@crawlee/browser-pool 4.0.0-beta.7 → 4.0.0-beta.70

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.
Files changed (62) hide show
  1. package/README.md +17 -13
  2. package/abstract-classes/browser-controller.d.ts +67 -7
  3. package/abstract-classes/browser-controller.d.ts.map +1 -1
  4. package/abstract-classes/browser-controller.js +6 -9
  5. package/abstract-classes/browser-controller.js.map +1 -1
  6. package/abstract-classes/browser-plugin.d.ts +37 -2
  7. package/abstract-classes/browser-plugin.d.ts.map +1 -1
  8. package/abstract-classes/browser-plugin.js +65 -6
  9. package/abstract-classes/browser-plugin.js.map +1 -1
  10. package/anonymize-proxy.d.ts +5 -1
  11. package/anonymize-proxy.d.ts.map +1 -1
  12. package/anonymize-proxy.js +9 -4
  13. package/anonymize-proxy.js.map +1 -1
  14. package/browser-pool.d.ts +88 -13
  15. package/browser-pool.d.ts.map +1 -1
  16. package/browser-pool.js +139 -25
  17. package/browser-pool.js.map +1 -1
  18. package/fingerprinting/hooks.d.ts.map +1 -1
  19. package/fingerprinting/hooks.js +24 -8
  20. package/fingerprinting/hooks.js.map +1 -1
  21. package/index.d.ts +14 -6
  22. package/index.d.ts.map +1 -1
  23. package/index.js +5 -2
  24. package/index.js.map +1 -1
  25. package/launch-context.d.ts +19 -2
  26. package/launch-context.d.ts.map +1 -1
  27. package/launch-context.js +11 -3
  28. package/launch-context.js.map +1 -1
  29. package/package.json +7 -8
  30. package/playwright/playwright-browser.d.ts.map +1 -1
  31. package/playwright/playwright-browser.js.map +1 -1
  32. package/playwright/playwright-controller.d.ts.map +1 -1
  33. package/playwright/playwright-controller.js +20 -9
  34. package/playwright/playwright-controller.js.map +1 -1
  35. package/playwright/playwright-plugin.d.ts +6 -0
  36. package/playwright/playwright-plugin.d.ts.map +1 -1
  37. package/playwright/playwright-plugin.js +27 -3
  38. package/playwright/playwright-plugin.js.map +1 -1
  39. package/puppeteer/puppeteer-controller.d.ts.map +1 -1
  40. package/puppeteer/puppeteer-controller.js +11 -8
  41. package/puppeteer/puppeteer-controller.js.map +1 -1
  42. package/puppeteer/puppeteer-plugin.d.ts +3 -0
  43. package/puppeteer/puppeteer-plugin.d.ts.map +1 -1
  44. package/puppeteer/puppeteer-plugin.js +81 -43
  45. package/puppeteer/puppeteer-plugin.js.map +1 -1
  46. package/remote-browser-pool.d.ts +173 -0
  47. package/remote-browser-pool.d.ts.map +1 -0
  48. package/remote-browser-pool.js +192 -0
  49. package/remote-browser-pool.js.map +1 -0
  50. package/remote-browser-provider.d.ts +84 -0
  51. package/remote-browser-provider.d.ts.map +1 -0
  52. package/remote-browser-provider.js +68 -0
  53. package/remote-browser-provider.js.map +1 -0
  54. package/utils.d.ts +7 -0
  55. package/utils.d.ts.map +1 -1
  56. package/utils.js +19 -0
  57. package/utils.js.map +1 -1
  58. package/logger.d.ts +0 -3
  59. package/logger.d.ts.map +0 -1
  60. package/logger.js +0 -5
  61. package/logger.js.map +0 -1
  62. package/tsconfig.build.tsbuildinfo +0 -1
@@ -0,0 +1,84 @@
1
+ /**
2
+ * Abstract base class for remote browser service providers.
3
+ *
4
+ * Implement this class to encapsulate the lifecycle of a remote browser session
5
+ * (creation, connection URL resolution, and cleanup). {@link RemoteBrowserPool}
6
+ * calls {@link connect} once per browser launch and {@link release} when the browser
7
+ * closes, crashes, the pool is destroyed, or the connection fails during launch.
