mirror of
https://git.mirrors.martin98.com/https://github.com/jina-ai/reader
synced 2025-07-31 09:21:59 +08:00
fix: saas issus
This commit is contained in:
parent
75a1283a91
commit
e6fdd87294
@ -6,7 +6,7 @@ import { marshalErrorLike } from 'civkit/lang';
|
|||||||
import { objHashMd5B64Of } from 'civkit/hash';
|
import { objHashMd5B64Of } from 'civkit/hash';
|
||||||
import _ from 'lodash';
|
import _ from 'lodash';
|
||||||
|
|
||||||
import { RateLimitControl, RateLimitDesc } from '../shared/services/rate-limit';
|
import { RateLimitControl, RateLimitDesc, RateLimitTriggeredError } from '../shared/services/rate-limit';
|
||||||
|
|
||||||
import { CrawlerHost, ExtraScrappingOptions } from './crawler';
|
import { CrawlerHost, ExtraScrappingOptions } from './crawler';
|
||||||
import { SerperSearchResult } from '../db/searched';
|
import { SerperSearchResult } from '../db/searched';
|
||||||
@ -19,8 +19,16 @@ import { AsyncLocalContext } from '../services/async-context';
|
|||||||
import { Context, Ctx, Method, Param, RPCReflect } from '../services/registry';
|
import { Context, Ctx, Method, Param, RPCReflect } from '../services/registry';
|
||||||
import { OutputServerEventStream } from '../lib/transform-server-event-stream';
|
import { OutputServerEventStream } from '../lib/transform-server-event-stream';
|
||||||
import { JinaEmbeddingsAuthDTO } from '../dto/jina-embeddings-auth';
|
import { JinaEmbeddingsAuthDTO } from '../dto/jina-embeddings-auth';
|
||||||
import { InsufficientBalanceError, RateLimitTriggeredError } from '../services/errors';
|
import { InsufficientBalanceError } from '../services/errors';
|
||||||
import { SerperImageSearchResponse, SerperNewsSearchResponse, SerperSearchQueryParams, SerperSearchResponse, SerperWebSearchResponse, WORLD_COUNTRIES, WORLD_LANGUAGES } from '../shared/3rd-party/serper-search';
|
import {
|
||||||
|
SerperImageSearchResponse,
|
||||||
|
SerperNewsSearchResponse,
|
||||||
|
SerperSearchQueryParams,
|
||||||
|
SerperSearchResponse,
|
||||||
|
SerperWebSearchResponse,
|
||||||
|
WORLD_COUNTRIES,
|
||||||
|
WORLD_LANGUAGES
|
||||||
|
} from '../shared/3rd-party/serper-search';
|
||||||
import { toAsyncGenerator } from '../utils/misc';
|
import { toAsyncGenerator } from '../utils/misc';
|
||||||
import type { JinaEmbeddingsTokenAccount } from '../shared/db/jina-embeddings-token-account';
|
import type { JinaEmbeddingsTokenAccount } from '../shared/db/jina-embeddings-token-account';
|
||||||
import { LRUCache } from 'lru-cache';
|
import { LRUCache } from 'lru-cache';
|
||||||
@ -218,10 +226,10 @@ export class SearcherHost extends RPCHost {
|
|||||||
}
|
}
|
||||||
const now = Date.now();
|
const now = Date.now();
|
||||||
let tgtDate;
|
let tgtDate;
|
||||||
if (err.retryAfter) {
|
if (err.retryAfterDate) {
|
||||||
tgtDate = new Date(now + err.retryAfter * 1000);
|
|
||||||
} else if (err.retryAfterDate) {
|
|
||||||
tgtDate = err.retryAfterDate;
|
tgtDate = err.retryAfterDate;
|
||||||
|
} else if (err.retryAfter) {
|
||||||
|
tgtDate = new Date(now + err.retryAfter * 1000);
|
||||||
}
|
}
|
||||||
|
|
||||||
if (tgtDate) {
|
if (tgtDate) {
|
||||||
|
@ -7,14 +7,14 @@ import {
|
|||||||
import { marshalErrorLike } from 'civkit/lang';
|
import { marshalErrorLike } from 'civkit/lang';
|
||||||
import _ from 'lodash';
|
import _ from 'lodash';
|
||||||
|
|
||||||
import { RateLimitControl, RateLimitDesc } from '../shared/services/rate-limit';
|
import { RateLimitControl, RateLimitDesc, RateLimitTriggeredError } from '../shared/services/rate-limit';
|
||||||
|
|
||||||
import { GlobalLogger } from '../services/logger';
|
import { GlobalLogger } from '../services/logger';
|
||||||
