@crawlee/core 4.0.0-beta.175 → 4.0.0-beta.176
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/configuration.d.ts
CHANGED
|
@@ -9,14 +9,8 @@ export declare function field<T extends z.ZodType>(schema: T, envVar?: string |
|
|
|
9
9
|
export declare const coerceBoolean: z.ZodPreprocess<z.ZodBoolean, unknown>;
|
|
10
10
|
export declare const coerceNumber: z.ZodPreprocess<z.ZodNumber, unknown>;
|
|
11
11
|
export declare const crawleeConfigFields: {
|
|
12
|
-
/** @default 'default' */
|
|
13
|
-
defaultDatasetId: ConfigField<z.ZodDefault<z.ZodString>>;
|
|
14
12
|
/** @default true */
|
|
15
13
|
purgeOnStart: ConfigField<z.ZodDefault<z.ZodPreprocess<z.ZodBoolean, unknown>>>;
|
|
16
|
-
/** @default 'default' */
|
|
17
|
-
defaultKeyValueStoreId: ConfigField<z.ZodDefault<z.ZodString>>;
|
|
18
|
-
/** @default 'default' */
|
|
19
|
-
defaultRequestQueueId: ConfigField<z.ZodDefault<z.ZodString>>;
|
|
20
14
|
/** @default 0.95 */
|
|
21
15
|
maxUsedCpuRatio: ConfigField<z.ZodDefault<z.ZodPreprocess<z.ZodNumber, unknown>>>;
|
|
22
16
|
/** @default 0.25 */
|
|
@@ -101,9 +95,6 @@ export interface Configuration extends ResolvedConfigValues {
|
|
|
101
95
|
* `memoryMbytes` | `CRAWLEE_MEMORY_MBYTES` | -
|
|
102
96
|
* `logLevel` | `CRAWLEE_LOG_LEVEL` | -
|
|
103
97
|
* `headless` | `CRAWLEE_HEADLESS` | `true`
|
|
104
|
-
* `defaultDatasetId` | `CRAWLEE_DEFAULT_DATASET_ID` | `'default'`
|
|
105
|
-
* `defaultKeyValueStoreId` | `CRAWLEE_DEFAULT_KEY_VALUE_STORE_ID` | `'default'`
|
|
106
|
-
* `defaultRequestQueueId` | `CRAWLEE_DEFAULT_REQUEST_QUEUE_ID` | `'default'`
|
|
107
98
|
* `persistStateIntervalMillis` | `CRAWLEE_PERSIST_STATE_INTERVAL_MILLIS` | `60_000`
|
|
108
99
|
* `internalTimeoutMillis` | `CRAWLEE_INTERNAL_TIMEOUT` | -
|
|
109
100
|
* `purgeOnStart` | `CRAWLEE_PURGE_ON_START` | `true`
|
package/configuration.js
CHANGED
|
@@ -39,14 +39,8 @@ const logLevelSchema = z.preprocess((val) => {
|
|
|
39
39
|
}, z.enum(LogLevel));
|
|
40
40
|
// --- Crawlee config field definitions ---
|
|
41
41
|
export const crawleeConfigFields = {
|
|
42
|
-
/** @default 'default' */
|
|
43
|
-
defaultDatasetId: field(z.string().default('default'), 'CRAWLEE_DEFAULT_DATASET_ID'),
|
|
44
42
|
/** @default true */
|
|
45
43
|
purgeOnStart: field(coerceBoolean.default(true), 'CRAWLEE_PURGE_ON_START'),
|
|
46
|
-
/** @default 'default' */
|
|
47
|
-
defaultKeyValueStoreId: field(z.string().default('default'), 'CRAWLEE_DEFAULT_KEY_VALUE_STORE_ID'),
|
|
48
|
-
/** @default 'default' */
|
|
49
|
-
defaultRequestQueueId: field(z.string().default('default'), 'CRAWLEE_DEFAULT_REQUEST_QUEUE_ID'),
|
|
50
44
|
/** @default 0.95 */
|
|
51
45
|
maxUsedCpuRatio: field(coerceNumber.default(0.95)),
|
|
52
46
|
/** @default 0.25 */
|
|
@@ -119,9 +113,6 @@ export const crawleeConfigFields = {
|
|
|
119
113
|
* `memoryMbytes` | `CRAWLEE_MEMORY_MBYTES` | -
|
|
120
114
|
* `logLevel` | `CRAWLEE_LOG_LEVEL` | -
|
|
121
115
|
