fix(reader): the Takeout import holds up under abuse, races and a run it cannot see

- Uploads: one at a time per user and ten an hour, each refused only after its body arrived;
  the history read an entry at a time (32 MiB cap), a Watch later list capped at YouTube's 5000,
  a user's kept videos at 200k; any unreadable shape is a 422, not a 500.
- Quota: Watch later lookups stop at the backfill reserve and at a per-user daily share; what is
  not asked about waits, known videos still go in, and the shorts job retries the rest.
- Staging and settling share one per-user advisory lock (no deadlock); a run settles only the
  rows it read, so a row staged meanwhile is never called gone.
- A failed run's notice names a translated failure, never the upstream's text.
- The status says when an import is queued; the page watches it to its end even when left, then
  reads the feed, its counts and the playlists again.
- Settings: one live region, the zone kept (unavailable) while a run goes, focus handed back,
  a missed drop kept from the browser, an oversize file refused before it is sent.
- Scheduler and run lists show Takeout runs by the importing user.
This commit is contained in:
2026-10-23 20:48:23 +02:00
parent e0b02a0c14
commit 9f580fd8c8
29 changed files with 1235 additions and 213 deletions
+140 -67
View File
@@ -1,4 +1,4 @@
import { useEffect, useState } from "react";
import { useEffect, useRef, useState } from "react";
import { useTranslation } from "react-i18next";
import { useQuery, useQueryClient } from "@tanstack/react-query";
import { RefreshCw, UploadCloud } from "lucide-react";
@@ -6,19 +6,19 @@ import { Section } from "../../components/ui/form";
import { Button } from "../../components/ui/Button";
import { DropZone } from "../../components/ui/DropZone";
import { StateMessage } from "../../components/ui/QueryState";
import { api, type TakeoutStatus } from "../../lib/api";
import { api, type TakeoutStatus, type TakeoutUpload } from "../../lib/api";
import type { TAKEOUT_OUTCOMES } from "../../lib/api/feed";
import { qk } from "../../lib/queryKeys";
import { notify } from "../../lib/notifications";
import { notifyActionError } from "../../lib/actionErrors";
import { formatDateTime, formatNumber } from "../../lib/format";
import { formatBytes, formatDateTime, formatNumber } from "../../lib/format";
import { importBusy, invalidateImported, watchTakeout } from "./takeoutWatch";
// Settings › Account: the user's Google Takeout export, imported. What they watched on YouTube and
// their YouTube Watch later list reach no API, so the export's two files are uploaded here, one
// request each; the server tells them apart and runs the import in the background, and its outcome
// also reaches the inbox. The import has no steps to count (the whole history is applied in one
// go), so while it runs the section says so and asks again until it is done.
const POLL_MS = 1500;
// go), so while it runs the section says so and asks again until it is done (`takeoutWatch`).
// The outcome counts, in the order they are said: the history's, then Watch later's. A count the
// import did not produce is not said at all.
@@ -35,6 +35,12 @@ const WATCH_LATER = [
["waiting", "laterWaiting"],
] as const;
// Every outcome the server can name is said by one of the two lists (feed.ts holds them to it).
type TakeoutOutcome = (typeof TAKEOUT_OUTCOMES)[number];
type Said = (typeof HISTORY)[number][0] | (typeof WATCH_LATER)[number][0];
const everyOutcomeSaid: [Exclude<TakeoutOutcome, Said>] extends [never] ? true : never = true;
void everyOutcomeSaid;
function countParts(s: TakeoutStatus, say: (key: string, n: number) => string): string[] {
const parts: string[] = [];
for (const [outcome, key] of HISTORY) {
@@ -52,58 +58,111 @@ export default function TakeoutImport() {
const { t, i18n } = useTranslation();
const lang = i18n.language;
const qc = useQueryClient();
const status = useQuery({
queryKey: qk.takeout(),
queryFn: api.takeoutStatus,
refetchInterval: (q) => (q.state.data?.run?.status === "running" ? POLL_MS : false),
});
const status = useQuery({ queryKey: qk.takeout(), queryFn: api.takeoutStatus });
const [uploading, setUploading] = useState<string | null>(null);
// "New import" reopens the drop zone over a finished one; before the first import it is open.
