Files
dragonsstash/worker/src/tdlib/download.ts
2026-03-25 22:23:47 +01:00

567 lines
16 KiB
TypeScript

import type { Client } from "tdl";
import { readFile, rename, copyFile, unlink, stat } from "fs/promises";
import { config } from "../util/config.js";
import { childLogger } from "../util/logger.js";
import { withFloodWait } from "../util/retry.js";
import { isArchiveAttachment } from "../archive/detect.js";
import type { TelegramMessage } from "../archive/multipart.js";
import type { TelegramPhoto } from "../preview/match.js";
const log = childLogger("download");
/** Maximum number of pages to scan per channel/topic to prevent infinite loops */
export const MAX_SCAN_PAGES = 5000;
/** Timeout for a single TDLib API call (ms) */
export const INVOKE_TIMEOUT_MS = 120_000; // 2 minutes
interface TdPhotoSize {
type: string;
photo: {
id: number;
size: number;
expected_size: number;
local?: {
path?: string;
is_downloading_active?: boolean;
is_downloading_completed?: boolean;
downloaded_size?: number;
};
};
width: number;
height: number;
}
interface TdMessage {
id: number;
date: number;
media_album_id?: string;
content: {
_: string;
document?: {
file_name?: string;
document?: {
id: number;
size: number;
local?: {
path?: string;
is_downloading_completed?: boolean;
};
};
};
photo?: {
sizes?: TdPhotoSize[];
};
caption?: {
text?: string;
};
};
}
interface TdFile {
id: number;
size: number;
expected_size: number;
local: {
path: string;
is_downloading_active: boolean;
is_downloading_completed: boolean;
downloaded_size: number;
download_offset: number;
};
}
export interface ChannelScanResult {
archives: TelegramMessage[];
photos: TelegramPhoto[];
totalScanned: number;
}
export type ScanProgressCallback = (messagesScanned: number) => void;
/**
* Invoke a TDLib method with a timeout to prevent indefinite hangs,
* and automatic retry on FLOOD_WAIT rate-limit errors.
*
* If TDLib does not respond within the timeout, the promise rejects.
* If Telegram returns a rate limit error, sleeps for the required
* duration and retries (up to maxRetries times).
*/
export async function invokeWithTimeout<T>(
client: Client,
// eslint-disable-next-line @typescript-eslint/no-explicit-any
request: Record<string, any>,
timeoutMs = INVOKE_TIMEOUT_MS
): Promise<T> {
return withFloodWait(
() =>
new Promise<T>((resolve, reject) => {
let settled = false;
const timer = setTimeout(() => {
if (!settled) {
settled = true;
reject(
new Error(
`TDLib invoke timed out after ${timeoutMs}ms for ${request._}`
)
);
}
}, timeoutMs);
(client.invoke(request) as Promise<T>)
.then((result) => {
if (!settled) {
settled = true;
clearTimeout(timer);
resolve(result);
}
})
.catch((err) => {
if (!settled) {
settled = true;
clearTimeout(timer);
reject(err);
}
});
}),
`TDLib:${request._}`
);
}
/**
* Fetch messages from a channel, stopping once we've scanned past the
* last-processed boundary (with one page of lookback for multipart safety).
* Collects both archive attachments AND photo messages (for preview matching).
* Returns messages in chronological order (oldest first).
*
* When `lastProcessedMessageId` is null (first run), scans everything.
* The worker applies a post-grouping filter to skip fully-processed sets,
* and keeps `packageExistsBySourceMessage` as a safety net.
*
* Safety features:
* - Max page limit to prevent infinite loops
* - Stuck detection: breaks if from_message_id stops advancing
* - Timeout on each TDLib API call
*/
export async function getChannelMessages(
client: Client,
chatId: bigint,
lastProcessedMessageId?: bigint | null,
limit = 100,
onProgress?: ScanProgressCallback
): Promise<ChannelScanResult> {
const archives: TelegramMessage[] = [];
const photos: TelegramPhoto[] = [];
const boundary = lastProcessedMessageId ? Number(lastProcessedMessageId) : null;
// Open the chat so TDLib can access it
try {
await invokeWithTimeout(client, { _: "openChat", chat_id: Number(chatId) });
} catch {
// Ignore — may already be open
}
let totalScanned = 0;
let pageCount = 0;
// Use searchChatMessages with document filter — this works even when
// getChatHistory is restricted (e.g. hidden history for new members).
