Merge branch 'main' into py-sdk-improve-response-handling

.github/workflows/clean-before-24h-complete-jobs.yml

@@ -0,0 +1,20 @@
+name: Clean Before 24h Completed Jobs
+on:
+  schedule:
+    - cron: '0 0 * * *'
+
+env:
+  BULL_AUTH_KEY: ${{ secrets.BULL_AUTH_KEY }}
+    
+jobs:
+  clean-jobs:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Send GET request to clean jobs
+        run: |
+          response=$(curl --write-out '%{http_code}' --silent --output /dev/null https://api.firecrawl.dev/admin/${{ secrets.BULL_AUTH_KEY }}/clean-before-24h-complete-jobs)
+          if [ "$response" -ne 200 ]; then
+            echo "Failed to clean jobs. Response: $response"
+            exit 1
+          fi
+          echo "Successfully cleaned jobs. Response: $response"

.github/workflows/fly.yml

@@ -183,6 +183,7 @@ jobs:
           FLY_API_TOKEN: ${{ secrets.FLY_API_TOKEN }}
 
   build-and-publish-python-sdk:
+    name: Build and publish Python SDK
     runs-on: ubuntu-latest
     needs: deploy
 
@@ -213,7 +214,7 @@ jobs:
         working-directory: ./apps/python-sdk
 
       - name: Publish to PyPI
-        if: ${{ env.VERSION_INCREMENTED == 'true' }}
+        if: ${{ env.PYTHON_SDK_VERSION_INCREMENTED == 'true' }}
         env:
           TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
           TWINE_PASSWORD: ${{ secrets.PYPI_PASSWORD }}
@@ -222,6 +223,7 @@ jobs:
         working-directory: ./apps/python-sdk
 
   build-and-publish-js-sdk:
+    name: Build and publish JavaScript SDK
     runs-on: ubuntu-latest
     needs: deploy
 

apps/api/fly.toml

@@ -54,7 +54,7 @@ kill_timeout = '5s'
     soft_limit = 12
 
 [[vm]]
-  size = 'performance-8x'
+  size = 'performance-4x'
   processes = ['app']
 
 

apps/api/openapi.json

@@ -51,10 +51,26 @@
                         "description": "Include the raw HTML content of the page. Will output a html key in the response.",
                         "default": false
                       },
+                      "screenshot": {
+                        "type": "boolean",
+                        "description": "Include a screenshot of the top of the page that you are scraping.",
+                        "default": false
+                      },
                       "waitFor": {
                         "type": "integer",
                         "description": "Wait x amount of milliseconds for the page to load to fetch content",
                         "default": 0
+                      },
+                      "removeTags": {
+                        "type": "array",
+                        "items": {
+                          "type": "string"
+                        },
+                        "description": "Tags, classes and ids to remove from the page. Use comma separated values. Example: 'script, .ad, #footer'"
+                      },
+                      "headers": {
+                        "type": "object",
+                        "description": "Headers to send with the request. Can be used to send cookies, user-agent, etc."
                       }
                     }
                   },
@@ -176,10 +192,20 @@
                         "description": "The crawling mode to use. Fast mode crawls 4x faster websites without sitemap, but may not be as accurate and shouldn't be used in heavy js-rendered websites.",
                         "default": "default"
                       },
+                      "ignoreSitemap": {
+                        "type": "boolean",
+                        "description": "Ignore the website sitemap when crawling",
+                        "default": false
+                      },
                       "limit": {
                         "type": "integer",
                         "description": "Maximum number of pages to crawl",
                         "default": 10000
+                      },
+                      "allowBackwardCrawling": {
+                        "type": "boolean",
+                        "description": "Allow backward crawling (crawl from the base URL to the previous URLs)",
+                        "default": false
                       }
                     }
                   },
@@ -195,6 +221,27 @@
                         "type": "boolean",
                         "description": "Include the raw HTML content of the page. Will output a html key in the response.",
                         "default": false
+                      },
+                      "screenshot": {
+                        "type": "boolean",
+                        "description": "Include a screenshot of the top of the page that you are scraping.",
+                        "default": false
+                      },
+                      "headers": {
+                        "type": "object",
+                        "description": "Headers to send with the request when scraping. Can be used to send cookies, user-agent, etc."
+                      },
+                      "removeTags": {
+                        "type": "array",
+                        "items": {
+                          "type": "string"
+                        },
+                        "description": "Tags, classes and ids to remove from the page. Use comma separated values. Example: 'script, .ad, #footer'"
+                      },
+                      "replaceAllPathsWithAbsolutePaths": {
+                        "type": "boolean",
+                        "description": "Replace all relative paths with absolute paths for images and links",
+                        "default": false
                       }
                     }
                   }
@@ -368,7 +415,7 @@
                       "items": {
                         "$ref": "#/components/schemas/CrawlStatusResponseObj"
                       },
-                      "description": "Partial documents returned as it is being crawls (streaming). When a page is ready it will append to the parial_data array - so no need to wait for all the website to be crawled."
+                      "description": "Partial documents returned as it is being crawled (streaming). **This feature is currently in alpha - expect breaking changes** When a page is ready, it will append to the partial_data array, so there is no need to wait for the entire website to be crawled. There is a max of 50 items in the array response. The oldest item (top of the array) will be removed when the new item is added to the array."
                     }
                   }
                 }
@@ -513,6 +560,10 @@
             "nullable": true,
             "description": "Raw HTML content of the page if `includeHtml`  is true"
           },
+          "index": {
+            "type": "integer",
+            "description": "The number of the page that was crawled. This is useful for `partial_data` so you know which page the data is from." 
+          },
           "metadata": {
             "type": "object",
             "properties": {

apps/api/src/__tests__/e2e_noAuth/index.test.ts

@@ -1,5 +1,4 @@
 import request from "supertest";
-import { app } from "../../index";
 import dotenv from "dotenv";
 const fs = require("fs");
 const path = require("path");

apps/api/src/__tests__/e2e_withAuth/index.test.ts

@@ -1,5 +1,4 @@
 import request from "supertest";
-import { app } from "../../index";
 import dotenv from "dotenv";
 import { v4 as uuidv4 } from "uuid";
 
@@ -35,7 +34,7 @@ describe("E2E Tests for API Routes", () => {
 
   describe("POST /v0/scrape", () => {
     it.concurrent("should require authorization", async () => {
-      const response = await request(app).post("/v0/scrape");
+      const response = await request(TEST_URL).post("/v0/scrape");
       expect(response.statusCode).toBe(401);
     });
 
@@ -136,6 +135,40 @@ describe("E2E Tests for API Routes", () => {
       expect(response.body.data.content).toContain('We present spectrophotometric observations of the Broad Line Radio Galaxy');
     }, 60000); // 60 seconds
 
