firecrawl 1.8.5 → 1.9.1

package/dist/index.cjs

@@ -405,6 +405,51 @@ var FirecrawlApp = class {
     }
     return { success: false, error: "Internal server error." };
   }
+  /**
+   * 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, params) {
+    const headers = this.prepareHeaders();
+    if (!params?.prompt) {
+      throw new FirecrawlError("Prompt is required", 400);
+    }
+    let jsonData = { urls, ...params };
+    let jsonSchema;
+    try {
+      jsonSchema = params?.schema ? (0, import_zod_to_json_schema.zodToJsonSchema)(params.schema) : void 0;
+    } catch (error) {
+      throw new FirecrawlError("Invalid schema. Use a valid Zod schema.", 400);
+    }
+    try {
+      const response = await this.postRequest(
+        this.apiUrl + `/v1/extract`,
+        { ...jsonData, schema: jsonSchema },
+        headers
+      );
+      if (response.status === 200) {
+        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");
+      }
+    } catch (error) {
+      throw new FirecrawlError(error.message, 500);
+    }
+    return { success: false, error: "Internal server error." };
+  }
   /**
    * Prepares the headers for an API request.
    * @param idempotencyKey - Optional key to ensure idempotency.

package/dist/index.d.cts

@@ -219,6 +219,26 @@ interface MapResponse {
     links?: string[];
     error?: string;
 }
+/**
+ * Parameters for extracting information from URLs.
+ * Defines options for extracting information from URLs.
+ */
+interface ExtractParams<LLMSchema extends zt.ZodSchema = any> {
+    prompt: string;
+    schema?: LLMSchema;
+    systemPrompt?: string;
+    allowExternalLinks?: boolean;
+}
+/**
+ * Response interface for extracting information from URLs.
+ * Defines the structure of the response received after extracting information from URLs.
+ */
+interface ExtractResponse<LLMSchema extends zt.ZodSchema = any> {
+    success: boolean;
+    data: LLMSchema;
+    error?: string;
+    warning?: string;
+}
 /**
  * Error response interface.
  * Defines the structure of the response received when an error occurs.
@@ -325,6 +345,14 @@ declare class FirecrawlApp {
      * @returns The response containing the job status.
      */
     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<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.
@@ -389,4 +417,4 @@ declare class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
     close(): void;
 }
 
-export { type Action, type ActionsResult, type BatchScrapeResponse, type BatchScrapeStatusResponse, type CrawlParams, type CrawlResponse, type CrawlScrapeOptions, type CrawlStatusResponse, CrawlWatcher, type ErrorResponse, type FirecrawlAppConfig, type FirecrawlDocument, type FirecrawlDocumentMetadata, FirecrawlError, type MapParams, type MapResponse, type ScrapeParams, type ScrapeResponse, FirecrawlApp as default };
+export { type Action, type ActionsResult, type BatchScrapeResponse, type BatchScrapeStatusResponse, type CrawlParams, type CrawlResponse, type CrawlScrapeOptions, type CrawlStatusResponse, CrawlWatcher, type ErrorResponse, type ExtractParams, type ExtractResponse, type FirecrawlAppConfig, type FirecrawlDocument, type FirecrawlDocumentMetadata, FirecrawlError, type MapParams, type MapResponse, type ScrapeParams, type ScrapeResponse, FirecrawlApp as default };

package/dist/index.d.ts

@@ -219,6 +219,26 @@ interface MapResponse {
     links?: string[];
     error?: string;
 }
+/**
+ * Parameters for extracting information from URLs.
+ * Defines options for extracting information from URLs.
+ */
+interface ExtractParams<LLMSchema extends zt.ZodSchema = any> {
+    prompt: string;
+    schema?: LLMSchema;
+    systemPrompt?: string;
+    allowExternalLinks?: boolean;
+}
+/**
+ * Response interface for extracting information from URLs.
+ * Defines the structure of the response received after extracting information from URLs.
+ */
+interface ExtractResponse<LLMSchema extends zt.ZodSchema = any> {
+    success: boolean;
+    data: LLMSchema;
+    error?: string;
+    warning?: string;
+}
 /**
  * Error response interface.
  * Defines the structure of the response received when an error occurs.
@@ -325,6 +345,14 @@ declare class FirecrawlApp {
      * @returns The response containing the job status.
      */
     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<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.
@@ -389,4 +417,4 @@ declare class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
     close(): void;
 }
 
