@crawlee/basic 3.17.1-beta.59 → 3.17.1-beta.60
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
|
@@ -436,7 +436,7 @@ export declare class BasicCrawler<Context extends CrawlingContext = BasicCrawlin
|
|
|
436
436
|
userAgent?: string;
|
|
437
437
|
};
|
|
438
438
|
protected onSkippedRequest?: SkippedRequestCallback;
|
|
439
|
-
private
|
|
439
|
+
private _ownsEventManager;
|
|
440
440
|
private loggedPerRun;
|
|
441
441
|
private experiments;
|
|
442
442
|
private readonly robotsTxtFileCache;
|
|
@@ -336,11 +336,11 @@ class BasicCrawler {
|
|
|
336
336
|
writable: true,
|
|
337
337
|
value: void 0
|
|
338
338
|
});
|
|
339
|
-
Object.defineProperty(this, "
|
|
339
|
+
Object.defineProperty(this, "_ownsEventManager", {
|
|
340
340
|
enumerable: true,
|
|
341
341
|
configurable: true,
|
|
342
342
|
writable: true,
|
|
343
|
-
value:
|
|
343
|
+
value: false
|
|
344
344
|
});
|
|
345
345
|
Object.defineProperty(this, "loggedPerRun", {
|
|
346
346
|
enumerable: true,
|
|
@@ -655,8 +655,8 @@ class BasicCrawler {
|
|
|
655
655
|
await this.autoscaledPool.run();
|
|
656
656
|
}
|
|
657
657
|
finally {
|
|
658
|
-
await this.teardown();
|
|
659
658
|
await this.stats.stopCapturing();
|
|
659
|
+
await this.teardown();
|
|
660
660
|
process.off('SIGINT', sigintHandler);
|
|
661
661
|
this.events.off("migrating" /* EventType.MIGRATING */, boundPauseOnMigration);
|
|
662
662
|
this.events.off("aborting" /* EventType.ABORTING */, boundPauseOnMigration);
|
|
@@ -932,7 +932,7 @@ class BasicCrawler {
|
|
|
932
932
|
async _init() {
|
|
933
933
|
if (!this.events.isInitialized()) {
|
|
934
934
|
await this.events.init();
|
|
935
|
-
this.
|
|
935
|
+
this._ownsEventManager = true;
|
|
936
936
|
}
|
|
937
937
|
this.autoscaledPool = new core_1.AutoscaledPool(this.autoscaledPoolOptions, this.config);
|
|
938
938
|
if (this.useSessionPool) {
|
|
@@ -1406,11 +1406,16 @@ class BasicCrawler {
|
|
|
1406
1406
|
* To stop the crawler gracefully (waiting for all running requests to finish), use {@link BasicCrawler.stop|`crawler.stop()`} instead.
|
|
1407
1407
|
*/
|
|
1408
1408
|
async teardown() {
|
|
1409
|
-
this
|
|
1409
|
+
// When this crawler initialized the event manager, its close() call emits
|
|
1410
|
+
// the final persistence event after the crawler-specific state has been
|
|
1411
|
+
// saved. External event managers still need an explicit event here.
|
|
1412
|
+
if (!this._ownsEventManager) {
|
|
1413
|
+
this.events.emit("persistState" /* EventType.PERSIST_STATE */, { isMigrating: false });
|
|
1414
|
+
}
|
|
1410
1415
|
if (this.useSessionPool) {
|
|
1411
|
-
await this.sessionPool.teardown();
|
|
1416
|
+
await this.sessionPool.teardown({ persistState: this._ownsEventManager });
|
|
1412
1417
|
}
|
|
1413
|
-
if (this.
|
|
1418
|
+
if (this._ownsEventManager) {
|
|
1414
1419
|
await this.events.close();
|
|
1415
1420
|
}
|
|
1416
1421
|
await this.autoscaledPool?.abort();
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@crawlee/basic",
|
|
3
|
-
"version": "3.17.1-beta.
|
|
3
|
+
"version": "3.17.1-beta.60",
|
|
4
4
|
"description": "The scalable web crawling and scraping library for JavaScript/Node.js. Enables development of data extraction and web automation jobs (not only) with headless Chrome and Puppeteer.",
|
|
5
5
|
"engines": {
|
|
6
6
|
"node": ">=16.0.0"
|
|
@@ -49,9 +49,9 @@
|
|
|
49
49
|
"@apify/log": "^2.4.0",
|
|
50
50
|
"@apify/timeout": "^0.4.0",
|
|
51
51
|
"@apify/utilities": "^2.7.10",
|
|
52
|
-
"@crawlee/core": "3.17.1-beta.
|
|
53
|
-
"@crawlee/types": "3.17.1-beta.
|
|
54
|
-
"@crawlee/utils": "3.17.1-beta.
|
|
52
|
+
"@crawlee/core": "3.17.1-beta.60",
|
|
53
|
+
"@crawlee/types": "3.17.1-beta.60",
|
|
54
|
+
"@crawlee/utils": "3.17.1-beta.60",
|
|
55
55
|
"csv-stringify": "^6.2.0",
|
|
56
56
|
"fs-extra": "^11.0.0",
|
|
57
57
|
"got-scraping": "^4.2.1",
|
|
@@ -67,5 +67,5 @@
|
|
|
67
67
|
}
|
|
68
68
|
}
|
|
69
69
|
},
|
|
70
|
-
"gitHead": "
|
|
70
|
+
"gitHead": "5ba2517b5f3bc6ef868b860c6c783a3e7e7d4a93"
|
|
71
71
|
}
|