+    it.concurrent("should return a successful response with a valid API key with removeTags option", async () => {
+      const responseWithoutRemoveTags = await request(TEST_URL)
+        .post("/v0/scrape")
+        .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`)
+        .set("Content-Type", "application/json")
+        .send({ url: "https://www.scrapethissite.com/" });
+      expect(responseWithoutRemoveTags.statusCode).toBe(200);
+      expect(responseWithoutRemoveTags.body).toHaveProperty("data");
+      expect(responseWithoutRemoveTags.body.data).toHaveProperty("content");
+      expect(responseWithoutRemoveTags.body.data).toHaveProperty("markdown");
+      expect(responseWithoutRemoveTags.body.data).toHaveProperty("metadata");
+      expect(responseWithoutRemoveTags.body.data).not.toHaveProperty("html");
+      expect(responseWithoutRemoveTags.body.data.content).toContain("Scrape This Site");
+      expect(responseWithoutRemoveTags.body.data.content).toContain("Lessons and Videos"); // #footer
+      expect(responseWithoutRemoveTags.body.data.content).toContain("[Sandbox]("); // .nav
+      expect(responseWithoutRemoveTags.body.data.content).toContain("web scraping"); // strong
+
+      const response = await request(TEST_URL)
+        .post("/v0/scrape")
+        .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`)
+        .set("Content-Type", "application/json")
+        .send({ url: "https://www.scrapethissite.com/", pageOptions: { removeTags: ['.nav', '#footer', 'strong'] } });
+      expect(response.statusCode).toBe(200);
+      expect(response.body).toHaveProperty("data");
+      expect(response.body.data).toHaveProperty("content");
+      expect(response.body.data).toHaveProperty("markdown");
+      expect(response.body.data).toHaveProperty("metadata");
+      expect(response.body.data).not.toHaveProperty("html");
+      expect(response.body.data.content).toContain("Scrape This Site");
+      expect(response.body.data.content).not.toContain("Lessons and Videos"); // #footer
+      expect(response.body.data.content).not.toContain("[Sandbox]("); // .nav
+      expect(response.body.data.content).not.toContain("web scraping"); // strong
+    }, 30000); // 30 seconds timeout
+
     // TODO: add this test back once we nail the waitFor option to be more deterministic
     // it.concurrent("should return a successful response with a valid API key and waitFor option", async () => {
     //   const startTime = Date.now();
@@ -596,7 +629,7 @@ describe("E2E Tests for API Routes", () => {
         .post("/v0/crawl")
         .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`)
         .set("Content-Type", "application/json")
-        .send({ url: "https://roastmywebsite.ai" });
+        .send({ url: "https://mendable.ai/blog" });
       expect(crawlResponse.statusCode).toBe(200);
 
       let isCompleted = false;
@@ -622,7 +655,13 @@ describe("E2E Tests for API Routes", () => {
       expect(completedResponse.body.data[0]).toHaveProperty("content");
       expect(completedResponse.body.data[0]).toHaveProperty("markdown");
       expect(completedResponse.body.data[0]).toHaveProperty("metadata");
-      expect(completedResponse.body.data[0].content).toContain("_Roast_");
+      expect(completedResponse.body.data[0].content).toContain("Mendable");
+
+      const childrenLinks = completedResponse.body.data.filter(doc => 
+        doc.metadata && doc.metadata.sourceURL && doc.metadata.sourceURL.includes("mendable.ai/blog")
+      );
+
+      expect(childrenLinks.length).toBe(completedResponse.body.data.length);
     }, 120000); // 120 seconds
     
     it.concurrent('should return a successful response for a valid crawl job with PDF files without explicit .pdf extension', async () => {
@@ -757,34 +796,82 @@ describe("E2E Tests for API Routes", () => {
     }, 60000);
   }); // 60 seconds
 
+  it.concurrent("should return a successful response for a valid crawl job with allowBackwardCrawling set to true option", async () => {
+    const crawlResponse = await request(TEST_URL)
+      .post("/v0/crawl")
+      .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`)
+      .set("Content-Type", "application/json")
+      .send({
+        url: "https://mendable.ai/blog",
+        pageOptions: { includeHtml: true },
+        crawlerOptions: { allowBackwardCrawling: true },
+      });
+    expect(crawlResponse.statusCode).toBe(200);
+    
+    let isFinished = false;
+    let completedResponse;
+
+    while (!isFinished) {
+      const response = await request(TEST_URL)
+        .get(`/v0/crawl/status/${crawlResponse.body.jobId}`)
+        .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`);
+      expect(response.statusCode).toBe(200);
+      expect(response.body).toHaveProperty("status");
+
+      if (response.body.status === "completed") {
+        isFinished = true;
+        completedResponse = response;
+      } else {
+        await new Promise((r) => setTimeout(r, 1000)); // Wait for 1 second before checking again
+      }
+    }
+
+    expect(completedResponse.statusCode).toBe(200);
+    expect(completedResponse.body).toHaveProperty("status");
+    expect(completedResponse.body.status).toBe("completed");
+    expect(completedResponse.body).toHaveProperty("data");
+    expect(completedResponse.body.data[0]).toHaveProperty("content");
+    expect(completedResponse.body.data[0]).toHaveProperty("markdown");
+    expect(completedResponse.body.data[0]).toHaveProperty("metadata");
+    expect(completedResponse.body.data[0]).toHaveProperty("html");
+    expect(completedResponse.body.data[0].content).toContain("Mendable");
+    expect(completedResponse.body.data[0].markdown).toContain("Mendable");
+
+    const onlyChildrenLinks = completedResponse.body.data.filter(doc => {
+      return doc.metadata && doc.metadata.sourceURL && doc.metadata.sourceURL.includes("mendable.ai/blog")
+    });
+
+    expect(completedResponse.body.data.length).toBeGreaterThan(onlyChildrenLinks.length);
+  }, 60000);
+
   it.concurrent("If someone cancels a crawl job, it should turn into failed status", async () => {
     const crawlResponse = await request(TEST_URL)
       .post("/v0/crawl")
       .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`)
       .set("Content-Type", "application/json")
       .send({ url: "https://jestjs.io" });
+
     expect(crawlResponse.statusCode).toBe(200);
 
-    // wait for 30 seconds
     await new Promise((r) => setTimeout(r, 20000));
 
-    const response = await request(TEST_URL)
+    const responseCancel = await request(TEST_URL)
       .delete(`/v0/crawl/cancel/${crawlResponse.body.jobId}`)
       .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`);
-    expect(response.statusCode).toBe(200);
-    expect(response.body).toHaveProperty("status");
-    expect(response.body.status).toBe("cancelled");
+    expect(responseCancel.statusCode).toBe(200);
+    expect(responseCancel.body).toHaveProperty("status");
+    expect(responseCancel.body.status).toBe("cancelled");
 
     await new Promise((r) => setTimeout(r, 10000));
-
     const completedResponse = await request(TEST_URL)
       .get(`/v0/crawl/status/${crawlResponse.body.jobId}`)
       .set("Authorization", `Bearer ${process.env.TEST_API_KEY}`);
+
     expect(completedResponse.statusCode).toBe(200);
     expect(completedResponse.body).toHaveProperty("status");
     expect(completedResponse.body.status).toBe("failed");
     expect(completedResponse.body).toHaveProperty("data");
-    expect(completedResponse.body.data).toEqual(null);
+    expect(completedResponse.body.data).toBeNull();
     expect(completedResponse.body).toHaveProperty("partial_data");
     expect(completedResponse.body.partial_data[0]).toHaveProperty("content");
     expect(completedResponse.body.partial_data[0]).toHaveProperty("markdown");

apps/api/src/controllers/auth.ts

@@ -143,7 +143,7 @@ export async function supaAuthenticateUser(
     const startDate = new Date();
     const endDate = new Date();
     endDate.setDate(endDate.getDate() + 7);
-    await sendNotification(team_id, NotificationType.RATE_LIMIT_REACHED, startDate.toISOString(), endDate.toISOString());
+    // await sendNotification(team_id, NotificationType.RATE_LIMIT_REACHED, startDate.toISOString(), endDate.toISOString());
     return {
       success: false,
       error: `Rate limit exceeded. Consumed points: ${rateLimiterRes.consumedPoints}, Remaining points: ${rateLimiterRes.remainingPoints}. Upgrade your plan at https://firecrawl.dev/pricing for increased rate limits or please retry after ${secs}s, resets at ${retryDate}`,