const [reopened, setReopened] = useState(false);
const run = status.data?.run ?? null;
const running = uploading !== null || run?.status === "running";
// Queued counts too: until its run row is written, `run` is still the import before it.
const busy = importBusy(status.data);
const running = uploading !== null || busy;
const finishedAt = run?.finished_at ?? null;
// A run that ended has changed what the feed and Watch later show: those lists are read again
// when the user next opens them, not 30 s later from the cache.
// A run under way is watched to its end, even if the user leaves Settings meanwhile.
useEffect(() => {
if (!finishedAt) return;
void qc.invalidateQueries({ queryKey: qk.feed() });
void qc.invalidateQueries({ queryKey: qk.playlist() });
if (busy) void watchTakeout(qc);
}, [busy, qc]);
// One that ended while nothing here watched it (a reload, another tab) has changed what the feed
// and Watch later show too: those lists are read again when next opened, not from the cache.
useEffect(() => {
if (finishedAt) invalidateImported(qc);
}, [finishedAt, qc]);
// A file dropped beside the zone, or while it is not there, is kept from the browser: it would
// open the file in the tab, unloading the app and the upload under way. The zone takes its own.
useEffect(() => {
const missed = (e: DragEvent) => {
if (e.defaultPrevented || !e.dataTransfer?.types.includes("Files")) return;
e.preventDefault();
if (e.type === "dragover") e.dataTransfer.dropEffect = "none";
};
window.addEventListener("dragover", missed);
window.addEventListener("drop", missed);
return () => {
window.removeEventListener("dragover", missed);
window.removeEventListener("drop", missed);
};
}, []);
// Where the focus goes when what holds it leaves: "New import" hands it to the zone it opens, and
// the zone, gone once the import it started is done, hands it to "New import".
const zone = useRef<HTMLButtonElement | null>(null);
const focusNext = useRef<"zone" | "newImport" | null>(null);
const focusIfNext = (which: "zone" | "newImport", el: HTMLButtonElement | null) => {
if (!el || focusNext.current !== which) return;
focusNext.current = null;
// Only focus that was lost with the control that held it: never taken from where the user went.
if (document.activeElement === null || document.activeElement === document.body) el.focus();
};
const zoneRef = (el: HTMLButtonElement | null) => {
zone.current = el;
focusIfNext("zone", el);
};
const newImportRef = (el: HTMLButtonElement | null) => focusIfNext("newImport", el);
const upload = async (files: File[]) => {
for (const file of files) await uploadOne(file);
if (document.activeElement === zone.current) focusNext.current = "newImport";
let taken = false;
for (const file of files) if (await uploadOne(file)) taken = true;
// The new run read first, then the upload's line let go: the other way round the previous
// import's line showed for a moment, as if this upload had done nothing.
await qc.invalidateQueries({ queryKey: qk.takeout() });
setUploading(null);
setReopened(false);
// Nothing taken, nothing imported: the zone stays open for the right file.
if (taken) setReopened(false);
else focusNext.current = null;
};
const uploadOne = async (file: File) => {
setUploading(file.name);
try {
const res = await api.takeoutUpload(file);
/** Send one file; whether the server took it. */
const uploadOne = async (file: File): Promise<boolean> => {
const max = status.data?.max_bytes ?? Infinity;
if (file.size > max) {
// Refused here, not after minutes of upload, and with the limit named.
notify({
level: "success",
message: t("settings.takeout.read", {
name: file.name,
count: res.videos,
n: formatNumber(res.videos, lang),
}),
level: "error",
message: t("settings.takeout.tooLarge", { name: file.name, limit: formatBytes(max, lang) }),
});
return false;
}
setUploading(file.name);
let res: TakeoutUpload | null = null;
try {
res = await api.takeoutUpload(file);
} catch (e) {
// A refusal names its reason (not a Takeout file, nothing in it); the next file still goes.
