Give backup album folders distinct names and remove stale entries (closes #103)

Within a collection's folder, files whose sanitized titles match (ignoring
case) each get their file ID added before the extension, and collections
whose sanitized names match get their ID added, so no symlink or JSON
replaces another. Names are chosen across all collections, so a scoped run
names folders the same as a full one.

Each run first removes symlinks into originals/ that no longer belong to a
collection, and the folders quak wrote (a sibling JSON with an album ID) for
collections that are gone or renamed. Anything else is left alone; a folder
still holding user files keeps its JSON.

Model: opus-5-5

README.md

@@ -471,6 +471,11 @@ accepted for backward compatibility but ignored. `backup-metadata --exif` (alias
 metadata. The listing and backup commands support `--json` for machine-readable
 output.
 
+`backup-metadata` fetches ML data in requests of up to 200 files. When a request
+still fails after its retries, the error is logged, each of its files is written
+with the reason in an `mlDataError` field instead of `mlData`, and the dump goes
+on. The exit code is non-zero if any ML data request failed.
+
 `helper fix-missing-thumbnails` regenerates thumbnails for baseline JPEG images
 only, because the bundled decoder (`jpeg-js`) decodes only JPEG. A non-JPEG
 image (PNG, HEIC) or a video is reported as `skipped` (unsupported format), kept
@@ -690,6 +695,11 @@ Under `cacheDirectory`:
         fetched.json              per-file fetch bookkeeping
 ```
 
+When `metadata.json` belongs to a different account than the client's,
+`Library.open` deletes it and `mldata/` and starts from an empty cache. Cached
+originals and thumbnails are kept; they are reached only through the files the
+current account's records name.
+
 A stored file appears only via an atomic temp-then-rename, so its presence means
 it is complete. The design also calls for a content-hash comparison against
 `FileMetadata.hash` on each fetched original; that check is deferred (issue

TODO.md

@@ -25,6 +25,17 @@ Tag v1.0.0.
   `originals/` for files no longer in the collection, and the folders of deleted
   or renamed collections, leaving anything else in `collections/` alone. The
   README backup layout states the naming rule.
+- 2026-09-23: Kept one account's cache from mixing with another's (issue 104).
+  When `metadata.json` in the cache directory was written for a different,
+  non-zero user ID than the client's, `Library.open` deletes it and `mldata/`
+  and starts empty, so the first refresh enumerates from 0. This only happens
+  with `--cache-dir` or an explicit `cacheDirectory`; the default path already
+  includes the user ID. A test opens one account's cache as another account.
+- 2026-09-23: `backup-metadata` no longer stops on one failed ML data request
+  (issue 101). Each request of up to 200 files is tried on its own; a failed one
+  is logged, its files are written with the reason in `mlDataError`, and the
+  command exits 1 once the dump is complete. `fetchMLData`, which only this
+  command used, is gone; the command calls `fetchMLDataBatch` per batch.
 - 2026-09-23: Single-sourced the version string (issue 5). `package.json` is the
   only place it is written: `src/index.ts` imports it for `VERSION` and
   `bin/quak.ts` passes `VERSION` to commander. tsc copies `package.json` to

src/cli-commands.ts

@@ -323,11 +323,11 @@ export const backupMetadataCommand = async (
     if (!client) return 1;
     const lib = await openReadLibrary(ctx, client);
     try {
-        await runMetadataBackup(lib, client, dir, {
+        const { failedMLBatches } = await runMetadataBackup(lib, client, dir, {
             exif: opts.exif || opts.all,
             onProgress: (msg) => ctx.stderr.write(msg + "\n"),
         });
-        return 0;
+        return failedMLBatches > 0 ? 1 : 0;
     } finally {
         await lib.close();
     }

src/library/index.ts

@@ -26,6 +26,7 @@
 // store marked unsaved until a later save actually lands, so a stuck disk is
 // never masked by a subsequent empty refresh.
 
+import { rm } from "node:fs/promises";
 import { join } from "node:path";
 import envPaths from "env-paths";
 
@@ -345,9 +346,20 @@ export class Library {
         const cacheDirectory =
             opts.cacheDirectory ??
             join(envPaths("quak", { suffix: "" }).cache, String(userID));
-        const store = await MetadataStore.load(
+        const metadataPath = join(cacheDirectory, "metadata.json");
-            join(cacheDirectory, "metadata.json"),
+        let store = await MetadataStore.load(metadataPath);
-        );
+        // A cache directory given explicitly can hold another account's cache.
+        // Its records and cursor are not this account's, so delete it and the
+        // ML data beside it and start empty. A user ID of 0 means the cache
+        // was never refreshed and so holds nothing to discard.