apps/api/src/controllers/crawl.ts

@@ -55,8 +55,14 @@ export async function crawlController(req: Request, res: Response) {
     }
 
     const mode = req.body.mode ?? "crawl";
-    const crawlerOptions = req.body.crawlerOptions ?? {};
-    const pageOptions = req.body.pageOptions ?? { onlyMainContent: false, includeHtml: false };
+    const crawlerOptions = req.body.crawlerOptions ?? {
+      allowBackwardCrawling: false
+    };
+    const pageOptions = req.body.pageOptions ?? {
+      onlyMainContent: false,
+      includeHtml: false,
+      removeTags: []
+    };
 
     if (mode === "single_urls" && !url.includes(",")) {
       try {
@@ -64,9 +70,7 @@ export async function crawlController(req: Request, res: Response) {
         await a.setOptions({
           mode: "single_urls",
           urls: [url],
-          crawlerOptions: {
-            returnOnlyUrls: true,
-          },
+          crawlerOptions: { ...crawlerOptions, returnOnlyUrls: true },
           pageOptions: pageOptions,
         });
 
@@ -91,7 +95,7 @@ export async function crawlController(req: Request, res: Response) {
     const job = await addWebScraperJob({
       url: url,
       mode: mode ?? "crawl", // fix for single urls not working
-      crawlerOptions: { ...crawlerOptions },
+      crawlerOptions: crawlerOptions,
       team_id: team_id,
       pageOptions: pageOptions,
       origin: req.body.origin ?? "api",

apps/api/src/controllers/crawlPreview.ts

@@ -26,7 +26,7 @@ export async function crawlPreviewController(req: Request, res: Response) {
 
     const mode = req.body.mode ?? "crawl";
     const crawlerOptions = req.body.crawlerOptions ?? {};
-    const pageOptions = req.body.pageOptions ?? { onlyMainContent: false, includeHtml: false };
+    const pageOptions = req.body.pageOptions ?? { onlyMainContent: false, includeHtml: false, removeTags: [] };
 
     const job = await addWebScraperJob({
       url: url,

apps/api/src/controllers/search.ts

@@ -85,6 +85,7 @@ export async function searchHelper(
       onlyMainContent: pageOptions?.onlyMainContent ?? true,
       fetchPageContent: pageOptions?.fetchPageContent ?? true,
       includeHtml: pageOptions?.includeHtml ?? false,
+      removeTags: pageOptions?.removeTags ?? [],
       fallback: false,
     },
   });
@@ -139,6 +140,7 @@ export async function searchController(req: Request, res: Response) {
       includeHtml: false,
       onlyMainContent: true,
       fetchPageContent: true,
+      removeTags: [],
       fallback: false,
     };
     const origin = req.body.origin ?? "api";

apps/api/src/index.ts

@@ -5,169 +5,215 @@ import "dotenv/config";
 import { getWebScraperQueue } from "./services/queue-service";
 import { redisClient } from "./services/rate-limiter";
 import { v0Router } from "./routes/v0";
-import { initSDK } from '@hyperdx/node-opentelemetry';
+import { initSDK } from "@hyperdx/node-opentelemetry";
+import cluster from "cluster";
+import os from "os";
 
 const { createBullBoard } = require("@bull-board/api");
 const { BullAdapter } = require("@bull-board/api/bullAdapter");
 const { ExpressAdapter } = require("@bull-board/express");
 
-export const app = express();
+const numCPUs = process.env.ENV === "local" ? 2 : os.cpus().length;
+console.log(`Number of CPUs: ${numCPUs} available`);
 
-global.isProduction = process.env.IS_PRODUCTION === "true";
+if (cluster.isMaster) {
+  console.log(`Master ${process.pid} is running`);
 
-app.use(bodyParser.urlencoded({ extended: true }));
-app.use(bodyParser.json({ limit: "10mb" }));
+  // Fork workers.
+  for (let i = 0; i < numCPUs; i++) {
+    cluster.fork();
+  }
 
-app.use(cors()); // Add this line to enable CORS
-
-const serverAdapter = new ExpressAdapter();
-serverAdapter.setBasePath(`/admin/${process.env.BULL_AUTH_KEY}/queues`);
-
-const { addQueue, removeQueue, setQueues, replaceQueues } = createBullBoard({
-  queues: [new BullAdapter(getWebScraperQueue())],
-  serverAdapter: serverAdapter,
-});
-
-app.use(
-  `/admin/${process.env.BULL_AUTH_KEY}/queues`,
-  serverAdapter.getRouter()
-);
-
-app.get("/", (req, res) => {
-  res.send("SCRAPERS-JS: Hello, world! Fly.io");
-});
-
-//write a simple test function
-app.get("/test", async (req, res) => {
-  res.send("Hello, world!");
-});
-
-// register router
-app.use(v0Router);
-
-const DEFAULT_PORT = process.env.PORT ?? 3002;
-const HOST = process.env.HOST ?? "localhost";
-redisClient.connect();
-
-// HyperDX OpenTelemetry
-if(process.env.ENV === 'production') {
-  initSDK({ consoleCapture: true, additionalInstrumentations: []});
-}
-
-
-export function startServer(port = DEFAULT_PORT) {
-  const server = app.listen(Number(port), HOST, () => {
-    console.log(`Server listening on port ${port}`);
-    console.log(
-      `For the UI, open http://${HOST}:${port}/admin/${process.env.BULL_AUTH_KEY}/queues`
-    );
-    console.log("");
-    console.log("1. Make sure Redis is running on port 6379 by default");
-    console.log(
-      "2. If you want to run nango, make sure you do port forwarding in 3002 using ngrok http 3002 "
-    );
+  cluster.on("exit", (worker, code, signal) => {
+    console.log(`Worker ${worker.process.pid} exited`);
+    console.log("Starting a new worker");
+    cluster.fork();
   });
-  return server;
-}
+} else {
+  const app = express();
 
-if (require.main === module) {
-  startServer();
-}
+  global.isProduction = process.env.IS_PRODUCTION === "true";
 
-// Use this as a "health check" that way we dont destroy the server
-app.get(`/admin/${process.env.BULL_AUTH_KEY}/queues`, async (req, res) => {
-  try {
-    const webScraperQueue = getWebScraperQueue();
-    const [webScraperActive] = await Promise.all([
-      webScraperQueue.getActiveCount(),
-    ]);
+  app.use(bodyParser.urlencoded({ extended: true }));
+  app.use(bodyParser.json({ limit: "10mb" }));
 
-    const noActiveJobs = webScraperActive === 0;
-    // 200 if no active jobs, 503 if there are active jobs
-    return res.status(noActiveJobs ? 200 : 500).json({
-      webScraperActive,
-      noActiveJobs,
-    });
-  } catch (error) {
-    console.error(error);
-    return res.status(500).json({ error: error.message });
+  app.use(cors()); // Add this line to enable CORS
+
+  const serverAdapter = new ExpressAdapter();
+  serverAdapter.setBasePath(`/admin/${process.env.BULL_AUTH_KEY}/queues`);
+
+  const { addQueue, removeQueue, setQueues, replaceQueues } = createBullBoard({
+    queues: [new BullAdapter(getWebScraperQueue())],
+    serverAdapter: serverAdapter,
+  });
+
+  app.use(
+    `/admin/${process.env.BULL_AUTH_KEY}/queues`,
+    serverAdapter.getRouter()
+  );
+
+  app.get("/", (req, res) => {
+    res.send("SCRAPERS-JS: Hello, world! Fly.io");
+  });
+
+  //write a simple test function
+  app.get("/test", async (req, res) => {
+    res.send("Hello, world!");
+  });
+
+  // register router
+  app.use(v0Router);
+
+  const DEFAULT_PORT = process.env.PORT ?? 3002;
+  const HOST = process.env.HOST ?? "localhost";
+  redisClient.connect();
+
+  // HyperDX OpenTelemetry
+  if (process.env.ENV === "production") {
+    initSDK({ consoleCapture: true, additionalInstrumentations: [] });
   }
-});
 
