firecrawl 1.9.0 → 1.9.2

package/dist/index.cjs

@@ -297,7 +297,21 @@ var FirecrawlApp = class {
    */
   async batchScrapeUrls(urls, params, pollInterval = 2, idempotencyKey, webhook) {
     const headers = this.prepareHeaders(idempotencyKey);
-    let jsonData = { urls, ...params ?? {}, webhook };
+    let jsonData = { urls, ...params };
+    if (jsonData?.extract?.schema) {
+      let schema = jsonData.extract.schema;
+      try {
+        schema = (0, import_zod_to_json_schema.zodToJsonSchema)(schema);
+      } catch (error) {
+      }
+      jsonData = {
+        ...jsonData,
+        extract: {
+          ...jsonData.extract,
+          schema
+        }
+      };
+    }
     try {
       const response = await this.postRequest(
         this.apiUrl + `/v1/batch/scrape`,
@@ -407,6 +421,7 @@ var FirecrawlApp = class {
   }
   /**
    * Extracts information from URLs using the Firecrawl API.
+   * Currently in Beta. Expect breaking changes on future minor versions.
    * @param url - The URL to extract information from.
    * @param params - Additional parameters for the extract request.
    * @returns The response from the extract operation.
@@ -430,7 +445,17 @@ var FirecrawlApp = class {
         headers
       );
       if (response.status === 200) {
-        return response.data;
+        const responseData = response.data;
+        if (responseData.success) {
+          return {
+            success: true,
+            data: responseData.data,
+            warning: responseData.warning,
+            error: responseData.error
+          };
+        } else {
+          throw new FirecrawlError(`Failed to scrape URL. Error: ${responseData.error}`, response.status);
+        }
       } else {
         this.handleError(response, "extract");
       }

package/dist/index.d.cts

@@ -223,9 +223,9 @@ interface MapResponse {
  * Parameters for extracting information from URLs.
  * Defines options for extracting information from URLs.
  */
-interface ExtractParams {
+interface ExtractParams<LLMSchema extends zt.ZodSchema = any> {
     prompt: string;
-    schema?: zt.ZodSchema;
+    schema?: LLMSchema;
     systemPrompt?: string;
     allowExternalLinks?: boolean;
 }
@@ -233,10 +233,11 @@ interface ExtractParams {
  * Response interface for extracting information from URLs.
  * Defines the structure of the response received after extracting information from URLs.
  */
-interface ExtractResponse {
-    success: true;
-    data: zt.infer<zt.ZodSchema>;
+interface ExtractResponse<LLMSchema extends zt.ZodSchema = any> {
+    success: boolean;
+    data: LLMSchema;
     error?: string;
+    warning?: string;
 }
 /**
  * Error response interface.
@@ -346,11 +347,12 @@ declare class FirecrawlApp {
     checkBatchScrapeStatus(id?: string, getAllData?: boolean): Promise<BatchScrapeStatusResponse | ErrorResponse>;
     /**
      * Extracts information from URLs using the Firecrawl API.
+     * Currently in Beta. Expect breaking changes on future minor versions.
      * @param url - The URL to extract information from.
      * @param params - Additional parameters for the extract request.
      * @returns The response from the extract operation.
      */
-    extract(urls: string[], params?: ExtractParams): Promise<ExtractResponse | ErrorResponse>;
+    extract<T extends zt.ZodSchema = any>(urls: string[], params?: ExtractParams<T>): Promise<ExtractResponse<zt.infer<T>> | ErrorResponse>;
     /**
      * Prepares the headers for an API request.
      * @param idempotencyKey - Optional key to ensure idempotency.

package/dist/index.d.ts

@@ -223,9 +223,9 @@ interface MapResponse {
  * Parameters for extracting information from URLs.
  * Defines options for extracting information from URLs.
  */
-interface ExtractParams {
+interface ExtractParams<LLMSchema extends zt.ZodSchema = any> {
     prompt: string;
-    schema?: zt.ZodSchema;
+    schema?: LLMSchema;
     systemPrompt?: string;
     allowExternalLinks?: boolean;
 }
@@ -233,10 +233,11 @@ interface ExtractParams {
  * Response interface for extracting information from URLs.
  * Defines the structure of the response received after extracting information from URLs.
  */
-interface ExtractResponse {
-    success: true;
-    data: zt.infer<zt.ZodSchema>;
+interface ExtractResponse<LLMSchema extends zt.ZodSchema = any> {
+    success: boolean;
+    data: LLMSchema;
     error?: string;
+    warning?: string;
 }
 /**
  * Error response interface.
@@ -346,11 +347,12 @@ declare class FirecrawlApp {
     checkBatchScrapeStatus(id?: string, getAllData?: boolean): Promise<BatchScrapeStatusResponse | ErrorResponse>;
     /**
      * Extracts information from URLs using the Firecrawl API.
+     * Currently in Beta. Expect breaking changes on future minor versions.
      * @param url - The URL to extract information from.
      * @param params - Additional parameters for the extract request.
      * @returns The response from the extract operation.
      */
-    extract(urls: string[], params?: ExtractParams): Promise<ExtractResponse | ErrorResponse>;
+    extract<T extends zt.ZodSchema = any>(urls: string[], params?: ExtractParams<T>): Promise<ExtractResponse<zt.infer<T>> | ErrorResponse>;
     /**
      * Prepares the headers for an API request.
      * @param idempotencyKey - Optional key to ensure idempotency.