+        if (store.userID !== 0 && store.userID !== userID) {
+            await rm(metadataPath, { force: true });
+            await rm(join(cacheDirectory, "mldata"), {
+                recursive: true,
+                force: true,
+            });
+            store = await MetadataStore.load(metadataPath);
+        }
         const intervalMs =
             (opts.refreshIntervalSeconds ?? DEFAULT_REFRESH_INTERVAL_SECONDS) *
             1000;

src/metadata-backup.ts

@@ -5,7 +5,11 @@ import exifReader from "exif-reader";
 import type { Client } from "./client.js";
 import type { Library, Photo } from "./library/index.js";
 import { sanitizeFileName } from "./filename.js";
-import { fetchMLData } from "./mldata-fetch.js";
+import {
+    fetchMLDataBatch,
+    MLDATA_BATCH_SIZE,
+    type MLData,
+} from "./mldata-fetch.js";
 import type { EnteFile } from "./model/types.js";
 
 export type ProgressCallback = (message: string) => void;
@@ -137,13 +141,14 @@ const extractExif = async (
 // of plain JSON: account, per-collection, and per-file records including the
 // private and public magic metadata and (by default) the ML data. Collections
 // and files are enumerated from the library's cache rather than a fresh server
-// scan; the ML fetch and EXIF extraction are unchanged.
+// scan. Returns how many ML data requests failed; their files are still
+// written, with `mlDataError` in place of `mlData`.
 export const runMetadataBackup = async (
     lib: Library,
     client: Client,
     outDir: string,
     opts?: MetadataBackupOptions,
-): Promise<void> => {
+): Promise<{ failedMLBatches: number }> => {
     const log = opts?.onProgress ?? (() => {});
     const wantExif = opts?.exif ?? false;
 
@@ -208,12 +213,31 @@ export const runMetadataBackup = async (
         }
     }
 
+    // One failed request (retries exhausted) must not end the dump: its files
+    // get the reason in `mlDataError` and the other batches go on.
     log("Fetching ML data (face detections, CLIP embeddings)...");
-    const mlDataMap = await fetchMLData(
+    const mlDataMap = new Map<number, MLData>();
+    const mlDataErrors = new Map<number, string>();
+    let failedMLBatches = 0;
+    const fileIDs = [...fileKeys.keys()];
+    for (let i = 0; i < fileIDs.length; i += MLDATA_BATCH_SIZE) {
+        const batch = fileIDs.slice(i, i + MLDATA_BATCH_SIZE);
+        try {
+            const result = await fetchMLDataBatch(
                 client.getApiClient(),
-        [...fileKeys.keys()],
+                batch,
                 fileKeys,
             );
+            for (const [id, payload] of result) mlDataMap.set(id, payload);
+        } catch (err) {
+            const reason = err instanceof Error ? err.message : String(err);
+            failedMLBatches++;
+            log(
+                `ML data request for ${batch.length} file(s) failed: ${reason}`,
+            );
+            for (const id of batch) mlDataErrors.set(id, reason);
+        }
+    }
     log(`Got ML data for ${mlDataMap.size} file(s)`);
 
     const writtenFileIDs = new Set<number>();
@@ -233,6 +257,8 @@ export const runMetadataBackup = async (
 
         const ml = mlDataMap.get(file.id);
         if (ml) fileMeta.mlData = ml;
+        const mlError = mlDataErrors.get(file.id);
+        if (mlError) fileMeta.mlDataError = mlError;
 
         if (wantExif && !writtenFileIDs.has(file.id)) {
             log(`[${file.metadata.title}] Extracting EXIF...`);
@@ -253,4 +279,5 @@ export const runMetadataBackup = async (
     }
 
     log("Metadata backup complete.");
+    return { failedMLBatches };
 };

src/mldata-fetch.ts

@@ -5,9 +5,8 @@
 // comes back encrypted under the file's own key and gzipped; decrypting and
 // gunzipping yields the JSON payload
 // `{ face: { faces: [...] }, clip: { embedding } }`. Ente caps a request at 200
-// ids, so `fetchMLData` batches for callers that want many at once while
+// ids, so callers that want many at once split them into batches of
-// `fetchMLDataBatch` is the single-request unit the library submits to its
+// `MLDATA_BATCH_SIZE` and call `fetchMLDataBatch` once per batch.
-// request pool.
 
 import { gunzipSync } from "node:zlib";
 
@@ -69,25 +68,3 @@ export const fetchMLDataBatch = async (
     }
     return result;
 };
-
-// Fetch ML data for arbitrarily many ids, batching at `MLDATA_BATCH_SIZE`. Used
-// by the one-shot metadata backup; the library fetches through its request pool
-// with `fetchMLDataBatch` instead.