-app.get(`/serverHealthCheck`, async (req, res) => {
-  try {
-    const webScraperQueue = getWebScraperQueue();
-    const [waitingJobs] = await Promise.all([
-      webScraperQueue.getWaitingCount(),
-    ]);
-
-    const noWaitingJobs = waitingJobs === 0;
-    // 200 if no active jobs, 503 if there are active jobs
-    return res.status(noWaitingJobs ? 200 : 500).json({
-      waitingJobs,
+  function startServer(port = DEFAULT_PORT) {
+    const server = app.listen(Number(port), HOST, () => {
+      console.log(`Worker ${process.pid} listening on port ${port}`);
+      console.log(
+        `For the UI, open http://${HOST}:${port}/admin/${process.env.BULL_AUTH_KEY}/queues`
+      );
+      console.log("");
+      console.log("1. Make sure Redis is running on port 6379 by default");
+      console.log(
+        "2. If you want to run nango, make sure you do port forwarding in 3002 using ngrok http 3002 "
+      );
     });
-  } catch (error) {
-    console.error(error);
-    return res.status(500).json({ error: error.message });
+    return server;
   }
-});
 
-app.get('/serverHealthCheck/notify', async (req, res) => {
-  if (process.env.SLACK_WEBHOOK_URL) {
-    const treshold = 1; // The treshold value for the active jobs
-    const timeout = 60000; // 1 minute // The timeout value for the check in milliseconds
+  if (require.main === module) {
+    startServer();
+  }
 
-    const getWaitingJobsCount = async () => {
+  // Use this as a "health check" that way we dont destroy the server
+  app.get(`/admin/${process.env.BULL_AUTH_KEY}/queues`, async (req, res) => {
+    try {
       const webScraperQueue = getWebScraperQueue();
-      const [waitingJobsCount] = await Promise.all([
+      const [webScraperActive] = await Promise.all([
+        webScraperQueue.getActiveCount(),
+      ]);
+
+      const noActiveJobs = webScraperActive === 0;
+      // 200 if no active jobs, 503 if there are active jobs
+      return res.status(noActiveJobs ? 200 : 500).json({
+        webScraperActive,
+        noActiveJobs,
+      });
+    } catch (error) {
+      console.error(error);
+      return res.status(500).json({ error: error.message });
+    }
+  });
+
+  app.get(`/serverHealthCheck`, async (req, res) => {
+    try {
+      const webScraperQueue = getWebScraperQueue();
+      const [waitingJobs] = await Promise.all([
         webScraperQueue.getWaitingCount(),
       ]);
 
-      return waitingJobsCount;
-    };
+      const noWaitingJobs = waitingJobs === 0;
+      // 200 if no active jobs, 503 if there are active jobs
+      return res.status(noWaitingJobs ? 200 : 500).json({
+        waitingJobs,
+      });
+    } catch (error) {
+      console.error(error);
+      return res.status(500).json({ error: error.message });
+    }
+  });
 
-    res.status(200).json({ message: "Check initiated" });
+  app.get("/serverHealthCheck/notify", async (req, res) => {
+    if (process.env.SLACK_WEBHOOK_URL) {
+      const treshold = 1; // The treshold value for the active jobs
+      const timeout = 60000; // 1 minute // The timeout value for the check in milliseconds
 
-    const checkWaitingJobs = async () => {
-      try {
-        let waitingJobsCount = await getWaitingJobsCount();
-        if (waitingJobsCount >= treshold) {
-          setTimeout(async () => {
-            // Re-check the waiting jobs count after the timeout
-            waitingJobsCount = await getWaitingJobsCount(); 
-            if (waitingJobsCount >= treshold) {
-              const slackWebhookUrl = process.env.SLACK_WEBHOOK_URL;
-              const message = {
-                text: `⚠️ Warning: The number of active jobs (${waitingJobsCount}) has exceeded the threshold (${treshold}) for more than ${timeout/60000} minute(s).`,
-              };
+      const getWaitingJobsCount = async () => {
+        const webScraperQueue = getWebScraperQueue();
+        const [waitingJobsCount] = await Promise.all([
+          webScraperQueue.getWaitingCount(),
+        ]);
 
-              const response = await fetch(slackWebhookUrl, {
-                method: 'POST',
-                headers: {
-                  'Content-Type': 'application/json',
-                },
-                body: JSON.stringify(message),
-              })
+        return waitingJobsCount;
+      };
 
-              if (!response.ok) {
-                console.error('Failed to send Slack notification')
+      res.status(200).json({ message: "Check initiated" });
+
+      const checkWaitingJobs = async () => {
+        try {
+          let waitingJobsCount = await getWaitingJobsCount();
+          if (waitingJobsCount >= treshold) {
+            setTimeout(async () => {
+              // Re-check the waiting jobs count after the timeout
+              waitingJobsCount = await getWaitingJobsCount();
+              if (waitingJobsCount >= treshold) {
+                const slackWebhookUrl = process.env.SLACK_WEBHOOK_URL;
+                const message = {
+                  text: `⚠️ Warning: The number of active jobs (${waitingJobsCount}) has exceeded the threshold (${treshold}) for more than ${
+                    timeout / 60000
+                  } minute(s).`,
+                };
+
+                const response = await fetch(slackWebhookUrl, {
+                  method: "POST",
+                  headers: {
+                    "Content-Type": "application/json",
+                  },
+                  body: JSON.stringify(message),
+                });
+
+                if (!response.ok) {
+                  console.error("Failed to send Slack notification");
+                }
               }
-            }
-          }, timeout);
+            }, timeout);
+          }
+        } catch (error) {
+          console.error(error);
         }
+      };
+
+      checkWaitingJobs();
+    }
+  });
+
+  app.get(
+    `/admin/${process.env.BULL_AUTH_KEY}/clean-before-24h-complete-jobs`,
+    async (req, res) => {
+      try {
+        const webScraperQueue = getWebScraperQueue();
+        const completedJobs = await webScraperQueue.getJobs(["completed"]);
+        const before24hJobs = completedJobs.filter(
+          (job) => job.finishedOn < Date.now() - 24 * 60 * 60 * 1000
+        );
+        const jobIds = before24hJobs.map((job) => job.id) as string[];
+        let count = 0;
+        for (const jobId of jobIds) {
+          try {
+            await webScraperQueue.removeJobs(jobId);
+            count++;
+          } catch (jobError) {
+            console.error(`Failed to remove job with ID ${jobId}:`, jobError);
+          }
+        }
+        res.status(200).send(`Removed ${count} completed jobs.`);
       } catch (error) {
-        console.error(error);
+        console.error("Failed to clean last 24h complete jobs:", error);
+        res.status(500).send("Failed to clean jobs");
       }
-    };
+    }
+  );
 
-    checkWaitingJobs();
-  }
-});
+  app.get("/is-production", (req, res) => {
+    res.send({ isProduction: global.isProduction });
+  });
 