// We search for documents first, then photos separately.
for (const filter of [
{ _: "searchMessagesFilterDocument" as const, kind: "document" },
{ _: "searchMessagesFilterPhoto" as const, kind: "photo" },
]) {
let fromMessageId = 0;
// eslint-disable-next-line no-constant-condition
while (true) {
if (pageCount >= MAX_SCAN_PAGES) {
log.warn(
{ chatId: chatId.toString(), pageCount, totalScanned },
"Hit max page limit for channel scan, stopping"
);
break;
}
pageCount++;
const result = await invokeWithTimeout<{ messages: TdMessage[]; total_count?: number }>(client, {
_: "searchChatMessages",
chat_id: Number(chatId),
query: "",
from_message_id: fromMessageId,
offset: 0,
limit: Math.min(limit, 100),
filter,
message_thread_id: 0,
});
if (!result.messages || result.messages.length === 0) break;
totalScanned += result.messages.length;
for (const msg of result.messages) {
// Check for archive documents
const doc = msg.content?.document;
if (doc?.file_name && doc.document && isArchiveAttachment(doc.file_name)) {
// Skip if we've already processed past this message
if (boundary && msg.id <= boundary) continue;
archives.push({
id: BigInt(msg.id),
fileName: doc.file_name,
fileId: String(doc.document.id),
fileSize: BigInt(doc.document.size),
date: new Date(msg.date * 1000),
mediaAlbumId: msg.media_album_id && msg.media_album_id !== "0" ? msg.media_album_id : undefined,
});
continue;
}
// Check for photo messages (potential previews)
const photo = msg.content?.photo;
const caption = msg.content?.caption?.text ?? "";
if (photo?.sizes && photo.sizes.length > 0) {
if (boundary && msg.id <= boundary) continue;
const smallest = photo.sizes[0];
photos.push({
id: BigInt(msg.id),
date: new Date(msg.date * 1000),
caption,
fileId: String(smallest.photo.id),
fileSize: smallest.photo.size || smallest.photo.expected_size,
mediaAlbumId: msg.media_album_id && msg.media_album_id !== "0" ? msg.media_album_id : undefined,
});
}
}
onProgress?.(totalScanned);
// Advance pagination
fromMessageId = result.messages[result.messages.length - 1].id;
if (result.messages.length < Math.min(limit, 100)) break;
await sleep(config.apiDelayMs);
}
}
// Close the chat after scanning
await invokeWithTimeout(client, {
_: "closeChat",
chat_id: Number(chatId),
}).catch(() => {});
log.info(
{ chatId: chatId.toString(), archives: archives.length, photos: photos.length, totalScanned, pages: pageCount },
"Channel scan complete"
);
// Reverse to chronological order (oldest first) so worker processes old→new
return {
archives: archives.reverse(),
photos: photos.reverse(),
totalScanned,
};
}
/**
* Download a photo thumbnail from Telegram and return its raw bytes.
* Uses synchronous download (photos are small, typically < 100KB).
* Returns null if download fails (non-critical).
*/
export async function downloadPhotoThumbnail(
client: Client,
fileId: string
): Promise<Buffer | null> {
const numericId = parseInt(fileId, 10);
try {
const result = (await client.invoke({
_: "downloadFile",
file_id: numericId,
priority: 1, // Low priority — thumbnails are nice-to-have
offset: 0,
limit: 0,
synchronous: true, // Small file — wait for it
})) as TdFile;
if (result?.local?.is_downloading_completed && result.local.path) {
const data = await readFile(result.local.path);
log.debug(
{ fileId, bytes: data.length },
"Downloaded photo thumbnail"
);
return data;
}
} catch (err) {
log.warn({ fileId, err }, "Failed to download photo thumbnail");
}
return null;
}
export interface DownloadProgress {
fileId: string;
fileName: string;
downloadedBytes: number;
totalBytes: number;
percent: number;
isComplete: boolean;
}
export type ProgressCallback = (progress: DownloadProgress) => void;
/**
* Download a file from Telegram to a local path with progress tracking
* and integrity verification.
*
* Progress flow:
* 1. Starts async download via TDLib
* 2. Listens for `updateFile` events to track download progress
* 3. Logs progress at every 10% increment
* 4. Once complete, verifies the local file size matches the expected size
* 5. Moves the file from TDLib's cache to the destination path
*
* Verification:
* - Compares actual file size on disk to the expected size from Telegram
* - Throws on mismatch (partial/corrupt download)
* - Throws on timeout (configurable, scales with file size)
* - Throws if download stops without completing (network error, etc.)