-export const fetchMLData = async (
-    api: ApiClient,
-    fileIDs: number[],
-    fileKeys: Map<number, Uint8Array>,
-): Promise<Map<number, MLData>> => {
-    const result = new Map<number, MLData>();
-    for (let i = 0; i < fileIDs.length; i += MLDATA_BATCH_SIZE) {
-        const batch = fileIDs.slice(i, i + MLDATA_BATCH_SIZE);
-        for (const [id, payload] of await fetchMLDataBatch(
-            api,
-            batch,
-            fileKeys,
-        )) {
-            result.set(id, payload);
-        }
-    }
-    return result;
-};

test/cli/backup.test.ts

@@ -735,6 +735,53 @@ describe("backup album folders", () => {
         expect(tree(outDir)).toEqual(before);
     });
 
+    it("leaves the albums an onlyAlbumNames run skips as they were", async () => {
+        const outDir = join(root, "backup");
+        // "trip" is skipped by the scoped run but its name clashes with the
+        // in-scope "Trip", so "Trip" must keep its ID suffix.
+        const lib = libraryOf([
+            {
+                collection: collection(10, "Trip"),
+                files: [file(1, 10, "a.jpg")],
+            },
+            {
+                collection: collection(11, "trip"),
+                files: [file(2, 11, "b.jpg")],
+            },
+            {
+                collection: collection(12, "Work"),
+                files: [file(3, 12, "c.jpg")],
+            },
+        ]);
+        const json = (name: string): string =>
+            readFileSync(join(outDir, "collections", name), "utf-8");
+
+        await runBackup(lib, { downloadDirectory: outDir });
+        const before = tree(outDir);
+        const skippedJSON = [json("trip (11).json"), json("Work.json")];
+        const scoped = await runBackup(lib, {
+            downloadDirectory: outDir,
+            onlyAlbumNames: ["Trip"],
+        });
+
+        expect(scoped.failed).toBe(0);
+        expect(before).toEqual([
+            "Trip (10)/",
+            "Trip (10)/a.jpg -> ../../originals/1.jpg",
+            "Trip (10).json",
+            "Work/",
+            "Work/c.jpg -> ../../originals/3.jpg",
+            "Work.json",
+            "trip (11)/",
+            "trip (11)/b.jpg -> ../../originals/2.jpg",
+            "trip (11).json",
+        ]);
+        expect(tree(outDir)).toEqual(before);
+        expect([json("trip (11).json"), json("Work.json")]).toEqual(
+            skippedJSON,
+        );
+    });
+
     it("removes links and album folders that are gone, and nothing the user added", async () => {
         const outDir = join(root, "backup");
         const albums: Album[] = [

test/cli/metadata-backup.test.ts

@@ -38,7 +38,7 @@ import { join } from "node:path";
 import { tmpdir } from "node:os";
 import sodium from "libsodium-wrappers-sumo";
 import { SRP, SrpServer } from "fast-srp-hap";
-import { beforeAll, afterAll, describe, expect, it } from "vitest";
+import { beforeAll, afterAll, describe, expect, it, vi } from "vitest";
 import {
     init,
     toBase64,
@@ -53,8 +53,16 @@ import {
     runMetadataBackup,
     type MetadataBackupOptions,
 } from "../../src/metadata-backup.js";
+import { backupMetadataCommand } from "../../src/cli-commands.js";
 import type { KeyAttributes } from "../../src/auth/types.js";
 
+// One file per ML data request, so the two files of the mock account are
+// fetched in two requests and one of them can fail on its own.
+vi.mock("../../src/mldata-fetch.js", async (importOriginal) => ({
+    ...(await importOriginal<typeof import("../../src/mldata-fetch.js")>()),
+    MLDATA_BATCH_SIZE: 1,
+}));
+
 const TEST_EMAIL = "metabackup@example.com";
 const TEST_PASSWORD = "metapass";
 const TEST_OPS = 2;
@@ -347,7 +355,8 @@ const buildMetaMock = async (): Promise<MetaMockState> => {
     };
 };
 
-const buildMetaFetch = (m: MetaMockState) => {
+// `failMLDataFor`: answer 500 to every ML data request that asks for this file.
+const buildMetaFetch = (m: MetaMockState, failMLDataFor?: number) => {
     let srpServer: SrpServer;
     return (async (
         input: RequestInfo | URL,
@@ -403,6 +412,8 @@ const buildMetaFetch = (m: MetaMockState) => {
         }
         if (path === "/files/data/fetch") {
             const body = JSON.parse(init?.body as string);
+            if ((body.fileIDs as number[]).includes(failMLDataFor!))
+                return new Response("server error", { status: 500 });
             const data = (body.fileIDs as number[])
                 .filter((id: number) => m.encryptedMLData[id])
                 .map((id: number) => ({
@@ -636,3 +647,63 @@ describe("quak backup-metadata", () => {
         expect(failedMeta.imageMetadataError).toEqual(expect.any(String));
     });
 });
+
+describe("quak backup-metadata when an ML data request fails", () => {
+    // Run the CLI command against the mock and return its exit code, stderr
+    // and output directory.