-
-app.get("/is-production", (req, res) => {
-  res.send({ isProduction: global.isProduction });
-});
-
-
-// /workers health check, cant act as load balancer, just has to be a pre deploy thing
+  console.log(`Worker ${process.pid} started`);
+}

apps/api/src/lib/entities.ts

@@ -18,6 +18,8 @@ export type PageOptions = {
   waitFor?: number;
   screenshot?: boolean;
   headers?: Record<string, string>;
+  replaceAllPathsWithAbsolutePaths?: boolean;
+  removeTags?: string | string[];
 };
 
 export type ExtractorOptions = {
@@ -35,20 +37,24 @@ export type SearchOptions = {
   location?: string;
 };
 
+export type CrawlerOptions = {
+  returnOnlyUrls?: boolean;
+  includes?: string[];
+  excludes?: string[];
+  maxCrawledLinks?: number;
+  maxDepth?: number;
+  limit?: number;
+  generateImgAltText?: boolean;
+  replaceAllPathsWithAbsolutePaths?: boolean;
+  ignoreSitemap?: boolean;
+  mode?: "default" | "fast"; // have a mode of some sort
+  allowBackwardCrawling?: boolean;
+}
+
 export type WebScraperOptions = {
   urls: string[];
   mode: "single_urls" | "sitemap" | "crawl";
-  crawlerOptions?: {
-    returnOnlyUrls?: boolean;
-    includes?: string[];
-    excludes?: string[];
-    maxCrawledLinks?: number;
-    maxDepth?: number;
-    limit?: number;
-    generateImgAltText?: boolean;
-    replaceAllPathsWithAbsolutePaths?: boolean;
-    mode?: "default" | "fast"; // have a mode of some sort
-  };
+  crawlerOptions?: CrawlerOptions;
   pageOptions?: PageOptions;
   extractorOptions?: ExtractorOptions;
   concurrentRequests?: number;

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

@@ -3,7 +3,7 @@ import cheerio, { load } from "cheerio";
 import { URL } from "url";
 import { getLinksFromSitemap } from "./sitemap";
 import async from "async";
-import { Progress } from "../../lib/entities";
+import { CrawlerOptions, PageOptions, Progress } from "../../lib/entities";
 import { scrapSingleUrl, scrapWithScrapingBee } from "./single_url";
 import robotsParser from "robots-parser";
 
@@ -20,6 +20,7 @@ export class WebCrawler {
   private robotsTxtUrl: string;
   private robots: any;
   private generateImgAltText: boolean;
+  private allowBackwardCrawling: boolean;
 
   constructor({
     initialUrl,
@@ -29,6 +30,7 @@ export class WebCrawler {
     limit = 10000,
     generateImgAltText = false,
     maxCrawledDepth = 10,
+    allowBackwardCrawling = false
   }: {
     initialUrl: string;
     includes?: string[];
@@ -37,6 +39,7 @@ export class WebCrawler {
     limit?: number;
     generateImgAltText?: boolean;
     maxCrawledDepth?: number;
+    allowBackwardCrawling?: boolean;
   }) {
     this.initialUrl = initialUrl;
     this.baseUrl = new URL(initialUrl).origin;
@@ -49,6 +52,7 @@ export class WebCrawler {
     this.maxCrawledLinks = maxCrawledLinks ?? limit;
     this.maxCrawledDepth = maxCrawledDepth ?? 10;
     this.generateImgAltText = generateImgAltText ?? false;
+    this.allowBackwardCrawling = allowBackwardCrawling ?? false;
   }
 
   private filterLinks(sitemapLinks: string[], limit: number, maxDepth: number): string[] {
@@ -90,10 +94,16 @@ export class WebCrawler {
         const linkHostname = normalizedLink.hostname.replace(/^www\./, '');
 
         // Ensure the protocol and hostname match, and the path starts with the initial URL's path
-        if (linkHostname !== initialHostname || !normalizedLink.pathname.startsWith(normalizedInitialUrl.pathname)) {
+        if (linkHostname !== initialHostname) {
           return false;
         }
 
+        if (!this.allowBackwardCrawling) {
+          if (!normalizedLink.pathname.startsWith(normalizedInitialUrl.pathname)) {
+            return false;
+          }
+        }
+
         const isAllowed = this.robots.isAllowed(link, "FireCrawlAgent") ?? true;
         // Check if the link is disallowed by robots.txt
         if (!isAllowed) {
@@ -108,6 +118,8 @@ export class WebCrawler {
 
   public async start(
     inProgress?: (progress: Progress) => void,
+    pageOptions?: PageOptions,
+    crawlerOptions?: CrawlerOptions,
     concurrencyLimit: number = 5,
     limit: number = 10000,
     maxDepth: number = 10
@@ -122,17 +134,21 @@ export class WebCrawler {
     }
 
 
-    const sitemapLinks = await this.tryFetchSitemapLinks(this.initialUrl);
-    if (sitemapLinks.length > 0) {
-      let filteredLinks = this.filterLinks(sitemapLinks, limit, maxDepth);
-      return filteredLinks.map(link => ({ url: link, html: "" }));
+    if(!crawlerOptions?.ignoreSitemap){
+      const sitemapLinks = await this.tryFetchSitemapLinks(this.initialUrl);
+      if (sitemapLinks.length > 0) {
+        let filteredLinks = this.filterLinks(sitemapLinks, limit, maxDepth);
+        return filteredLinks.map(link => ({ url: link, html: "" }));
+      }
     }
 
     const urls = await this.crawlUrls(
       [this.initialUrl],
+      pageOptions,
       concurrencyLimit,
       inProgress
     );
+    
     if (
       urls.length === 0 &&
       this.filterLinks([this.initialUrl], limit, this.maxCrawledDepth).length > 0
@@ -140,14 +156,15 @@ export class WebCrawler {
       return [{ url: this.initialUrl, html: "" }];
     }
 
-
     // make sure to run include exclude here again
     const filteredUrls = this.filterLinks(urls.map(urlObj => urlObj.url), limit, this.maxCrawledDepth);
+
     return filteredUrls.map(url => ({ url, html: urls.find(urlObj => urlObj.url === url)?.html || "" }));
   }
 
   private async crawlUrls(
     urls: string[],
+    pageOptions: PageOptions,
     concurrencyLimit: number,
     inProgress?: (progress: Progress) => void,
   ): Promise<{ url: string, html: string }[]> {
@@ -158,7 +175,7 @@ export class WebCrawler {
         }
         return;
       }
-      const newUrls = await this.crawl(task);
+      const newUrls = await this.crawl(task, pageOptions);
       // add the initial url if not already added
       // if (this.visited.size === 1) {
       //   let normalizedInitial = this.initialUrl;
@@ -188,7 +205,7 @@ export class WebCrawler {
           currentDocumentUrl: task,
         });
       }
-      await this.crawlUrls(newUrls.map((p) => p.url), concurrencyLimit, inProgress);
+      await this.crawlUrls(newUrls.map((p) => p.url), pageOptions, concurrencyLimit, inProgress);
       if (callback && typeof callback === "function") {
         callback();
       }
@@ -207,20 +224,18 @@ export class WebCrawler {
     return Array.from(this.crawledUrls.entries()).map(([url, html]) => ({ url, html }));
   }
 