-export { type Action, type ActionsResult, type BatchScrapeResponse, type BatchScrapeStatusResponse, type CrawlParams, type CrawlResponse, type CrawlScrapeOptions, type CrawlStatusResponse, CrawlWatcher, type ErrorResponse, type FirecrawlAppConfig, type FirecrawlDocument, type FirecrawlDocumentMetadata, FirecrawlError, type MapParams, type MapResponse, type ScrapeParams, type ScrapeResponse, FirecrawlApp as default };
+export { type Action, type ActionsResult, type BatchScrapeResponse, type BatchScrapeStatusResponse, type CrawlParams, type CrawlResponse, type CrawlScrapeOptions, type CrawlStatusResponse, CrawlWatcher, type ErrorResponse, type ExtractParams, type ExtractResponse, type FirecrawlAppConfig, type FirecrawlDocument, type FirecrawlDocumentMetadata, FirecrawlError, type MapParams, type MapResponse, type ScrapeParams, type ScrapeResponse, FirecrawlApp as default };

package/dist/index.js

@@ -369,6 +369,51 @@ var FirecrawlApp = class {
     }
     return { success: false, error: "Internal server error." };
   }
+  /**
+   * 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, params) {
+    const headers = this.prepareHeaders();
+    if (!params?.prompt) {
+      throw new FirecrawlError("Prompt is required", 400);
+    }
+    let jsonData = { urls, ...params };
+    let jsonSchema;
+    try {
+      jsonSchema = params?.schema ? zodToJsonSchema(params.schema) : void 0;
+    } catch (error) {
+      throw new FirecrawlError("Invalid schema. Use a valid Zod schema.", 400);
+    }
+    try {
+      const response = await this.postRequest(
+        this.apiUrl + `/v1/extract`,
+        { ...jsonData, schema: jsonSchema },
+        headers
+      );
+      if (response.status === 200) {
+        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");
+      }
+    } catch (error) {
+      throw new FirecrawlError(error.message, 500);
+    }
+    return { success: false, error: "Internal server error." };
+  }
   /**
    * Prepares the headers for an API request.
    * @param idempotencyKey - Optional key to ensure idempotency.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "firecrawl",
-  "version": "1.8.5",
+  "version": "1.9.1",
   "description": "JavaScript SDK for Firecrawl API",
   "main": "dist/index.js",
   "types": "dist/index.d.ts",

package/src/index.ts

@@ -236,6 +236,28 @@ export interface MapResponse {
   error?: string;
 }
 
+/**
+ * Parameters for extracting information from URLs.
+ * Defines options for extracting information from URLs.
+ */
+export interface ExtractParams<LLMSchema extends zt.ZodSchema = any> {
+  prompt: string;
+  schema?: LLMSchema;
+  systemPrompt?: string;
+  allowExternalLinks?: boolean;
+}
+
+/**
+ * Response interface for extracting information from URLs.
+ * Defines the structure of the response received after extracting information from URLs.
+ */
+export interface ExtractResponse<LLMSchema extends zt.ZodSchema = any> {
+  success: boolean;
+  data: LLMSchema;
+  error?: string;
+  warning?: string;
+}
+
 /**
  * Error response interface.
  * Defines the structure of the response received when an error occurs.
@@ -245,7 +267,6 @@ export interface ErrorResponse {
   error: string;
 }
 
-
 /**
  * Custom error class for Firecrawl.
  * Extends the built-in Error class to include a status code.
@@ -679,6 +700,55 @@ export default class FirecrawlApp {
     return { success: false, error: "Internal server error." };
   }
 
+  /**
+   * 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<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<T> = { urls,  ...params };
+    let jsonSchema: any;
+    try {
+      jsonSchema = params?.schema ? zodToJsonSchema(params.schema) : undefined;
+    } catch (error: any) {
+      throw new FirecrawlError("Invalid schema. Use a valid Zod schema.", 400);
+    }
+
+    try {
+      const response: AxiosResponse = await this.postRequest(
+        this.apiUrl + `/v1/extract`,
+        { ...jsonData, schema: jsonSchema },
+        headers
+      );
+      if (response.status === 200) {
+        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");
+      }
+    } catch (error: any) {
+      throw new FirecrawlError(error.message, 500);
+    }
+    return { success: false, error: "Internal server error." };
+  }
+
   /**
    * Prepares the headers for an API request.
    * @param idempotencyKey - Optional key to ensure idempotency.