import { AsyncLocalContext } from '../services/async-context';
|
import { AsyncLocalContext } from '../services/async-context';
|
||||||
import { Context, Ctx, Method, Param, RPCReflect } from '../services/registry';
|
import { Context, Ctx, Method, Param, RPCReflect } from '../services/registry';
|
||||||
import { OutputServerEventStream } from '../lib/transform-server-event-stream';
|
import { OutputServerEventStream } from '../lib/transform-server-event-stream';
|
||||||
import { JinaEmbeddingsAuthDTO } from '../dto/jina-embeddings-auth';
|
import { JinaEmbeddingsAuthDTO } from '../dto/jina-embeddings-auth';
|
||||||
import { InsufficientBalanceError, RateLimitTriggeredError } from '../services/errors';
|
import { InsufficientBalanceError } from '../services/errors';
|
||||||
import { WORLD_COUNTRIES, WORLD_LANGUAGES } from '../shared/3rd-party/serper-search';
|
import { WORLD_COUNTRIES, WORLD_LANGUAGES } from '../shared/3rd-party/serper-search';
|
||||||
import { GoogleSERP } from '../services/serp/google';
|
import { GoogleSERP } from '../services/serp/google';
|
||||||
import { WebSearchEntry } from '../services/serp/compat';
|
import { WebSearchEntry } from '../services/serp/compat';
|
||||||
@ -172,10 +172,11 @@ export class SerpHost extends RPCHost {
|
|||||||
const now = new Date();
|
const now = new Date();
|
||||||
const blockedTimeRemaining = (highFreqKey.blockedUntil.valueOf() - now.valueOf());
|
const blockedTimeRemaining = (highFreqKey.blockedUntil.valueOf() - now.valueOf());
|
||||||
if (blockedTimeRemaining > 0) {
|
if (blockedTimeRemaining > 0) {
|
||||||
throw RateLimitTriggeredError.from({
|
this.logger.warn(`Rate limit triggered for ${uid}, this request should have been blocked`);
|
||||||
message: `Per UID rate limit exceeded (async)`,
|
// throw RateLimitTriggeredError.from({
|
||||||
retryAfter: Math.ceil(blockedTimeRemaining / 1000),
|
// message: `Per UID rate limit exceeded (async)`,
|
||||||
});
|
// retryAfter: Math.ceil(blockedTimeRemaining / 1000),
|
||||||
|
// });
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -229,10 +230,10 @@ export class SerpHost extends RPCHost {
|
|||||||
}
|
}
|
||||||
const now = Date.now();
|
const now = Date.now();
|
||||||
let tgtDate;
|
let tgtDate;
|
||||||
if (err.retryAfter) {
|
if (err.retryAfterDate) {
|
||||||
tgtDate = new Date(now + err.retryAfter * 1000);
|
|
||||||
} else if (err.retryAfterDate) {
|
|
||||||
tgtDate = err.retryAfterDate;
|
tgtDate = err.retryAfterDate;
|
||||||
|
} else if (err.retryAfter) {
|
||||||
|
tgtDate = new Date(now + err.retryAfter * 1000);
|
||||||
}
|
}
|
||||||
|
|
||||||
if (tgtDate) {
|
if (tgtDate) {
|
||||||
|
@ -61,10 +61,23 @@ export class CurlControl extends AsyncService {
|
|||||||
}
|
}
|
||||||
|
|
||||||
curlImpersonateHeader(curl: Curl, headers?: object) {
|
curlImpersonateHeader(curl: Curl, headers?: object) {
|
||||||
|
let uaPlatform = this.platform;
|
||||||
|
if (this.ua.includes('Windows')) {
|
||||||
|
uaPlatform = 'Windows';
|
||||||
|
} else if (this.ua.includes('Android')) {
|
||||||
|
uaPlatform = 'Android';
|
||||||
|
} else if (this.ua.includes('iPhone') || this.ua.includes('iPad') || this.ua.includes('iPod')) {
|
||||||
|
uaPlatform = 'iOS';
|
||||||
|
} else if (this.ua.includes('CrOS')) {
|
||||||
|
uaPlatform = 'Chrome OS';
|
||||||
|
} else if (this.ua.includes('Macintosh')) {
|
||||||
|
uaPlatform = 'macOS';
|
||||||
|
}
|
||||||
|
|
||||||
const mixinHeaders: Record<string, string> = {
|
const mixinHeaders: Record<string, string> = {
|
||||||
'sch-ch-ua': `Not A(Brand";v="8", "Chromium";v="${this.chromeVersion}", "Google Chrome";v="${this.chromeVersion}"`,
|