-  async crawl(url: string): Promise<{url: string, html: string}[]> {
-    if (this.visited.has(url) || !this.robots.isAllowed(url, "FireCrawlAgent")){
+  async crawl(url: string, pageOptions: PageOptions): Promise<{url: string, html: string}[]> {
+    const normalizedUrl = this.normalizeCrawlUrl(url);
+    if (this.visited.has(normalizedUrl) || !this.robots.isAllowed(url, "FireCrawlAgent")) {
       return [];
     }
-    this.visited.add(url);
-    
+    this.visited.add(normalizedUrl);
 
     if (!url.startsWith("http")) {
       url = "https://" + url;
-
     }
     if (url.endsWith("/")) {
       url = url.slice(0, -1);
-
     }
 
     if (this.isFile(url) || this.isSocialMediaOrEmail(url)) {
@@ -228,25 +243,23 @@ export class WebCrawler {
     }
 
     try {
-      let content : string = "";
+      let content: string = "";
       // If it is the first link, fetch with single url
       if (this.visited.size === 1) {
-        const page = await scrapSingleUrl(url, {includeHtml: true});
-        content = page.html ?? ""
+        const page = await scrapSingleUrl(url, { ...pageOptions, includeHtml: true });
+        content = page.html ?? "";
       } else {
         const response = await axios.get(url);
         content = response.data ?? "";
       }
       const $ = load(content);
-      let links: {url: string, html: string}[] = [];
+      let links: { url: string, html: string }[] = [];
 
       // Add the initial URL to the list of links
-      if(this.visited.size === 1)
-      {
-        links.push({url, html: content});
+      if (this.visited.size === 1) {
+        links.push({ url, html: content });
       }
 
-
       $("a").each((_, element) => {
         const href = $(element).attr("href");
         if (href) {
@@ -254,32 +267,43 @@ export class WebCrawler {
           if (!href.startsWith("http")) {
             fullUrl = new URL(href, this.baseUrl).toString();
           }
-          const url = new URL(fullUrl);
-          const path = url.pathname;
+          const urlObj = new URL(fullUrl);
+          const path = urlObj.pathname;
 
           if (
             this.isInternalLink(fullUrl) &&
             this.matchesPattern(fullUrl) &&
             this.noSections(fullUrl) &&
-            this.matchesIncludes(path) &&
+            // The idea here to comment this out is to allow wider website coverage as we filter this anyway afterwards
+            // this.matchesIncludes(path) &&
             !this.matchesExcludes(path) &&
             this.robots.isAllowed(fullUrl, "FireCrawlAgent")
           ) {
-            links.push({url: fullUrl, html: content});
+            links.push({ url: fullUrl, html: content });
           }
         }
       });
 
-      if(this.visited.size === 1){
+      if (this.visited.size === 1) {
         return links;
       }
       // Create a new list to return to avoid modifying the visited list
-      return links.filter((link) => !this.visited.has(link.url));
+      return links.filter((link) => !this.visited.has(this.normalizeCrawlUrl(link.url)));
     } catch (error) {
       return [];
     }
   }
 
+  private normalizeCrawlUrl(url: string): string {
+    try{
+      const urlObj = new URL(url);
+      urlObj.searchParams.sort(); // Sort query parameters to normalize
+      return urlObj.toString();
+    } catch (error) {
+      return url;
+    }
+  }
+
   private matchesIncludes(url: string): boolean {
     if (this.includes.length === 0 || this.includes[0] == "") return true;
     return this.includes.some((pattern) => new RegExp(pattern).test(url));
@@ -388,7 +412,6 @@ export class WebCrawler {
 
     // Normalize and check if the URL is present in any of the sitemaps
     const normalizedUrl = normalizeUrl(url);
-
     const normalizedSitemapLinks = sitemapLinks.map(link => normalizeUrl(link));
 
     // has to be greater than 0 to avoid adding the initial URL to the sitemap links, and preventing crawler to crawl

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

@@ -31,12 +31,14 @@ export class WebScraperDataProvider {
   private limit: number = 10000;
   private concurrentRequests: number = 20;
   private generateImgAltText: boolean = false;
+  private ignoreSitemap: boolean = false;
   private pageOptions?: PageOptions;
   private extractorOptions?: ExtractorOptions;
   private replaceAllPathsWithAbsolutePaths?: boolean = false;
   private generateImgAltTextModel: "gpt-4-turbo" | "claude-3-opus" =
     "gpt-4-turbo";
   private crawlerMode: string = "default";
+  private allowBackwardCrawling: boolean = false;
 
   authorize(): void {
     throw new Error("Method not implemented.");
@@ -169,10 +171,15 @@ export class WebScraperDataProvider {
       maxCrawledDepth: this.maxCrawledDepth,
       limit: this.limit,
       generateImgAltText: this.generateImgAltText,
+      allowBackwardCrawling: this.allowBackwardCrawling,
     });
 
     let links = await crawler.start(
       inProgress,
+      this.pageOptions,
+      {
+        ignoreSitemap: this.ignoreSitemap,
+      },
       5,
       this.limit,
       this.maxCrawledDepth
@@ -296,9 +303,10 @@ export class WebScraperDataProvider {
   }
 
   private applyPathReplacements(documents: Document[]): Document[] {
-    return this.replaceAllPathsWithAbsolutePaths
-      ? replacePathsWithAbsolutePaths(documents)
-      : replaceImgPathsWithAbsolutePaths(documents);
+    if (this.replaceAllPathsWithAbsolutePaths) {
+      documents = replacePathsWithAbsolutePaths(documents);
+    }
+    return replaceImgPathsWithAbsolutePaths(documents);
   }
 
   private async applyImgAltText(documents: Document[]): Promise<Document[]> {
@@ -467,12 +475,19 @@ export class WebScraperDataProvider {
     this.limit = options.crawlerOptions?.limit ?? 10000;
     this.generateImgAltText =
       options.crawlerOptions?.generateImgAltText ?? false;
-    this.pageOptions = options.pageOptions ?? { onlyMainContent: false, includeHtml: false };
+    this.pageOptions = options.pageOptions ?? {
+      onlyMainContent: false,
+      includeHtml: false,
+      replaceAllPathsWithAbsolutePaths: false,
+      removeTags: []
+    };
     this.extractorOptions = options.extractorOptions ?? {mode: "markdown"}
-    this.replaceAllPathsWithAbsolutePaths = options.crawlerOptions?.replaceAllPathsWithAbsolutePaths ?? false;
+    this.replaceAllPathsWithAbsolutePaths = options.crawlerOptions?.replaceAllPathsWithAbsolutePaths ?? options.pageOptions?.replaceAllPathsWithAbsolutePaths ?? false;
     //! @nicolas, for some reason this was being injected and breaking everything. Don't have time to find source of the issue so adding this check
     this.excludes = this.excludes.filter((item) => item !== "");
     this.crawlerMode = options.crawlerOptions?.mode ?? "default";
+    this.ignoreSitemap = options.crawlerOptions?.ignoreSitemap ?? false;
+    this.allowBackwardCrawling = options.crawlerOptions?.allowBackwardCrawling ?? false;
 
     // make sure all urls start with https://
     this.urls = this.urls.map((url) => {

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

@@ -8,6 +8,7 @@ import { excludeNonMainTags } from "./utils/excludeTags";
 import { urlSpecificParams } from "./utils/custom/website_params";
 import { fetchAndProcessPdf } from "./utils/pdfProcessor";
 import { handleCustomScraping } from "./custom/handleCustomScraping";
+import axios from "axios";
 
 dotenv.config();
 
@@ -19,6 +20,8 @@ const baseScrapers = [
   "fetch",
 ] as const;
 
+const universalTimeout = 15000;
+
 export async function generateRequestParams(
   url: string,
   wait_browser: string = "domcontentloaded",
@@ -59,21 +62,24 @@ export async function scrapWithFireEngine(
       `[Fire-Engine] Scraping ${url} with wait: ${waitParam} and screenshot: ${screenshotParam}`
     );
 
