7 +171,7 @@ Object.assign(report, {
   ok: true, promoted: ids.length, bytes,
   added: ids.filter(id => !prev[id]).length, removed: Object.keys(prev).filter(id => !chosen[id]).length,
   promotedViews: ids.reduce((a, id) => a + chosen[id].views, 0),
-  top: ids.slice(0, 50).map(id => ({ id, views: chosen[id].views })),
+  top: ids.slice(0, 10).map(id => ({ id, views: chosen[id].views })),
 });
 await writeFile(`${CACHE}/report.json`, JSON.stringify(report, null, 2));
 console.log(JSON.stringify(report, null, 2));

src/_utils/keys.js

@@ -1,5 +1,9 @@
 // Shared by the worker and scripts/promote.js, so a promoted image's R2 key always matches the one the worker stored.
 
+// How long an image is kept: the R2 bucket's lifecycle rule deletes it then, so the query is searched again, and
+// promote.js stops serving it as a static file
+export const TTL_SECONDS = 90 * 24 * 60 * 60;
+
 // Highest ?i= per site: 40 on the main site (100 Brave results; a Firecrawl query's ?i= past FIRECRAWL_RESULTS gets
 // an AI image), 20 on free.* (~50 Openverse + Wikimedia results) and 1000 on ai.* (each is a new image, so there's no
 // list to run out).

src/worker.js

@@ -4,10 +4,9 @@ import { freeImageUrls, parseFreeSource, FREE_UA } from "./_utils/free.js";
 import { isFlickrPlaceholder } from "./_utils/flickr.js";
 import { aiImage, parseAiSource, DEFAULT_AI_SOURCE } from "./_utils/ai.js";
 import { surreal } from "./_utils/surreal.js";
-import { MAX_INDEX, siteOf, folder, normalizeQuery, cacheKey as keyFor, r2Key as r2KeyFor, sha256 } from "./_utils/keys.js";
+import { TTL_SECONDS, MAX_INDEX, siteOf, folder, normalizeQuery, cacheKey as keyFor, r2Key as r2KeyFor, sha256 } from "./_utils/keys.js";
 import * as LIMITS from "../public/assets/rate-limit.js";
 
-const TTL_SECONDS = 90 * 24 * 60 * 60;
 // How long a failed search keeps serving bad.webp before the query is searched again
 const ERR_SECONDS = 24 * 60 * 60;
 // How long a search may hold its query's lock; longer than the 20s image deadline plus the search itself