community[patch]: Support for specifying api url for firecrawl docume…

…nt loader (#6488) * Support for specifying api url for firecrawl document loader * refactor: remove default apiUrl value * refactor: change comment * Fix --------- Co-authored-by: jacoblee93 <[email protected]>
langchain-ai · Aug 12, 2024 · 3ff65cb · 3ff65cb
1 parent 4bc1621
commit 3ff65cb
Show file tree

Hide file tree

Showing 3 changed files with 31 additions and 4 deletions.
diff --git a/libs/langchain-community/package.json b/libs/langchain-community/package.json
@@ -82,7 +82,7 @@
     "@langchain/scripts": "~0.0.20",
     "@langchain/standard-tests": "0.0.0",
     "@layerup/layerup-security": "^1.5.12",
-    "@mendable/firecrawl-js": "^0.0.13",
+    "@mendable/firecrawl-js": "^0.0.36",
     "@mlc-ai/web-llm": "0.2.46",
     "@mozilla/readability": "^0.4.4",
     "@neondatabase/serverless": "^0.9.1",

diff --git a/libs/langchain-community/src/document_loaders/web/firecrawl.ts b/libs/langchain-community/src/document_loaders/web/firecrawl.ts
@@ -18,6 +18,10 @@ interface FirecrawlLoaderParameters {
    */
   apiKey?: string;
 
+  /**
+   * API URL for Firecrawl.
+   */
+  apiUrl?: string;
   /**
    * Mode of operation. Can be either "crawl" or "scrape". If not provided, the default value is "crawl".
    */
@@ -46,6 +50,8 @@ interface FirecrawlDocument {
 export class FireCrawlLoader extends BaseDocumentLoader {
   private apiKey: string;
 
+  private apiUrl?: string;
+
   private url: string;
 
   private mode: "crawl" | "scrape";
@@ -56,6 +62,7 @@ export class FireCrawlLoader extends BaseDocumentLoader {
     super();
     const {
       apiKey = getEnvironmentVariable("FIRECRAWL_API_KEY"),
+      apiUrl,
       url,
       mode = "crawl",
       params,
@@ -67,18 +74,25 @@ export class FireCrawlLoader extends BaseDocumentLoader {
     }
 
     this.apiKey = apiKey;
+    this.apiUrl = apiUrl;
     this.url = url;
     this.mode = mode;
     this.params = params;
   }
 
   /**
-   * Loads the data from the Firecrawl.
+   * Loads data from Firecrawl.
    * @returns An array of Documents representing the retrieved data.
    * @throws An error if the data could not be loaded.
    */
   public async load(): Promise<DocumentInterface[]> {
-    const app = new FirecrawlApp({ apiKey: this.apiKey });
+    const params: ConstructorParameters<typeof FirecrawlApp>[0] = {
+      apiKey: this.apiKey,
+    };
+    if (this.apiUrl !== undefined) {
+      params.apiUrl = this.apiUrl;
+    }
+    const app = new FirecrawlApp(params);
     let firecrawlDocs: FirecrawlDocument[];
 
     if (this.mode === "scrape") {

diff --git a/yarn.lock b/yarn.lock
@@ -11185,7 +11185,7 @@ __metadata:
     "@langchain/scripts": ~0.0.20
     "@langchain/standard-tests": 0.0.0
     "@layerup/layerup-security": ^1.5.12
-    "@mendable/firecrawl-js": ^0.0.13
+    "@mendable/firecrawl-js": ^0.0.36
     "@mlc-ai/web-llm": 0.2.46
     "@mozilla/readability": ^0.4.4
     "@neondatabase/serverless": ^0.9.1
@@ -12704,6 +12704,19 @@ __metadata:
   languageName: node
   linkType: hard
 
+"@mendable/firecrawl-js@npm:^0.0.36":
+  version: 0.0.36
+  resolution: "@mendable/firecrawl-js@npm:0.0.36"
+  dependencies:
+    axios: ^1.6.8
+    dotenv: ^16.4.5
+    uuid: ^9.0.1
+    zod: ^3.23.8
+    zod-to-json-schema: ^3.23.0
+  checksum: 93ac8a7d9d25c04d4f618e282c136af06cf7712ec3402922531094c3cdab0e59d6f484a7f583022032eb58f914a0494193f2fd22986edd0f6712a29545edf95a
+  languageName: node
+  linkType: hard
+
 "@microsoft/fetch-event-source@npm:^2.0.1":
   version: 2.0.1
   resolution: "@microsoft/fetch-event-source@npm:2.0.1"