-    const response = await fetch(process.env.FIRE_ENGINE_BETA_URL + "/scrape", {
-      method: "POST",
-      headers: {
-        "Content-Type": "application/json",
-      },
-      body: JSON.stringify({
+    const response = await axios.post(
+      process.env.FIRE_ENGINE_BETA_URL + "/scrape",
+      {
         url: url,
         wait: waitParam,
         screenshot: screenshotParam,
         headers: headers,
-        pageOptions: pageOptions
-      }),
-    });
+        pageOptions: pageOptions,
+      },
+      {
+        headers: {
+          "Content-Type": "application/json",
+        },
+        timeout: universalTimeout + waitParam
+      }
+    );
 
-    if (!response.ok) {
+    if (response.status !== 200) {
       console.error(
         `[Fire-Engine] Error fetching url: ${url} with status: ${response.status}`
       );
@@ -84,13 +90,17 @@ export async function scrapWithFireEngine(
     if (contentType && contentType.includes("application/pdf")) {
       return { html: await fetchAndProcessPdf(url), screenshot: "" };
     } else {
-      const data = await response.json();
+      const data = response.data;
       const html = data.content;
       const screenshot = data.screenshot;
       return { html: html ?? "", screenshot: screenshot ?? "" };
     }
   } catch (error) {
-    console.error(`[Fire-Engine][c] Error fetching url: ${url} -> ${error}`);
+    if (error.code === 'ECONNABORTED') {
+      console.log(`[Fire-Engine] Request timed out for ${url}`);
+    } else {
+      console.error(`[Fire-Engine][c] Error fetching url: ${url} -> ${error}`);
+    }
     return { html: "", screenshot: "" };
   }
 }
@@ -98,7 +108,7 @@ export async function scrapWithFireEngine(
 export async function scrapWithScrapingBee(
   url: string,
   wait_browser: string = "domcontentloaded",
-  timeout: number = 15000
+  timeout: number = universalTimeout
 ): Promise<string> {
   try {
     const client = new ScrapingBeeClient(process.env.SCRAPING_BEE_API_KEY);
@@ -141,15 +151,19 @@ export async function scrapWithPlaywright(
     // If the user has passed a wait parameter in the request, use that
     const waitParam = reqParams["params"]?.wait ?? waitFor;
 
-    const response = await fetch(process.env.PLAYWRIGHT_MICROSERVICE_URL, {
-      method: "POST",
+    const response = await axios.post(process.env.PLAYWRIGHT_MICROSERVICE_URL, {
+      url: url,
+      wait_after_load: waitParam,
+      headers: headers,
+    }, {
       headers: {
         "Content-Type": "application/json",
       },
-      body: JSON.stringify({ url: url, wait_after_load: waitParam, headers: headers }),
+      timeout: universalTimeout + waitParam, // Add waitParam to timeout to account for the wait time
+      transformResponse: [(data) => data] // Prevent axios from parsing JSON automatically
     });
 
-    if (!response.ok) {
+    if (response.status !== 200) {
       console.error(
         `[Playwright] Error fetching url: ${url} with status: ${response.status}`
       );
@@ -160,7 +174,7 @@ export async function scrapWithPlaywright(
     if (contentType && contentType.includes("application/pdf")) {
       return fetchAndProcessPdf(url);
     } else {
-      const textData = await response.text();
+      const textData = response.data;
       try {
         const data = JSON.parse(textData);
         const html = data.content;
@@ -171,17 +185,28 @@ export async function scrapWithPlaywright(
       }
     }
   } catch (error) {
-    console.error(`[Playwright] Error fetching url: ${url} -> ${error}`);
+    if (error.code === 'ECONNABORTED') {
+      console.log(`[Playwright] Request timed out for ${url}`);
+    } else {
+      console.error(`[Playwright] Error fetching url: ${url} -> ${error}`);
+    }
     return "";
   }
 }
 
 export async function scrapWithFetch(url: string): Promise<string> {
   try {
-    const response = await fetch(url);
-    if (!response.ok) {
+    const response = await axios.get(url, {
+      headers: {
+        "Content-Type": "application/json",
+      },
+      timeout: universalTimeout,
+      transformResponse: [(data) => data] // Prevent axios from parsing JSON automatically
+    });
+
+    if (response.status !== 200) {
       console.error(
-        `[Fetch] Error fetching url: ${url} with status: ${response.status}`
+        `[Axios] Error fetching url: ${url} with status: ${response.status}`
       );
       return "";
     }
@@ -190,11 +215,15 @@ export async function scrapWithFetch(url: string): Promise<string> {
     if (contentType && contentType.includes("application/pdf")) {
       return fetchAndProcessPdf(url);
     } else {
-      const text = await response.text();
+      const text = response.data;
       return text;
     }
   } catch (error) {
-    console.error(`[Fetch][c] Error fetching url: ${url} -> ${error}`);
+    if (error.code === 'ECONNABORTED') {
+      console.log(`[Axios] Request timed out for ${url}`);
+    } else {
+      console.error(`[Axios] Error fetching url: ${url} -> ${error}`);
+    }
     return "";
   }
 }
@@ -275,6 +304,19 @@ export async function scrapSingleUrl(
   const removeUnwantedElements = (html: string, pageOptions: PageOptions) => {
     const soup = cheerio.load(html);
     soup("script, style, iframe, noscript, meta, head").remove();
+
+    if (pageOptions.removeTags) {
+      if (typeof pageOptions.removeTags === 'string') {
+        pageOptions.removeTags.split(',').forEach((tag) => {
+          soup(tag.trim()).remove();
+        });
+      } else if (Array.isArray(pageOptions.removeTags)) {
+        pageOptions.removeTags.forEach((tag) => {
+          soup(tag).remove();
+        });
+      }
+    }
+    
     if (pageOptions.onlyMainContent) {
       // remove any other tags that are not in the main content
       excludeNonMainTags.forEach((tag) => {

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

@@ -12,6 +12,7 @@ export async function getLinksFromSitemap(
       content = response.data;
     } catch (error) {
       console.error(`Request failed for ${sitemapUrl}: ${error}`);
+
       return allUrls;
     }
 

apps/api/src/scraper/WebScraper/utils/__tests__/replacePaths.test.ts

@@ -6,12 +6,14 @@ describe('replacePaths', () => {
     it('should replace relative paths with absolute paths', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'This is a [link](/path/to/resource) and an image ![alt text](/path/to/image.jpg).'
+        content: 'This is a [link](/path/to/resource).',
+        markdown: 'This is a [link](/path/to/resource).'
       }];
 
       const expectedDocuments: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'This is a [link](https://example.com/path/to/resource) and an image ![alt text](https://example.com/path/to/image.jpg).'
+        content: 'This is a [link](https://example.com/path/to/resource).',
+        markdown: 'This is a [link](https://example.com/path/to/resource).'
       }];
 
       const result = replacePathsWithAbsolutePaths(documents);
@@ -21,7 +23,8 @@ describe('replacePaths', () => {
     it('should not alter absolute URLs', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'This is an [external link](https://external.com/path) and an image ![alt text](https://example.com/path/to/image.jpg).'
+        content: 'This is an [external link](https://external.com/path).',
+        markdown: 'This is an [external link](https://external.com/path).'
       }];
 
       const result = replacePathsWithAbsolutePaths(documents);
@@ -31,7 +34,8 @@ describe('replacePaths', () => {
     it('should not alter data URLs for images', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'This is an image: ![alt text](data:image/png;base64,ABC123==).'
+        content: 'This is an image: ![alt text](data:image/png;base64,ABC123==).',
+        markdown: 'This is an image: ![alt text](data:image/png;base64,ABC123==).'
       }];
 