+    const runCommand = async (failMLDataFor?: number) => {
+        const client = await Client.login({
+            email: TEST_EMAIL,
+            password: TEST_PASSWORD,
+            apiOptions: {
+                fetch: buildMetaFetch(mock, failMLDataFor),
+                retry: { sleep: async () => {} },
+            },
+        });
+        const outDir = mkdtempSync(join(testDir, "ml-fail-"));
+        let stderr = "";
+        const code = await backupMetadataCommand(
+            {
+                stdout: { write: () => true },
+                stderr: { write: (text: string) => (stderr += text) },
+                sessionDir: testDir,
+                cacheDir: mkdtempSync(join(testDir, "cache-")),
+                loadSession: () => client,
+            },
+            outDir,
+            {},
+        );
+        return { code, stderr, outDir };
+    };
+
+    it("writes every file, marks the failed batch's files, and exits 1", async () => {
+        const { code, stderr, outDir } = await runCommand(200);
+
+        expect(code).toBe(1);
+        expect(stderr).toContain("ML data request for 1 file(s) failed");
+
+        const ok = JSON.parse(
+            readFileSync(
+                join(outDir, "collections", "10-Vacation", "100.json"),
+                "utf-8",
+            ),
+        );
+        expect(ok.mlData.clip.embedding).toEqual([0.5, 0.6, 0.7]);
+        expect(ok.mlDataError).toBeUndefined();
+
+        const failed = JSON.parse(
+            readFileSync(
+                join(outDir, "collections", "20-__Work", "200.json"),
+                "utf-8",
+            ),
+        );
+        expect(failed.metadata.title).toBe("diagram.png");
+        expect(failed.mlData).toBeUndefined();
+        expect(failed.mlDataError).toContain("500");
+    });
+
+    it("exits 0 when every ML data request succeeds", async () => {
+        const { code } = await runCommand();
+        expect(code).toBe(0);
+    });
+});

test/library/mldata.test.ts

@@ -13,6 +13,8 @@
  *   2. `Library` wiring: after each refresh the library fetches ML data through
  *      the metadata pool for every known file not yet cached, is incremental on
  *      later refreshes, and refetches a file whose `updationTime` advanced.
+ *      Opening a cache directory written by another account starts empty,
+ *      its ML data included (issue #104).
  *
  * Embedding values are chosen to be exactly representable as float32 so the
  * round-trip through `clip.f32` compares equal.
@@ -498,4 +500,63 @@ describe("Library ML-data fetch on refresh", () => {
             await lib.close();
         }
     });
+
+    it("starts empty when the cache directory holds another account's cache", async () => {
+        // Account A fills the cache directory: metadata and ML data.
+        const clientA = new MLMockClient();
+        clientA.collectionsQueue.push({
+            collections: [collection(1, 100)],
+            deleted: [],
+            cursor: 100,
+        });
+        clientA.filesFor(1, {
+            files: [file(1001, 1, 90)],
+            deleted: [],
+            cursor: 90,
+        });
+        clientA.mlByFile.set(1001, payload([0.5, 0.25, 0.75]));
+        const libA = await Library.open({
+            client: clientA,
+            cacheDirectory,
+            refreshIntervalSeconds: 3600,
+        });
+        try {
+            await vi.waitFor(
+                () => expect(libA.status().lastMLFetchAt).toBeGreaterThan(0),
+                { timeout: 2000, interval: 5 },
+            );
+        } finally {
+            await libA.close();
+        }
+
+        // Account B opens the same directory.
+        const clientB = new MLMockClient();
+        clientB.userID = USER_ID + 1;
+        const sinceTimes: number[] = [];
+        const realCollectionsSince = clientB.collectionsSince.bind(clientB);
+        clientB.collectionsSince = async (args) => {
+            sinceTimes.push(args.sinceTime);
+            return realCollectionsSince(args);
+        };
+        const libB = await Library.open({
+            client: clientB,
+            cacheDirectory,
+            refreshIntervalSeconds: 3600,
+        });
+        try {
+            expect(sinceTimes[0]).toBe(0);
+            expect(libB.status().userID).toBe(USER_ID + 1);
+            expect(libB.listCollections()).toEqual([]);
+            expect(libB.getFile(1, 1001)).toBeUndefined();
+            expect(await libB.mldata.forFile({ fileID: 1001 })).toBeUndefined();
+            expect(
+                libB.mldata.searchByEmbedding({ embedding: [0.5, 0.25, 0.75] }),
+            ).toEqual([]);
+            expect(
+                existsSync(join(cacheDirectory, "mldata", "1001.json")),
+            ).toBe(false);
+        } finally {
+            await libB.close();
+        }
+    });
 });