package/dist/index.js

@@ -261,7 +261,21 @@ var FirecrawlApp = class {
    */
   async batchScrapeUrls(urls, params, pollInterval = 2, idempotencyKey, webhook) {
     const headers = this.prepareHeaders(idempotencyKey);
-    let jsonData = { urls, ...params ?? {}, webhook };
+    let jsonData = { urls, ...params };
+    if (jsonData?.extract?.schema) {
+      let schema = jsonData.extract.schema;
+      try {
+        schema = zodToJsonSchema(schema);
+      } catch (error) {
+      }
+      jsonData = {
+        ...jsonData,
+        extract: {
+          ...jsonData.extract,
+          schema
+        }
+      };
+    }
     try {
       const response = await this.postRequest(
         this.apiUrl + `/v1/batch/scrape`,
@@ -371,6 +385,7 @@ var FirecrawlApp = class {
   }
   /**
    * Extracts information from URLs using the Firecrawl API.
+   * Currently in Beta. Expect breaking changes on future minor versions.
    * @param url - The URL to extract information from.
    * @param params - Additional parameters for the extract request.
    * @returns The response from the extract operation.
@@ -394,7 +409,17 @@ var FirecrawlApp = class {
         headers
       );
       if (response.status === 200) {
-        return response.data;
+        const responseData = response.data;
+        if (responseData.success) {
+          return {
+            success: true,
+            data: responseData.data,
+            warning: responseData.warning,
+            error: responseData.error
+          };
+        } else {
+          throw new FirecrawlError(`Failed to scrape URL. Error: ${responseData.error}`, response.status);
+        }
       } else {
         this.handleError(response, "extract");
       }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "firecrawl",
-  "version": "1.9.0",
+  "version": "1.9.2",
   "description": "JavaScript SDK for Firecrawl API",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",

package/src/index.ts

@@ -240,9 +240,9 @@ export interface MapResponse {
  * Parameters for extracting information from URLs.
  * Defines options for extracting information from URLs.
  */
-export interface ExtractParams {
+export interface ExtractParams<LLMSchema extends zt.ZodSchema = any> {
   prompt: string;
-  schema?: zt.ZodSchema;
+  schema?: LLMSchema;
   systemPrompt?: string;
   allowExternalLinks?: boolean;
 }
@@ -251,10 +251,11 @@ export interface ExtractParams {
  * Response interface for extracting information from URLs.
  * Defines the structure of the response received after extracting information from URLs.
  */
-export interface ExtractResponse {
-  success: true;
-  data: zt.infer<zt.ZodSchema>;
+export interface ExtractResponse<LLMSchema extends zt.ZodSchema = any> {
+  success: boolean;
+  data: LLMSchema;
   error?: string;
+  warning?: string;
 }
 
 /**
@@ -576,7 +577,24 @@ export default class FirecrawlApp {
     webhook?: CrawlParams["webhook"],
   ): Promise<BatchScrapeStatusResponse | ErrorResponse> {
     const headers = this.prepareHeaders(idempotencyKey);
-    let jsonData: any = { urls, ...(params ?? {}), webhook };
+    let jsonData: any = { urls, ...params };
+    if (jsonData?.extract?.schema) {
+      let schema = jsonData.extract.schema;
+
+      // Try parsing the schema as a Zod schema
+      try {
+        schema = zodToJsonSchema(schema);
+      } catch (error) {
+        
+      }
+      jsonData = {
+        ...jsonData,
+        extract: {
+          ...jsonData.extract,
+          schema: schema,
+        },
+      };
+    }
     try {
       const response: AxiosResponse = await this.postRequest(
         this.apiUrl + `/v1/batch/scrape`,
@@ -701,18 +719,19 @@ export default class FirecrawlApp {
 
   /**
    * Extracts information from URLs using the Firecrawl API.
+   * Currently in Beta. Expect breaking changes on future minor versions.
    * @param url - The URL to extract information from.
    * @param params - Additional parameters for the extract request.
    * @returns The response from the extract operation.
    */
-  async extract(urls: string[], params?: ExtractParams): Promise<ExtractResponse | ErrorResponse> {
+  async extract<T extends zt.ZodSchema = any>(urls: string[], params?: ExtractParams<T>): Promise<ExtractResponse<zt.infer<T>> | ErrorResponse> {
     const headers = this.prepareHeaders();
 
     if (!params?.prompt) {
       throw new FirecrawlError("Prompt is required", 400);
     }
 
-    let jsonData: { urls: string[] } & ExtractParams= { urls,  ...params };
+    let jsonData: { urls: string[] } & ExtractParams<T> = { urls,  ...params };
     let jsonSchema: any;
     try {
       jsonSchema = params?.schema ? zodToJsonSchema(params.schema) : undefined;
@@ -727,7 +746,17 @@ export default class FirecrawlApp {
         headers
       );
       if (response.status === 200) {
-        return response.data as ExtractResponse;
+        const responseData = response.data as ExtractResponse<T>;
+        if (responseData.success) {
+          return {
+            success: true,
+            data: responseData.data,
+            warning: responseData.warning,
+            error: responseData.error
+          };
+        } else {
+          throw new FirecrawlError(`Failed to scrape URL. Error: ${responseData.error}`, response.status);
+        }
       } else {
         this.handleError(response, "extract");
       }