8
+ *
9
+ * Pass the provider instance as the `endpoint` of a {@link RemoteBrowserPool}, then
10
+ * hand the pool to a crawler via its `browserPool` option:
11
+ *
12
+ * ```typescript
13
+ * const browserPool = new RemoteBrowserPool({
14
+ * browserPlugins: [new PlaywrightPlugin(playwright.chromium)],
15
+ * endpoint: new MyProvider(),
16
+ * });
17
+ *
18
+ * const crawler = new PlaywrightCrawler({ browserPool });
19
+ * ```
20
+ *
21
+ * **Example — simple static endpoint (e.g. Browserless):**
22
+ * ```typescript
23
+ * class BrowserlessProvider extends RemoteBrowserProvider {
24
+ * maxOpenBrowsers = 2; // respect the service's concurrent session limit
25
+ *
26
+ * async connect() {
27
+ * return { url: `wss://production-sfo.browserless.io?token=${token}` };
28
+ * }
29
+ * }
30
+ * ```
31
+ *
32
+ * **Example — session lifecycle with concurrency limit (e.g. Browserbase):**
33
+ * ```typescript
34
+ * class BrowserbaseProvider extends RemoteBrowserProvider<{ id: string }> {
35
+ * maxOpenBrowsers = 2; // respect the service's concurrent session limit
36
+ *
37
+ * async connect({ proxyUrl } = {}) {
38
+ * const session = await createSession(apiKey, projectId, {
39
+ * proxies: proxyUrl ? [{ type: 'external', server: proxyUrl }] : undefined,
40
+ * });
41
+ * return { url: session.connectUrl, context: { id: session.id } };
42
+ * }
43
+ *
44
+ * async release(context: { id: string }) {
45
+ * await releaseSession(apiKey, context.id);
46
+ * }
47
+ * }
48
+ * ```
49
+ */
50
+ export declare abstract class RemoteBrowserProvider<TContext extends Record<string, unknown> = Record<string, unknown>> {
51
+ /**
52
+ * Maximum number of browsers that can be open at the same time.
53
+ * Set this to your remote service's concurrent session limit to avoid 429 errors.
54
+ */
55
+ maxOpenBrowsers?: number;
56
+ /**
57
+ * Called once per browser launch. Return the WebSocket/CDP endpoint URL
58
+ * and an optional `context` object that will be passed back to {@link release}.
59
+ *
60
+ * @param options.proxyUrl - The proxy URL resolved by Crawlee's proxy configuration
61
+ * for this browser session. Pass it to your remote service's proxy API if supported.
62
+ */
63
+ abstract connect(options?: {
64
+ proxyUrl?: string;
65
+ }): Promise<{
66
+ url: string;
67
+ context?: TContext;
68
+ }> | {
69
+ url: string;
70
+ context?: TContext;
71
+ };
72
+ /**
73
+ * Called when the browser closes, crashes, the pool is destroyed, or the
74
+ * connection fails right after {@link connect} succeeds.
75
+ * Override this to clean up remote sessions, release API resources, etc.
76
+ *
77
+ * Errors thrown here are caught and logged as warnings — they never crash the crawler.
78
+ * Safe to assume this is called at most once per {@link connect} call.
79
+ *
80
+ * @param _context The same `context` object returned by {@link connect}.
81
+ */
82
+ release(_context: TContext): Promise<void>;
83
+ }
84
+ //# sourceMappingURL=remote-browser-provider.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"remote-browser-provider.d.ts","sourceRoot":"","sources":["../src/remote-browser-provider.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAgDG;AACH,8BAAsB,qBAAqB,CAAC,QAAQ,SAAS,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC;IAC1G;;;OAGG;IACH,eAAe,CAAC,EAAE,MAAM,CAAC;IAEzB;;;;;;OAMG;IACH,QAAQ,CAAC,OAAO,CAAC,OAAO,CAAC,EAAE;QACvB,QAAQ,CAAC,EAAE,MAAM,CAAC;KACrB,GAAG,OAAO,CAAC;QAAE,GAAG,EAAE,MAAM,CAAC;QAAC,OAAO,CAAC,EAAE,QAAQ,CAAA;KAAE,CAAC,GAAG;QAAE,GAAG,EAAE,MAAM,CAAC;QAAC,OAAO,CAAC,EAAE,QAAQ,CAAA;KAAE;IAEtF;;;;;;;;;OASG;IACG,OAAO,CAAC,QAAQ,EAAE,QAAQ,GAAG,OAAO,CAAC,IAAI,CAAC;CACnD"}
@@ -0,0 +1,68 @@
1
+ /**
2
+ * Abstract base class for remote browser service providers.
