@crawlee/basic 4.0.0-beta.236 → 4.0.0-beta.237
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.
- package/internals/basic-crawler.js +12 -12
- package/internals/crawler-run.js +2 -0
- package/package.json +7 -7
|
@@ -795,18 +795,6 @@ export class BasicCrawler {
|
|
|
795
795
|
if (this.running) {
|
|
796
796
|
throw new Error('This crawler instance is already running, you can add more requests to it via `crawler.addRequests()`.');
|
|
797
797
|
}
|
|
798
|
-
if (this.#hasFinishedBefore) {
|
|
799
|
-
// A supplied statistics instance keeps whatever state it was handed - only wipe a default we built.
|
|
800
|
-
await this.#statisticsDep.ifOwned(async (stats) => {
|
|
801
|
-
stats.reset();
|
|
802
|
-
await stats.resetStore();
|
|
803
|
-
});
|
|
804
|
-
// A session pool the crawler built follows the same blank-slate rule as its statistics.
|
|
805
|
-
await this.#sessionPoolDep.ifOwned(async (pool) => {
|
|
806
|
-
pool.reset();
|
|
807
|
-
await pool.resetStore();
|
|
808
|
-
});
|
|
809
|
-
}
|
|
810
798
|
// An owned governor is rebuilt for every run, so it always starts from a clean slate — stale
|
|
811
799
|
// resource snapshots or a previous run's scaled desired concurrency would otherwise distort this
|
|
812
800
|
// run's scaling. An injected one is long-lived and its lifecycle belongs to the caller.
|
|
@@ -819,6 +807,18 @@ export class BasicCrawler {
|
|
|
819
807
|
});
|
|
820
808
|
this.#run = run;
|
|
821
809
|
try {
|
|
810
|
+
if (this.#hasFinishedBefore) {
|
|
811
|
+
// A supplied statistics instance keeps whatever state it was handed - only wipe a default we built.
|
|
812
|
+
await this.#statisticsDep.ifOwned(async (stats) => {
|
|
813
|
+
stats.reset();
|
|
814
|
+
await stats.resetStore();
|
|
815
|
+
});
|
|
816
|
+
// A session pool the crawler built follows the same blank-slate rule as its statistics.
|
|
817
|
+
await this.#sessionPoolDep.ifOwned(async (pool) => {
|
|
818
|
+
pool.reset();
|
|
819
|
+
await pool.resetStore();
|
|
820
|
+
});
|
|
821
|
+
}
|
|
822
822
|
await purgeDefaultStorages({
|
|
823
823
|
onlyPurgeOnce: true,
|
|
824
824
|
storageBackend: serviceLocator.getStorageBackend(),
|
package/internals/crawler-run.js
CHANGED
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@crawlee/basic",
|
|
3
|
-
"version": "4.0.0-beta.
|
|
3
|
+
"version": "4.0.0-beta.237",
|
|
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": ">=22.13.0"
|
|
@@ -42,10 +42,10 @@
|
|
|
42
42
|
"@apify/datastructures": "^3.0.1",
|
|
43
43
|
"@apify/timeout": "^1.0.1",
|
|
44
44
|
"@apify/utilities": "^3.0.1",
|
|
45
|
-
"@crawlee/core": "4.0.0-beta.
|
|
46
|
-
"@crawlee/http-client": "4.0.0-beta.
|
|
47
|
-
"@crawlee/types": "4.0.0-beta.
|
|
48
|
-
"@crawlee/utils": "4.0.0-beta.
|
|
45
|
+
"@crawlee/core": "4.0.0-beta.237",
|
|
46
|
+
"@crawlee/http-client": "4.0.0-beta.237",
|
|
47
|
+
"@crawlee/types": "4.0.0-beta.237",
|
|
48
|
+
"@crawlee/utils": "4.0.0-beta.237",
|
|
49
49
|
"@sapphire/async-queue": "^1.5.5",
|
|
50
50
|
"@standard-schema/spec": "^1.0.0",
|
|
51
51
|
"csv-stringify": "^6.5.2",
|
|
@@ -57,7 +57,7 @@
|
|
|
57
57
|
"zod": "^4.5.4"
|
|
58
58
|
},
|
|
59
59
|
"optionalDependencies": {
|
|
60
|
-
"@crawlee/impit-client": "4.0.0-beta.
|
|
60
|
+
"@crawlee/impit-client": "4.0.0-beta.237"
|
|
61
61
|
},
|
|
62
62
|
"lerna": {
|
|
63
63
|
"command": {
|
|
@@ -66,5 +66,5 @@
|
|
|
66
66
|
}
|
|
67
67
|
}
|
|
68
68
|
},
|
|
69
|
-
"gitHead": "
|
|
69
|
+
"gitHead": "e661302cd88d98ea3c54721440cd5b23ec876102"
|
|
70
70
|
}
|