* `headless` | `CRAWLEE_HEADLESS` | `true`
|
|
122
|
-
* `defaultDatasetId` | `CRAWLEE_DEFAULT_DATASET_ID` | `'default'`
|
|
123
|
-
* `defaultKeyValueStoreId` | `CRAWLEE_DEFAULT_KEY_VALUE_STORE_ID` | `'default'`
|
|
124
|
-
* `defaultRequestQueueId` | `CRAWLEE_DEFAULT_REQUEST_QUEUE_ID` | `'default'`
|
|
125
116
|
* `persistStateIntervalMillis` | `CRAWLEE_PERSIST_STATE_INTERVAL_MILLIS` | `60_000`
|
|
126
117
|
* `internalTimeoutMillis` | `CRAWLEE_INTERNAL_TIMEOUT` | -
|
|
127
118
|
* `purgeOnStart` | `CRAWLEE_PURGE_ON_START` | `true`
|
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.176",
|
|
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"
|
|
@@ -53,10 +53,10 @@
|
|
|
53
53
|
"@apify/log": "^3.0.1",
|
|
54
54
|
"@apify/timeout": "^1.0.1",
|
|
55
55
|
"@apify/utilities": "^3.0.1",
|
|
56
|
-
"@crawlee/fs-storage": "4.0.0-beta.
|
|
57
|
-
"@crawlee/http-client": "4.0.0-beta.
|
|
58
|
-
"@crawlee/types": "4.0.0-beta.
|
|
59
|
-
"@crawlee/utils": "4.0.0-beta.
|
|
56
|
+
"@crawlee/fs-storage": "4.0.0-beta.176",
|
|
57
|
+
"@crawlee/http-client": "4.0.0-beta.176",
|
|
58
|
+
"@crawlee/types": "4.0.0-beta.176",
|
|
59
|
+
"@crawlee/utils": "4.0.0-beta.176",
|
|
60
60
|
"@sapphire/async-queue": "^1.5.5",
|
|
61
61
|
"@standard-schema/spec": "^1.0.0",
|
|
62
62
|
"@vladfrangu/async_event_emitter": "^2.4.6",
|
|
@@ -76,5 +76,5 @@
|
|
|
76
76
|
}
|
|
77
77
|
}
|
|
78
78
|
},
|
|
79
|
-
"gitHead": "
|
|
79
|
+
"gitHead": "3e9e27d524dff8fe72aaf337e3d24fc2232eb241"
|
|
80
80
|
}
|
|
@@ -29,8 +29,7 @@ import type { StorageIdentifier } from './storage_instance_manager.js';
|
|
|
29
29
|
* ```
|
|
30
30
|
* {CRAWLEE_STORAGE_DIR}/key_value_stores/{STORE_ID}/{INDEX}.{EXT}
|
|
31
31
|
* ```
|
|
32
|
-
* Note that `{STORE_ID}` is the name or ID of the key-value store. The default key-value store has ID: `default
|
|
33
|
-
* unless you override it by setting the `CRAWLEE_DEFAULT_KEY_VALUE_STORE_ID` environment variable.
|
|
32
|
+
* Note that `{STORE_ID}` is the name or ID of the key-value store. The default key-value store has ID: `default`.
|
|
34
33
|
* The `{KEY}` is the key of the record and `{EXT}` corresponds to the MIME content type of the data value.
|
|
35
34
|
*
|
|
36
35
|
* **Example usage:**
|
|
@@ -51,8 +51,7 @@ const openOptionsSchema = z.strictObject({
|
|
|
51
51
|
* ```
|
|
52
52
|
* {CRAWLEE_STORAGE_DIR}/key_value_stores/{STORE_ID}/{INDEX}.{EXT}
|
|
53
53
|
* ```
|
|
54
|
-
* Note that `{STORE_ID}` is the name or ID of the key-value store. The default key-value store has ID: `default
|
|
55
|
-
* unless you override it by setting the `CRAWLEE_DEFAULT_KEY_VALUE_STORE_ID` environment variable.
|
|
54
|
+
* Note that `{STORE_ID}` is the name or ID of the key-value store. The default key-value store has ID: `default`.
|
|
56
55
|
* The `{KEY}` is the key of the record and `{EXT}` corresponds to the MIME content type of the data value.
|
|
57
56
|
*
|
|
58
57
|
* **Example usage:**
|