an image's header, by path, because ffprobe cannot read a HEIC

`/v1/probe` is ffprobe, which is the wrong tool for a still and fails at it
loudly: a HEIC out of an iPhone is not a container it understands, so it rejects
the file outright with "moov atom not found" rather than reporting no
dimensions.

So anything needing a photograph's size had nowhere to ask — and a caller
building a width ladder needs exactly that, to know which rungs would be an
upscale.

sharp reads it, HEIC included, because `metadata()` only parses the header. That
is precisely the part of a HEIC that works; the pixels are what its bundled
libheif cannot reach, which is what decodable.ts routes around. No decode
happens, so this needs no job slot.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

sidecar/index.ts

@@ -1,6 +1,6 @@
 import { timingSafeEqual } from "node:crypto";
 import { ASSETS_DIR, MAX_CONCURRENT_JOBS, MAX_UPLOAD_BYTES, PORT, requireToken } from "./config";
-import { ApiError, invalid, notFound, toErrorResponse } from "./errors";
+import { ApiError, invalid, notFound, toErrorResponse, unprocessable } from "./errors";
 import { contentPath, deleteAsset, putAsset, readAsset } from "./assets/store";
 import { ffprobe } from "./probe/ffprobe";
 import { runSharp, type SharpRequest } from "./sharp/run";
@@ -225,6 +225,38 @@ async function route(request: Request): Promise<Response> {
     return Response.json(await withSlot(() => ffprobe(target)));
   }
 
+  /*
+    An image's header, by path.
+
+    Separate from `/v1/probe` because ffprobe is the wrong tool for a still and fails at it loudly: a
+    HEIC out of an iPhone is not a media container it understands, and it rejects the file outright
+    ("moov atom not found") rather than reporting no dimensions. Anything that needs a photograph's
+    size — a caller building a width ladder, and therefore needing to know which rungs would be an
+    upscale — had nowhere to ask.
+
+    sharp reads it, including HEIC, because `metadata()` only parses the header. That is precisely the
+    part of a HEIC that works; the pixels are what its bundled libheif cannot reach, which is what
+    `decodable.ts` exists to route around. No decode happens here, so this is cheap and needs no job
+    slot: it is a file header read, not an image operation.
+  */
+  if (path === "/v1/image" && request.method === "GET") {
+    const target = await safeInputPath(url.searchParams.get("path"), "path");
+    const meta = await sharp(target)
+      .metadata()
+      .catch(() => {
+        throw unprocessable(`Could not read that image: ${target}`);
+      });
+    return Response.json({
+      format: meta.format ?? null,
+      width: meta.width ?? null,
+      height: meta.height ?? null,
+      space: meta.space ?? null,
+      hasAlpha: meta.hasAlpha ?? false,
+      pages: meta.pages ?? 1,
+      orientation: meta.orientation ?? null,
+    });
+  }
+
   const asset = path.match(/^\/v1\/assets\/([^/]+)(\/content|\/probe)?$/);
   if (asset) {
     const [, id, suffix] = asset;