Merge pull request #34 from mendableai/nsc/returnOnlyUrls

Implements the ability for the crawler to output all the links it found, without scraping

apps/api/requests.http

@@ -13,13 +13,22 @@ GET http://localhost:3002/v0/jobs/active HTTP/1.1
 
 
 ### Scrape Website
-POST https://api.firecrawl.dev/v0/scrape HTTP/1.1
+POST http://localhost:3002/v0/crawl HTTP/1.1
 Authorization: Bearer 
 content-type: application/json
 
 {
-    "url":"https://www.mendable.ai"
+    "url":"https://www.mendable.ai",
+    "crawlerOptions": {
+        "returnOnlyUrls": true
     }
+}
+
+
+
+
+
+
 
 
 ### Scrape Website
@@ -34,7 +43,7 @@ content-type: application/json
 
 
 ### Check Job Status
-GET http://localhost:3002/v0/crawl/status/333ab225-dc3e-418b-9d4b-8fb833cbaf89 HTTP/1.1
+GET http://localhost:3002/v0/crawl/status/a6053912-d602-4709-841f-3d2cb46fea0a HTTP/1.1
 Authorization: Bearer 
 
 ### Get Job Result

apps/api/src/lib/entities.ts

@@ -38,6 +38,10 @@ export type WebScraperOptions = {
   concurrentRequests?: number;
 };
 
+export interface DocumentUrl {
+  url: string;
+}
+
 export class Document {
   id?: string;
   url?: string; // Used only in /search for now

apps/api/src/main/runWebScraper.ts

@@ -1,9 +1,10 @@
 import { Job } from "bull";
 import { CrawlResult, WebScraperOptions } from "../types";
 import { WebScraperDataProvider } from "../scraper/WebScraper";
-import { Progress } from "../lib/entities";
+import { DocumentUrl, Progress } from "../lib/entities";
 import { billTeam } from "../services/billing/credit_billing";
 import { Document } from "../lib/entities";
+
 export async function startWebScraperPipeline({
   job,
 }: {
@@ -47,7 +48,7 @@ export async function runWebScraper({
 }): Promise<{
   success: boolean;
   message: string;
-  docs: CrawlResult[];
+  docs: Document[] | DocumentUrl[];
 }> {
   try {
     const provider = new WebScraperDataProvider();
@@ -68,7 +69,7 @@ export async function runWebScraper({
     }
     const docs = (await provider.getDocuments(false, (progress: Progress) => {
       inProgress(progress);
-    })) as CrawlResult[];
+    })) as Document[];
 
     if (docs.length === 0) {
       return {
@@ -79,7 +80,14 @@ export async function runWebScraper({
     }
 
     // remove docs with empty content
-    const filteredDocs = docs.filter((doc) => doc.content.trim().length > 0);
+    const filteredDocs = crawlerOptions.returnOnlyUrls
+      ? docs.map((doc) => {
+          if (doc.metadata.sourceURL) {
+            return { url: doc.metadata.sourceURL };
+          }
+        })
+      : docs.filter((doc) => doc.content.trim().length > 0);
+
 
     const { success, credit_usage } = await billTeam(
       team_id,

apps/api/src/scraper/WebScraper/index.ts

@@ -80,11 +80,16 @@ export class WebScraperDataProvider {
         });
         let links = await crawler.start(inProgress, 5, this.limit);
         if (this.returnOnlyUrls) {
+          inProgress({
+            current: links.length,
+            total: links.length,
+            status: "COMPLETED",
+            currentDocumentUrl: this.urls[0],
+          });
           return links.map((url) => ({
             content: "",
+            markdown: "",
             metadata: { sourceURL: url },
-            provider: "web",
-            type: "text",
           }));
         }