apify 4.0.0-beta.47 → 4.0.0-beta.48

package/dist/proxy_configuration.d.ts

@@ -172,8 +172,12 @@ export declare class ProxyConfiguration extends CoreProxyConfiguration {
      * Returns a new {@link ProxyInfo} object with a fresh proxy URL. Each call mints an
      * independent URL; for Apify Proxy a random session id is embedded so consecutive
      * calls resolve to different IPs.
+     *
+     * A passed `proxyInfo` (e.g. one restored with a persisted session) is returned unchanged, unless it is an
+     * Apify Proxy one. Its endpoint is specific to the host, so it is moved to the current endpoint, keeping the
+     * username and with it the proxy session.
      */
-    newProxyInfo(): Promise<ProxyInfo | undefined>;
+    newProxyInfo(proxyInfo?: ProxyInfo): Promise<ProxyInfo | undefined>;
     /**
      * Returns a new proxy URL. For Apify Proxy, each call generates a URL with a fresh
      * random session id, so consecutive calls return independent URLs. For custom
@@ -185,6 +189,7 @@ export declare class ProxyConfiguration extends CoreProxyConfiguration {
      */
     protected getUsername(sessionId: string): string;
     protected composeDefaultUrl(sessionId: string): string;
+    private composeUrl;
     /**
      * Fetch & set the proxy password from Apify API if an Apify token is provided.
      */

package/dist/proxy_configuration.js

@@ -18,6 +18,7 @@ const SUBDIVISION_CODE_REGEX = /^[A-Z0-9]{1,3}$/;
 // users; a fresh one is minted for every URL the SDK hands out so that the
 // returned proxy URLs are independent.
 const SESSION_ID_LENGTH = 12;
+const SESSION_USERNAME_REGEX = /(?:^|,)session-[^,]+/;
 /**
  * Configures connection to a proxy server with the provided options. Proxy servers are used to prevent target websites from blocking
  * your crawlers based on IP address rate limits or blacklists. Setting proxy configuration in your crawlers automatically configures
@@ -146,9 +147,18 @@ export class ProxyConfiguration extends CoreProxyConfiguration {
      * Returns a new {@link ProxyInfo} object with a fresh proxy URL. Each call mints an
      * independent URL; for Apify Proxy a random session id is embedded so consecutive
      * calls resolve to different IPs.
+     *
+     * A passed `proxyInfo` (e.g. one restored with a persisted session) is returned unchanged, unless it is an
+     * Apify Proxy one. Its endpoint is specific to the host, so it is moved to the current endpoint, keeping the
+     * username and with it the proxy session.
      */
-    async newProxyInfo() {
-        const url = await this.newUrl();
+    async newProxyInfo(proxyInfo) {
+        // Only Apify Proxy infos carry `groups` and a proxy session in the username.
+        if (proxyInfo &&
+            !(this.#usesApifyProxy && 'groups' in proxyInfo && SESSION_USERNAME_REGEX.test(proxyInfo.username ?? ''))) {
+            return proxyInfo;
+        }
+        const url = proxyInfo?.username ? this.composeUrl(proxyInfo.username) : await this.newUrl();
         if (!url)
             return undefined;
         const parsed = new URL(url);
@@ -159,6 +169,9 @@ export class ProxyConfiguration extends CoreProxyConfiguration {
             hostname: parsed.hostname,
             port: parsed.port,
         };
+        // A refreshed info keeps its own groups/country, which match its username rather than this config.
+        if (proxyInfo)
+            return { ...proxyInfo, ...result };
         if (this.#usesApifyProxy) {
             result.groups = this.#groups;
             if (this.#countryCode !== undefined)
@@ -197,7 +210,9 @@ export class ProxyConfiguration extends CoreProxyConfiguration {
         return parts.join(',');
     }
     composeDefaultUrl(sessionId) {
-        const username = this.getUsername(sessionId);
+        return this.composeUrl(this.getUsername(sessionId));
+    }
+    composeUrl(username) {
         const url = new URL(`http://${this.#hostname}:${this.#port}`);
         url.username = `${username}`;
         url.password = `${this.#password}`;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "apify",
-  "version": "4.0.0-beta.47",
+  "version": "4.0.0-beta.48",
   "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.",
   "engines": {
     "node": ">=22.13.0"
@@ -61,10 +61,10 @@
     "@apify/timeout": "^1.0.1",
     "@apify/utilities": "^3.1.0",
     "@apify/validations": "^1.1.1",
-    "@crawlee/core": "4.0.0-beta.201",
-    "@crawlee/fs-storage": "4.0.0-beta.201",
-    "@crawlee/types": "4.0.0-beta.201",
-    "@crawlee/utils": "4.0.0-beta.201",
+    "@crawlee/core": "4.0.0-beta.213",
+    "@crawlee/fs-storage": "4.0.0-beta.213",
+    "@crawlee/types": "4.0.0-beta.213",
+    "@crawlee/utils": "4.0.0-beta.213",
     "apify-client": "^3.0.0-beta.28",
     "minimatch": "^10.2.6",
     "semver": "^7.8.5",
@@ -81,7 +81,7 @@
     "@types/semver": "^7.8.0",
     "@types/ws": "^8.18.1",
     "commitlint": "^21.2.3",
-    "crawlee": "4.0.0-beta.201",
+    "crawlee": "4.0.0-beta.213",
     "husky": "^9.1.7",
     "lint-staged": "^17.6.0",
     "oxfmt": "0.71.0",