'Sec-Ch-Ua': `Not A(Brand";v="8", "Chromium";v="${this.chromeVersion}", "Google Chrome";v="${this.chromeVersion}"`,
|
||||||
'sec-ch-ua-mobile': '?0',
|
'Sec-Ch-Ua-Mobile': '?0',
|
||||||
'sec-ch-ua-platform': this.platform,
|
'Sec-Ch-Ua-Platform': `"${uaPlatform}"`,
|
||||||
'Upgrade-Insecure-Requests': '1',
|
'Upgrade-Insecure-Requests': '1',
|
||||||
'User-Agent': this.ua,
|
'User-Agent': this.ua,
|
||||||
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7',
|
'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/avif,image/webp,image/apng,*/*;q=0.8,application/signed-exchange;v=b3;q=0.7',
|
||||||
|
@ -1,4 +1,4 @@
|
|||||||
import { ApplicationError, Prop, RPC_TRANSFER_PROTOCOL_META_SYMBOL, StatusCode } from 'civkit/civ-rpc';
|
import { ApplicationError, StatusCode } from 'civkit/civ-rpc';
|
||||||
import _ from 'lodash';
|
import _ from 'lodash';
|
||||||
import dayjs from 'dayjs';
|
import dayjs from 'dayjs';
|
||||||
import utc from 'dayjs/plugin/utc';
|
import utc from 'dayjs/plugin/utc';
|
||||||
@ -46,31 +46,3 @@ export class SecurityCompromiseError extends ApplicationError { }
|
|||||||
|
|
||||||
@StatusCode(41201)
|
@StatusCode(41201)
|
||||||
export class BatchSizeTooLargeError extends ApplicationError { }
|
export class BatchSizeTooLargeError extends ApplicationError { }
|
||||||
|
|
||||||
|
|
||||||
@StatusCode(42903)
|
|
||||||
export class RateLimitTriggeredError extends ApplicationError {
|
|
||||||
|
|
||||||
@Prop({
|
|
||||||
desc: 'Retry after seconds',
|
|
||||||
})
|
|
||||||
retryAfter?: number;
|
|
||||||
|
|
||||||
@Prop({
|
|
||||||
desc: 'Retry after date',
|
|
||||||
})
|
|
||||||
retryAfterDate?: Date;
|
|
||||||
|
|
||||||
protected override get [RPC_TRANSFER_PROTOCOL_META_SYMBOL]() {
|
|
||||||
const retryAfter = this.retryAfter || this.retryAfterDate;
|
|
||||||
if (!retryAfter) {
|
|
||||||
return super[RPC_TRANSFER_PROTOCOL_META_SYMBOL];
|
|
||||||
}
|
|
||||||
|
|
||||||
return _.merge(_.cloneDeep(super[RPC_TRANSFER_PROTOCOL_META_SYMBOL]), {
|
|
||||||
headers: {
|
|
||||||
'Retry-After': `${retryAfter instanceof Date ? dayjs(retryAfter).utc().format('ddd, DD MMM YYYY HH:mm:ss [GMT]') : retryAfter}`,
|
|
||||||
}
|
|
||||||
});
|
|
||||||
}
|
|
||||||
}
|
|
||||||
|
@ -56,6 +56,7 @@ export class SerperSearchService extends AsyncService {
|
|||||||
let maxTries = 3;
|
let maxTries = 3;
|
||||||
|
|
||||||
while (maxTries--) {
|
while (maxTries--) {
|
||||||
|
const t0 = Date.now();
|
||||||
try {
|
try {
|
||||||
this.logger.debug(`Doing external search`, query);
|
this.logger.debug(`Doing external search`, query);
|
||||||
let r;
|
let r;
|
||||||
@ -101,11 +102,14 @@ export class SerperSearchService extends AsyncService {
|
|||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
const dt = Date.now() - t0;
|
||||||
this.blackHoleDetector.itWorked();
|
this.blackHoleDetector.itWorked();
|
||||||
|
this.logger.debug(`External search took ${dt}ms`, { searchDt: dt, variant });
|
||||||
|
|
||||||
return r.parsed;
|
return r.parsed;
|
||||||
} catch (err: any) {
|
} catch (err: any) {
|
||||||
this.logger.error(`${variant} search failed: ${err?.message}`, { err: marshalErrorLike(err) });
|
const dt = Date.now() - t0;
|
||||||
|
this.logger.error(`${variant} search failed: ${err?.message}`, { searchDt: dt, err: marshalErrorLike(err) });
|
||||||
if (err?.status === 429) {
|
if (err?.status === 429) {
|
||||||
await delay(500 + 1000 * Math.random());
|
await delay(500 + 1000 * Math.random());
|
||||||
continue;
|
continue;
|
||||||
|
@ -1 +1 @@
|
|||||||
Subproject commit f30577e73eca151f9d62198ae3b60d0172befaed
|
Subproject commit 580ea72e0eddaa115b85dabf29de41d079ecd2d0
|
Loading…
x
Reference in New Issue
Block a user