notifyActionError(e, t("errors.writeFailed.takeoutUpload"));
}
if (res === null) return false;
notify({
level: "success",
message: t("settings.takeout.read", {
name: file.name,
count: res.videos,
n: formatNumber(res.videos, lang),
}),
});
return true;
};
const say = (key: string, n: number) =>
t(`settings.takeout.count.${key}`, { count: n, n: formatNumber(n, lang) });
const when = (iso: string | null) => (iso ? formatDateTime(iso, lang) : "");
// Not while the last import is still being asked for: a zone that flashes before the line that
// replaces it would offer a first import to someone who has done one.
const showZone = status.isSuccess && !running && (run === null || reopened);
// replaces it would offer a first import to someone who has done one. While an import runs the
// zone stays, unavailable: the control that started it keeps the focus.
const showZone = status.isSuccess && (running || run === null || reopened);
// A failed re-read keeps the previous answer: it is not shown under the line saying the read
// failed, as if it were current — a run's spinner no more than its outcome.
const current = !status.isError || uploading !== null;
return (
<Section title={t("settings.takeout.title")}>
@@ -117,7 +176,14 @@ export default function TakeoutImport() {
{showZone && (
<>
<DropZone className="px-4 py-5" multiple accept=".json,.csv" onFiles={upload}>
<DropZone
ref={zoneRef}
className="px-4 py-5"
multiple
accept=".json,.csv"
disabled={running}
onFiles={upload}
>
{(dragOver) => (
<>
<UploadCloud className="w-6 h-6 text-muted" />
@@ -138,42 +204,49 @@ export default function TakeoutImport() {
</>
)}
{running ? (
<div
className="mt-2 flex items-center gap-2 text-xs text-muted"
role="status"
data-testid="takeout-running"
>
<RefreshCw className="w-3.5 h-3.5 animate-spin" />
{uploading !== null
? t("settings.takeout.uploading", { name: uploading })
: t("settings.takeout.running")}
<div className="mt-2 flex items-center justify-between gap-3">
{/* One live region for the run and its outcome, mounted throughout: what replaces
"Processing…" is announced, which a region mounted with its text is not. */}
<div role="status" className="min-w-0 text-xs text-muted">
{running && current ? (
<span className="flex items-center gap-2" data-testid="takeout-running">
<RefreshCw className="w-3.5 h-3.5 animate-spin" />
{uploading !== null
? t("settings.takeout.uploading", { name: uploading })
: t("settings.takeout.running")}
</span>
) : (
!running &&
run &&
status.data &&
current && (
<p data-testid="takeout-last">
{run.status === "error"
? t("settings.takeout.failed", { when: when(run.finished_at) })
: run.status === "interrupted"
? t("settings.takeout.interrupted", { when: when(run.started_at) })
: [
t("settings.takeout.last", { when: when(run.finished_at) }),
...countParts(status.data, say),
].join(" · ")}
</p>
)
)}
</div>
) : (
// A failed re-read keeps the previous answer: it is not shown under the line saying the
// read failed, as if it were current.
run &&
status.data &&
!status.isError && (
<div className="mt-2 flex items-center justify-between gap-3">
<p className="text-xs text-muted" data-testid="takeout-last">
{run.status === "error"
? t("settings.takeout.failed", { when: when(run.finished_at) })
: run.status === "interrupted"
? t("settings.takeout.interrupted", { when: when(run.started_at) })
: [
t("settings.takeout.last", { when: when(run.finished_at) }),
...countParts(status.data, say),
].join(" · ")}
</p>
{!reopened && (
<Button variant="outline" size="sm" onClick={() => setReopened(true)}>
{t("settings.takeout.newImport")}
</Button>
)}
</div>
)
)}
{!running && run && status.data && current && !reopened && (
<Button
ref={newImportRef}
variant="outline"
size="sm"
onClick={() => {
focusNext.current = "zone";
setReopened(true);
}}
>
{t("settings.takeout.newImport")}
</Button>
)}
</div>
</Section>
);
}