3
+ *
4
+ * Implement this class to encapsulate the lifecycle of a remote browser session
5
+ * (creation, connection URL resolution, and cleanup). {@link RemoteBrowserPool}
6
+ * calls {@link connect} once per browser launch and {@link release} when the browser
7
+ * closes, crashes, the pool is destroyed, or the connection fails during launch.
8
+ *
9
+ * Pass the provider instance as the `endpoint` of a {@link RemoteBrowserPool}, then
10
+ * hand the pool to a crawler via its `browserPool` option:
11
+ *
12
+ * ```typescript
13
+ * const browserPool = new RemoteBrowserPool({
14
+ * browserPlugins: [new PlaywrightPlugin(playwright.chromium)],
15
+ * endpoint: new MyProvider(),
16
+ * });
17
+ *
18
+ * const crawler = new PlaywrightCrawler({ browserPool });
19
+ * ```
20
+ *
21
+ * **Example — simple static endpoint (e.g. Browserless):**
22
+ * ```typescript
23
+ * class BrowserlessProvider extends RemoteBrowserProvider {
24
+ * maxOpenBrowsers = 2; // respect the service's concurrent session limit
25
+ *
26
+ * async connect() {
27
+ * return { url: `wss://production-sfo.browserless.io?token=${token}` };
28
+ * }
29
+ * }
30
+ * ```
31
+ *
32
+ * **Example — session lifecycle with concurrency limit (e.g. Browserbase):**
33
+ * ```typescript
34
+ * class BrowserbaseProvider extends RemoteBrowserProvider<{ id: string }> {
35
+ * maxOpenBrowsers = 2; // respect the service's concurrent session limit
36
+ *
37
+ * async connect({ proxyUrl } = {}) {
38
+ * const session = await createSession(apiKey, projectId, {
39
+ * proxies: proxyUrl ? [{ type: 'external', server: proxyUrl }] : undefined,
40
+ * });
41
+ * return { url: session.connectUrl, context: { id: session.id } };
42
+ * }
43
+ *
44
+ * async release(context: { id: string }) {
45
+ * await releaseSession(apiKey, context.id);
46
+ * }
47
+ * }
48
+ * ```
49
+ */
50
+ export class RemoteBrowserProvider {
51
+ /**
52
+ * Maximum number of browsers that can be open at the same time.
53
+ * Set this to your remote service's concurrent session limit to avoid 429 errors.
54
+ */
55
+ maxOpenBrowsers;
56
+ /**
57
+ * Called when the browser closes, crashes, the pool is destroyed, or the
58
+ * connection fails right after {@link connect} succeeds.
59
+ * Override this to clean up remote sessions, release API resources, etc.
60
+ *
61
+ * Errors thrown here are caught and logged as warnings — they never crash the crawler.
62
+ * Safe to assume this is called at most once per {@link connect} call.
63
+ *
64
+ * @param _context The same `context` object returned by {@link connect}.
65
+ */
66
+ async release(_context) { }
67
+ }
68
+ //# sourceMappingURL=remote-browser-provider.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"remote-browser-provider.js","sourceRoot":"","sources":["../src/remote-browser-provider.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAgDG;AACH,MAAM,OAAgB,qBAAqB;IACvC;;;OAGG;IACH,eAAe,CAAU;IAazB;;;;;;;;;OASG;IACH,KAAK,CAAC,OAAO,CAAC,QAAkB,IAAkB,CAAC;CACtD"}
package/utils.d.ts CHANGED
@@ -3,6 +3,13 @@ import type { PlaywrightPlugin } from './playwright/playwright-plugin.js';
3
3
  import type { PuppeteerPlugin } from './puppeteer/puppeteer-plugin.js';
4
4
  export type UnwrapPromise<T> = T extends PromiseLike<infer R> ? UnwrapPromise<R> : T;
5
5
  export declare function noop(..._args: unknown[]): void;
6
+ /**
7
+ * Strips secrets from a URL so it can be safely included in logs and error messages. Removes userinfo
8
+ * credentials and the entire query string and fragment — remote browser services routinely carry tokens
9
+ * there (e.g. Browserless `?token=…`), and we can't tell which params are sensitive. Keeps the
10
+ * protocol, host, port, and path, which are enough to diagnose connection failures.
11
+ */
12
+ export declare function sanitizeEndpointForLog(endpoint: string): string;
6
13
  /**
7
14
  * This is required when using optional dependencies.
