ML data included by default, --exif is the opt-in, --all aliases --exif

ML data (face detections, CLIP embeddings) is now fetched by default
in backup-metadata. Use --no-ml to skip it. EXIF extraction (which
requires downloading every file) remains opt-in via --exif. --all is
an alias for --exif.
This commit is contained in:
2026-06-09 17:38:15 -04:00
parent 8cd57f4d12
commit 21a1a78f07
3 changed files with 41 additions and 16 deletions

View File

@@ -340,23 +340,26 @@ program
"Dump all decrypted account metadata to a directory of JSON files", "Dump all decrypted account metadata to a directory of JSON files",
) )
.argument("<dir>", "Output directory") .argument("<dir>", "Output directory")
.option( .option("--no-ml", "Skip ML data (face detections, CLIP embeddings)")
"--ml",
"Include ML data (face detections, CLIP embeddings) from the Ente server",
)
.option( .option(
"--exif", "--exif",
"Download each file and extract full EXIF/IPTC/XMP metadata (slow)", "Download each file and extract full EXIF/IPTC/XMP metadata (slow)",
) )
.action(async (dir: string, opts: { ml?: boolean; exif?: boolean }) => { .option("--all", "Alias for --exif")
.action(
async (
dir: string,
opts: { ml?: boolean; exif?: boolean; all?: boolean },
) => {
await init(); await init();
const client = requireSession(); const client = requireSession();
await runMetadataBackup(client, dir, { await runMetadataBackup(client, dir, {
mlData: opts.ml, mlData: opts.ml,
exif: opts.exif, exif: opts.exif || opts.all,
onProgress: (msg) => stderr.write(msg + "\n"), onProgress: (msg) => stderr.write(msg + "\n"),
}); });
}); },
);
program program
.command("backup") .command("backup")

View File

@@ -120,7 +120,7 @@ export const runMetadataBackup = async (
opts?: MetadataBackupOptions, opts?: MetadataBackupOptions,
): Promise<void> => { ): Promise<void> => {
const log = opts?.onProgress ?? (() => {}); const log = opts?.onProgress ?? (() => {});
const wantML = opts?.mlData ?? false; const wantML = opts?.mlData ?? true;
const wantExif = opts?.exif ?? false; const wantExif = opts?.exif ?? false;
mkdirSync(outDir, { recursive: true }); mkdirSync(outDir, { recursive: true });

View File

@@ -587,8 +587,8 @@ describe("quak backup-metadata", () => {
expect(fileMeta.mlData.clip.embedding).toEqual([0.5, 0.6, 0.7]); expect(fileMeta.mlData.clip.embedding).toEqual([0.5, 0.6, 0.7]);
}); });
it("does not include ML data when --ml is not set", async () => { it("includes ML data by default", async () => {
const outDir = join(testDir, "no-ml"); const outDir = join(testDir, "ml-default");
const client = await Client.login({ const client = await Client.login({
email: TEST_EMAIL, email: TEST_EMAIL,
password: TEST_PASSWORD, password: TEST_PASSWORD,
@@ -597,6 +597,28 @@ describe("quak backup-metadata", () => {
await runMetadataBackup(client, outDir); await runMetadataBackup(client, outDir);
const collDirs = readdirSync(join(outDir, "collections"));
const vacDir = collDirs.find((d) => d.includes("Vacation"))!;
const fileMeta = JSON.parse(
readFileSync(
join(outDir, "collections", vacDir, "100.json"),
"utf-8",
),
);
expect(fileMeta.mlData).toBeDefined();
expect(fileMeta.mlData.face.faces[0].faceID).toBe("face-abc");
});
it("excludes ML data when mlData: false", async () => {
const outDir = join(testDir, "no-ml");
const client = await Client.login({
email: TEST_EMAIL,
password: TEST_PASSWORD,
apiOptions: { fetch: buildMetaFetch(mock) },
});
await runMetadataBackup(client, outDir, { mlData: false });
const collDirs = readdirSync(join(outDir, "collections")); const collDirs = readdirSync(join(outDir, "collections"));
const vacDir = collDirs.find((d) => d.includes("Vacation"))!; const vacDir = collDirs.find((d) => d.includes("Vacation"))!;
const fileMeta = JSON.parse( const fileMeta = JSON.parse(