@crawlee/core 4.0.0-beta.92 → 4.0.0-beta.93
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.
|
@@ -1,11 +1,10 @@
|
|
|
1
1
|
import type { ConcurrencyConsumer, IConcurrencySystem } from './concurrency_system.js';
|
|
2
2
|
import type { CrawleeLogger } from '../log.js';
|
|
3
3
|
/**
|
|
4
|
-
* The
|
|
5
|
-
*
|
|
6
|
-
* `runTaskFunction`.
|
|
4
|
+
* The two predicates that steer a task loop: *is there work ready?* and *are we done?* These are the parts of the loop
|
|
5
|
+
* a caller legitimately overrides, as opposed to the task itself (`runTaskFunction`), which the loop's driver owns.
|
|
7
6
|
*/
|
|
8
|
-
export interface
|
|
7
|
+
export interface TaskLoopPredicates {
|
|
9
8
|
/**
|
|
10
9
|
* A function that indicates whether `runTaskFunction` should be called.
|
|
11
10
|
* This function is called every time there is free capacity for a new task and it should
|
|
@@ -15,14 +14,14 @@ export interface AutoscaledPoolPredicateOptions {
|
|
|
15
14
|
isTaskReadyFunction?: () => Promise<boolean>;
|
|
16
15
|
/**
|
|
17
16
|
* A function that is called only when there are no tasks to be processed.
|
|
18
|
-
* If it resolves to `true` then the
|
|
17
|
+
* If it resolves to `true` then the run finishes. Being called only
|
|
19
18
|
* when there are no tasks being processed means that as long as `isTaskReadyFunction()`
|
|
20
19
|
* keeps resolving to `true`, `isFinishedFunction()` will never be called.
|
|
21
|
-
* To abort a run, use the {@link AutoscaledPool.abort} method.
|
|
22
20
|
*/
|
|
23
21
|
isFinishedFunction?: () => Promise<boolean>;
|
|
24
22
|
}
|
|
25
|
-
|
|
23
|
+
/** @internal */
|
|
24
|
+
export interface AutoscaledPoolOptions extends TaskLoopPredicates {
|
|
26
25
|
/**
|
|
27
26
|
* The governor that decides whether there is free compute for one more task. Typically a
|
|
28
27
|
* {@link ConcurrencySystem}, but any {@link IConcurrencySystem} works. Share a single instance across
|
|
@@ -62,8 +61,8 @@ export interface AutoscaledPoolOptions extends AutoscaledPoolPredicateOptions {
|
|
|
62
61
|
*
|
|
63
62
|
* Before running the pool, you need to implement the following three functions:
|
|
64
63
|
* {@link AutoscaledPoolOptions.runTaskFunction|`runTaskFunction`},
|
|
65
|
-
* {@link
|
|
66
|
-
* {@link
|
|
64
|
+
* {@link TaskLoopPredicates.isTaskReadyFunction|`isTaskReadyFunction`} and
|
|
65
|
+
* {@link TaskLoopPredicates.isFinishedFunction|`isFinishedFunction`}.
|
|
67
66
|
*
|
|
68
67
|
* The auto-scaled pool is started by calling the {@link AutoscaledPool.run} function.
|
|
69
68
|
* The pool periodically queries `isTaskReadyFunction` for more tasks, managing optimal concurrency, until the function
|
|
@@ -104,7 +103,8 @@ export interface AutoscaledPoolOptions extends AutoscaledPoolPredicateOptions {
|
|
|
104
103
|
* await concurrencySystem.stop();
|
|
105
104
|
* }
|
|
106
105
|
* ```
|
|
107
|
-
*
|
|
106
|
+
*
|
|
107
|
+
* @internal
|
|
108
108
|
*/
|
|
109
109
|
export declare class AutoscaledPool {
|
|
110
110
|
private readonly log;
|
|
@@ -10,8 +10,8 @@ import { serviceLocator } from '../service_locator.js';
|
|
|
10
10
|
*
|
|
11
11
|
* Before running the pool, you need to implement the following three functions:
|
|
12
12
|
* {@link AutoscaledPoolOptions.runTaskFunction|`runTaskFunction`},
|
|
13
|
-
* {@link
|
|
14
|
-
* {@link
|
|
13
|
+
* {@link TaskLoopPredicates.isTaskReadyFunction|`isTaskReadyFunction`} and
|
|
14
|
+
* {@link TaskLoopPredicates.isFinishedFunction|`isFinishedFunction`}.
|
|
15
15
|
*
|
|
16
16
|
* The auto-scaled pool is started by calling the {@link AutoscaledPool.run} function.
|
|
17
17
|
* The pool periodically queries `isTaskReadyFunction` for more tasks, managing optimal concurrency, until the function
|
|
@@ -52,7 +52,8 @@ import { serviceLocator } from '../service_locator.js';
|
|
|
52
52
|
* await concurrencySystem.stop();
|
|
53
53
|
* }
|
|
54
54
|
* ```
|
|
55
|
-
*
|
|
55
|
+
*
|
|
56
|
+
* @internal
|
|
56
57
|
*/
|
|
57
58
|
export class AutoscaledPool {
|
|
58
59
|
log;
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@crawlee/core",
|
|
3
|
-
"version": "4.0.0-beta.
|
|
3
|
+
"version": "4.0.0-beta.93",
|
|
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": ">=22.0.0"
|
|
@@ -53,9 +53,9 @@
|
|
|
53
53
|
"@apify/pseudo_url": "^2.0.59",
|
|
54
54
|
"@apify/timeout": "^0.3.2",
|
|
55
55
|
"@apify/utilities": "^2.15.5",
|
|
56
|
-
"@crawlee/fs-storage": "4.0.0-beta.
|
|
57
|
-
"@crawlee/types": "4.0.0-beta.
|
|
58
|
-
"@crawlee/utils": "4.0.0-beta.
|
|
56
|
+
"@crawlee/fs-storage": "4.0.0-beta.93",
|
|
57
|
+
"@crawlee/types": "4.0.0-beta.93",
|
|
58
|
+
"@crawlee/utils": "4.0.0-beta.93",
|
|
59
59
|
"@sapphire/async-queue": "^1.5.5",
|
|
60
60
|
"@sapphire/shapeshift": "^4.0.0",
|
|
61
61
|
"@vladfrangu/async_event_emitter": "^2.4.6",
|
|
@@ -79,5 +79,5 @@
|
|
|
79
79
|
}
|
|
80
80
|
}
|
|
81
81
|
},
|
|
82
|
-
"gitHead": "
|
|
82
|
+
"gitHead": "b9d21e80d94d01e21d4e1c19191610d0157dd172"
|
|
83
83
|
}
|