8
15
  * Importing a type gives `any`, but `Parameters<any>` gives `unknown[]` instead of `any`
package/utils.d.ts.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"utils.d.ts","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,sCAAsC,CAAC;AAC1E,OAAO,KAAK,EAAE,gBAAgB,EAAE,MAAM,mCAAmC,CAAC;AAC1E,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,iCAAiC,CAAC;AAEvE,MAAM,MAAM,aAAa,CAAC,CAAC,IAAI,CAAC,SAAS,WAAW,CAAC,MAAM,CAAC,CAAC,GAAG,aAAa,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC;AAErF,wBAAgB,IAAI,CAAC,GAAG,KAAK,EAAE,OAAO,EAAE,GAAG,IAAI,CAAG;AAElD;;;GAGG;AACH,MAAM,MAAM,cAAc,CAAC,CAAC,SAAS,CAAC,GAAG,IAAI,EAAE,GAAG,KAAK,GAAG,IAAI,OAAO,EAAE,SAAS,UAAU,CAAC,CAAC,CAAC,GAAG,GAAG,GAAG,UAAU,CAAC,CAAC,CAAC,CAAC;AAEpH,MAAM,MAAM,uBAAuB,CAE/B,KAAK,SAAS,SAAS,OAAO,EAAE,EAEhC,MAAM,SAAS,aAAa,EAAE,GAAG,EAAE,IACnC,KAAK,SAAS,SAAS,CAAC,MAAM,UAAU,EAAE,GAAG,MAAM,IAAI,CAAC,GAAG,CAAC,MAAM,UAAU,EAAE,GAAG,MAAM,IAAI,CAAC,GAE1F,UAAU,SAAS,gBAAgB,GAE/B,uBAAuB,CAAC,IAAI,EAAE,CAAC,GAAG,MAAM,EAAE,gBAAgB,CAAC,CAAC,GAE5D,UAAU,SAAS,eAAe,GAEhC,uBAAuB,CAAC,IAAI,EAAE,CAAC,GAAG,MAAM,EAAE,eAAe,CAAC,CAAC,GAE3D,KAAK,GAEX,KAAK,SAAS,EAAE,GAEd,MAAM,GAEN,KAAK,SAAS,SAAS,CAAC,MAAM,CAAC,CAAC,EAAE,GAEhC;IAAC,CAAC;CAAC,SAAS,CAAC,eAAe,GAAG,gBAAgB,CAAC,GAE5C,CAAC,EAAE,GAEH,KAAK,GAET,MAAM,CAAC"}
1
+ {"version":3,"file":"utils.d.ts","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,aAAa,EAAE,MAAM,sCAAsC,CAAC;AAC1E,OAAO,KAAK,EAAE,gBAAgB,EAAE,MAAM,mCAAmC,CAAC;AAC1E,OAAO,KAAK,EAAE,eAAe,EAAE,MAAM,iCAAiC,CAAC;AAEvE,MAAM,MAAM,aAAa,CAAC,CAAC,IAAI,CAAC,SAAS,WAAW,CAAC,MAAM,CAAC,CAAC,GAAG,aAAa,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC;AAErF,wBAAgB,IAAI,CAAC,GAAG,KAAK,EAAE,OAAO,EAAE,GAAG,IAAI,CAAG;AAElD;;;;;GAKG;AACH,wBAAgB,sBAAsB,CAAC,QAAQ,EAAE,MAAM,GAAG,MAAM,CAW/D;AAED;;;GAGG;AACH,MAAM,MAAM,cAAc,CAAC,CAAC,SAAS,CAAC,GAAG,IAAI,EAAE,GAAG,KAAK,GAAG,IAAI,OAAO,EAAE,SAAS,UAAU,CAAC,CAAC,CAAC,GAAG,GAAG,GAAG,UAAU,CAAC,CAAC,CAAC,CAAC;AAEpH,MAAM,MAAM,uBAAuB,CAE/B,KAAK,SAAS,SAAS,OAAO,EAAE,EAEhC,MAAM,SAAS,aAAa,EAAE,GAAG,EAAE,IACnC,KAAK,SAAS,SAAS,CAAC,MAAM,UAAU,EAAE,GAAG,MAAM,IAAI,CAAC,GAAG,CAAC,MAAM,UAAU,EAAE,GAAG,MAAM,IAAI,CAAC,GAE1F,UAAU,SAAS,gBAAgB,GAE/B,uBAAuB,CAAC,IAAI,EAAE,CAAC,GAAG,MAAM,EAAE,gBAAgB,CAAC,CAAC,GAE5D,UAAU,SAAS,eAAe,GAEhC,uBAAuB,CAAC,IAAI,EAAE,CAAC,GAAG,MAAM,EAAE,eAAe,CAAC,CAAC,GAE3D,KAAK,GAEX,KAAK,SAAS,EAAE,GAEd,MAAM,GAEN,KAAK,SAAS,SAAS,CAAC,MAAM,CAAC,CAAC,EAAE,GAEhC;IAAC,CAAC;CAAC,SAAS,CAAC,eAAe,GAAG,gBAAgB,CAAC,GAE5C,CAAC,EAAE,GAEH,KAAK,GAET,MAAM,CAAC"}
package/utils.js CHANGED
@@ -1,2 +1,21 @@
1
1
  export function noop(..._args) { }
2
+ /**
3
+ * Strips secrets from a URL so it can be safely included in logs and error messages. Removes userinfo
4
+ * credentials and the entire query string and fragment — remote browser services routinely carry tokens
5
+ * there (e.g. Browserless `?token=…`), and we can't tell which params are sensitive. Keeps the
6
+ * protocol, host, port, and path, which are enough to diagnose connection failures.