*/
export async function downloadFile(
client: Client,
fileId: string,
destPath: string,
expectedSize: bigint,
fileName: string,
onProgress?: ProgressCallback
): Promise<void> {
const numericId = parseInt(fileId, 10);
const totalBytes = Number(expectedSize);
log.info(
{ fileId, fileName, destPath, totalBytes },
"Starting file download"
);
// Report initial progress
onProgress?.({
fileId,
fileName,
downloadedBytes: 0,
totalBytes,
percent: 0,
isComplete: false,
});
return new Promise<void>((resolve, reject) => {
let lastLoggedPercent = 0;
let settled = false;
let downloadStarted = false; // True once TDLib reports is_downloading_active
let lastProgressBytes = 0;
let lastProgressTime = Date.now();
// Timeout: 20 minutes per GB, minimum 15 minutes
const timeoutMs = Math.max(
15 * 60_000,
(totalBytes / (1024 * 1024 * 1024)) * 20 * 60_000
);
const timer = setTimeout(() => {
if (!settled) {
settled = true;
cleanup();
reject(
new Error(
`Download timed out after ${Math.round(timeoutMs / 60_000)}min for ${fileName}`
)
);
}
}, timeoutMs);
// Stall detection: no progress for 5 minutes after download started → reject
const STALL_TIMEOUT_MS = 5 * 60_000;
const stallChecker = setInterval(() => {
if (settled || !downloadStarted) return;
const stallMs = Date.now() - lastProgressTime;
if (stallMs >= STALL_TIMEOUT_MS) {
settled = true;
cleanup();
reject(
new Error(
`Download stalled for ${fileName} — no progress for ${Math.round(stallMs / 60_000)}min ` +
`(${lastProgressBytes}/${totalBytes} bytes)`
)
);
}
}, 30_000);
// Listen for file update events to track progress
// eslint-disable-next-line @typescript-eslint/no-explicit-any
const handleUpdate = (update: any) => {
if (update?._ !== "updateFile") return;
const file = update.file as TdFile | undefined;
if (!file || file.id !== numericId) return;
const downloaded = file.local.downloaded_size;
const percent =
totalBytes > 0 ? Math.round((downloaded / totalBytes) * 100) : 0;
// Track whether the download has actually started
if (file.local.is_downloading_active) {
downloadStarted = true;
}
// Reset stall timer when bytes advance
if (downloaded > lastProgressBytes) {
lastProgressBytes = downloaded;
lastProgressTime = Date.now();
}
// Log at every 10% increment
if (percent >= lastLoggedPercent + 10) {
lastLoggedPercent = percent - (percent % 10);
log.info(
{ fileId, fileName, downloaded, totalBytes, percent: `${percent}%` },
"Download progress"
);
}
// Report to callback
onProgress?.({
fileId,
fileName,
downloadedBytes: downloaded,
totalBytes,
percent,
isComplete: file.local.is_downloading_completed,
});
// Download finished
if (file.local.is_downloading_completed) {
if (!settled) {
settled = true;
cleanup();
verifyAndMove(file.local.path, destPath, totalBytes, fileName, fileId)
.then(resolve)
.catch(reject);
}
}
// Download stopped without completing — only if it had actually started.
// TDLib may emit an initial updateFile with is_downloading_active=false
// before the download begins; ignoring that prevents false positives.
if (
downloadStarted &&
!file.local.is_downloading_active &&
!file.local.is_downloading_completed
) {
if (!settled) {
settled = true;
cleanup();
reject(
new Error(
`Download stopped unexpectedly for ${fileName} ` +
`(${downloaded}/${totalBytes} bytes, ${percent}%)`
)
);
}
}
};
const cleanup = () => {
clearTimeout(timer);
clearInterval(stallChecker);
client.off("update", handleUpdate);
};
// Subscribe to updates BEFORE starting download
client.on("update", handleUpdate);
// Start async download (non-blocking — progress via updateFile events)
// Wrapped in withFloodWait: if the initial invoke is rate-limited,
// it will sleep and retry before the download event loop begins.
withFloodWait(
() =>
client.invoke({
_: "downloadFile",
file_id: numericId,
priority: 32,
offset: 0,
limit: 0,
synchronous: false,
}),
`downloadFile:${fileName}`
)
.then((result: unknown) => {
// If the file was already cached locally, invoke returns immediately
const file = result as TdFile | undefined;
if (file?.local?.is_downloading_completed && !settled) {
settled = true;
cleanup();
verifyAndMove(file.local.path, destPath, totalBytes, fileName, fileId)
.then(resolve)
.catch(reject);
}
})
.catch((err: unknown) => {
if (!settled) {
settled = true;
cleanup();
reject(err);
}
});
});
}
/**
* Verify the downloaded file's size matches the expected size,
* then move it to the destination path.
*/
async function verifyAndMove(
localPath: string,
destPath: string,
expectedBytes: number,
fileName: string,
fileId: string
): Promise<void> {
const stats = await stat(localPath);
const actualBytes = stats.size;
if (expectedBytes > 0 && actualBytes !== expectedBytes) {
log.error(
{ fileId, fileName, expectedBytes, actualBytes },
"Download size mismatch — file is incomplete or corrupted"
);
throw new Error(
`Download verification failed for ${fileName}: ` +
`expected ${expectedBytes} bytes, got ${actualBytes} bytes`
);
}
log.info(
{ fileId, fileName, bytes: actualBytes, destPath },
"File verified and complete"
);
// Move from TDLib's cache to our temp directory.
// Use rename first (fast, same filesystem), fall back to copy+delete
// when source and destination are on different filesystems (EXDEV).
try {
await rename(localPath, destPath);
} catch (err: unknown) {
if ((err as NodeJS.ErrnoException).code === "EXDEV") {
log.debug(
{ fileId, fileName },
"Cross-device rename — falling back to copy + unlink"
);
await copyFile(localPath, destPath);
await unlink(localPath);
} else {
throw err;
}
}
}
function sleep(ms: number): Promise<void> {
return new Promise((resolve) => setTimeout(resolve, ms));
}