7
+ */
8
+ export function sanitizeEndpointForLog(endpoint) {
9
+ try {
10
+ const url = new URL(endpoint);
11
+ url.username = '';
12
+ url.password = '';
13
+ url.search = '';
14
+ url.hash = '';
15
+ return url.toString();
16
+ }
17
+ catch {
18
+ return '<invalid URL>';
19
+ }
20
+ }
2
21
  //# sourceMappingURL=utils.js.map
package/utils.js.map CHANGED
@@ -1 +1 @@
1
- {"version":3,"file":"utils.js","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":"AAMA,MAAM,UAAU,IAAI,CAAC,GAAG,KAAgB,IAAS,CAAC"}
1
+ {"version":3,"file":"utils.js","sourceRoot":"","sources":["../src/utils.ts"],"names":[],"mappings":"AAMA,MAAM,UAAU,IAAI,CAAC,GAAG,KAAgB,IAAS,CAAC;AAElD;;;;;GAKG;AACH,MAAM,UAAU,sBAAsB,CAAC,QAAgB;IACnD,IAAI,CAAC;QACD,MAAM,GAAG,GAAG,IAAI,GAAG,CAAC,QAAQ,CAAC,CAAC;QAC9B,GAAG,CAAC,QAAQ,GAAG,EAAE,CAAC;QAClB,GAAG,CAAC,QAAQ,GAAG,EAAE,CAAC;QAClB,GAAG,CAAC,MAAM,GAAG,EAAE,CAAC;QAChB,GAAG,CAAC,IAAI,GAAG,EAAE,CAAC;QACd,OAAO,GAAG,CAAC,QAAQ,EAAE,CAAC;IAC1B,CAAC;IAAC,MAAM,CAAC;QACL,OAAO,eAAe,CAAC;IAC3B,CAAC;AACL,CAAC"}
package/logger.d.ts DELETED
@@ -1,3 +0,0 @@
1
- // @ts-ignore optional peer dependency or compatibility with es2022
2
- export declare const log: import("@apify/log").Log;
3
- //# sourceMappingURL=logger.d.ts.map
package/logger.d.ts.map DELETED
@@ -1 +0,0 @@
1
- {"version":3,"file":"logger.d.ts","sourceRoot":"","sources":["../src/logger.ts"],"names":[],"mappings":"AAEA,eAAO,MAAM,GAAG,0BAEd,CAAC"}
package/logger.js DELETED
@@ -1,5 +0,0 @@
1
- import defaultLog from '@apify/log';
2
- export const log = defaultLog.child({
3
- prefix: 'BrowserPool',
4
- });
5
- //# sourceMappingURL=logger.js.map
package/logger.js.map DELETED
@@ -1 +0,0 @@
1
- {"version":3,"file":"logger.js","sourceRoot":"","sources":["../src/logger.ts"],"names":[],"mappings":"AAAA,OAAO,UAAU,MAAM,YAAY,CAAC;AAEpC,MAAM,CAAC,MAAM,GAAG,GAAG,UAAU,CAAC,KAAK,CAAC;IAChC,MAAM,EAAE,aAAa;CACxB,CAAC,CAAC"}