       const result = replacePathsWithAbsolutePaths(documents);
@@ -41,12 +45,14 @@ describe('replacePaths', () => {
     it('should handle multiple links and images correctly', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Here are two links: [link1](/path1) and [link2](/path2), and two images: ![img1](/img1.jpg) ![img2](/img2.jpg).'
+        content: 'Here are two links: [link1](/path1) and [link2](/path2).',
+        markdown: 'Here are two links: [link1](/path1) and [link2](/path2).'
       }];
 
       const expectedDocuments: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Here are two links: [link1](https://example.com/path1) and [link2](https://example.com/path2), and two images: ![img1](https://example.com/img1.jpg) ![img2](https://example.com/img2.jpg).'
+        content: 'Here are two links: [link1](https://example.com/path1) and [link2](https://example.com/path2).',
+        markdown: 'Here are two links: [link1](https://example.com/path1) and [link2](https://example.com/path2).'
       }];
 
       const result = replacePathsWithAbsolutePaths(documents);
@@ -56,12 +62,14 @@ describe('replacePaths', () => {
     it('should correctly handle a mix of absolute and relative paths', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Mixed paths: [relative](/path), [absolute](https://example.com/path), and [data image](data:image/png;base64,ABC123==).'
+        content: 'Mixed paths: [relative](/path), [absolute](https://example.com/path), and [data image](data:image/png;base64,ABC123==).',
+        markdown: 'Mixed paths: [relative](/path), [absolute](https://example.com/path), and [data image](data:image/png;base64,ABC123==).'
       }];
 
       const expectedDocuments: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Mixed paths: [relative](https://example.com/path), [absolute](https://example.com/path), and [data image](data:image/png;base64,ABC123==).'
+        content: 'Mixed paths: [relative](https://example.com/path), [absolute](https://example.com/path), and [data image](data:image/png;base64,ABC123==).',
+        markdown: 'Mixed paths: [relative](https://example.com/path), [absolute](https://example.com/path), and [data image](data:image/png;base64,ABC123==).'
       }];
 
       const result = replacePathsWithAbsolutePaths(documents);
@@ -74,12 +82,14 @@ describe('replacePaths', () => {
     it('should replace relative image paths with absolute paths', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Here is an image: ![alt text](/path/to/image.jpg).'
+        content: 'Here is an image: ![alt text](/path/to/image.jpg).',
+        markdown: 'Here is an image: ![alt text](/path/to/image.jpg).'
       }];
 
       const expectedDocuments: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Here is an image: ![alt text](https://example.com/path/to/image.jpg).'
+        content: 'Here is an image: ![alt text](https://example.com/path/to/image.jpg).',
+        markdown: 'Here is an image: ![alt text](https://example.com/path/to/image.jpg).'
       }];
 
       const result = replaceImgPathsWithAbsolutePaths(documents);
@@ -89,7 +99,8 @@ describe('replacePaths', () => {
     it('should not alter data:image URLs', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'An image with a data URL: ![alt text](data:image/png;base64,ABC123==).'
+        content: 'An image with a data URL: ![alt text](data:image/png;base64,ABC123==).',
+        markdown: 'An image with a data URL: ![alt text](data:image/png;base4,ABC123==).'
       }];
 
       const result = replaceImgPathsWithAbsolutePaths(documents);
@@ -99,12 +110,14 @@ describe('replacePaths', () => {
     it('should handle multiple images with a mix of data and relative URLs', () => {
       const documents: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Multiple images: ![img1](/img1.jpg) ![img2](data:image/png;base64,ABC123==) ![img3](/img3.jpg).'
+        content: 'Multiple images: ![img1](/img1.jpg) ![img2](data:image/png;base64,ABC123==) ![img3](/img3.jpg).',
+        markdown: 'Multiple images: ![img1](/img1.jpg) ![img2](data:image/png;base64,ABC123==) ![img3](/img3.jpg).'
       }];
 
       const expectedDocuments: Document[] = [{
         metadata: { sourceURL: 'https://example.com' },
-        content: 'Multiple images: ![img1](https://example.com/img1.jpg) ![img2](data:image/png;base64,ABC123==) ![img3](https://example.com/img3.jpg).'
+        content: 'Multiple images: ![img1](https://example.com/img1.jpg) ![img2](data:image/png;base64,ABC123==) ![img3](https://example.com/img3.jpg).',
+        markdown: 'Multiple images: ![img1](https://example.com/img1.jpg) ![img2](data:image/png;base64,ABC123==) ![img3](https://example.com/img3.jpg).'
       }];
 
       const result = replaceImgPathsWithAbsolutePaths(documents);

apps/api/src/scraper/WebScraper/utils/replacePaths.ts

@@ -10,7 +10,8 @@ export const replacePathsWithAbsolutePaths = (documents: Document[]): Document[]
         ) || [];
 
       paths.forEach((path: string) => {
-        const isImage = path.startsWith("!");
+        try {
+          const isImage = path.startsWith("!");
         let matchedUrl = path.match(/\(([^)]+)\)/) || path.match(/href="([^"]+)"/);
         let url = matchedUrl[1];
 
@@ -22,18 +23,18 @@ export const replacePathsWithAbsolutePaths = (documents: Document[]): Document[]
         }
 
         const markdownLinkOrImageText = path.match(/(!?\[.*?\])/)[0];
-        if (isImage) {
-          document.content = document.content.replace(
-            path,
-            `${markdownLinkOrImageText}(${url})`
-          );
-        } else {
+        // Image is handled afterwards
+        if (!isImage) {
           document.content = document.content.replace(
             path,
             `${markdownLinkOrImageText}(${url})`
           );
+          }
+        } catch (error) {
+          
         }
       });
+      document.markdown = document.content;
     });
 
     return documents;
@@ -60,8 +61,10 @@ export const replaceImgPathsWithAbsolutePaths = (documents: Document[]): Documen
           if (!imageUrl.startsWith("http")) {
             if (imageUrl.startsWith("/")) {
               imageUrl = imageUrl.substring(1);
+              imageUrl = new URL(imageUrl, baseUrl).toString();
+            } else {
+              imageUrl = new URL(imageUrl, document.metadata.sourceURL).toString();
             }
-            imageUrl = new URL(imageUrl, baseUrl).toString();
           }
         }
 
@@ -70,6 +73,7 @@ export const replaceImgPathsWithAbsolutePaths = (documents: Document[]): Documen
           `![${altText}](${imageUrl})`
         );
       });
+      document.markdown = document.content;
     });
 
     return documents;

apps/api/src/services/queue-worker.ts

@@ -38,7 +38,7 @@ getWebScraperQueue().process(
         error: message /* etc... */,
       };
 
-      await callWebhook(job.data.team_id, data);
+      await callWebhook(job.data.team_id, job.id as string, data);
 
       await logJob({
         success: success,
@@ -78,7 +78,7 @@ getWebScraperQueue().process(
         error:
           "Something went wrong... Contact help@mendable.ai or try again." /* etc... */,
       };
-      await callWebhook(job.data.team_id, data);
+      await callWebhook(job.data.team_id, job.id as string, data);
       await logJob({
         success: false,
         message: typeof error === 'string' ? error : (error.message ?? "Something went wrong... Contact help@mendable.ai"),

apps/api/src/services/redis.ts

@@ -1,8 +1,35 @@
-import Redis from 'ioredis';
+import Redis from "ioredis";
 
 // Initialize Redis client
 const redis = new Redis(process.env.REDIS_URL);
 
+// Listen to 'error' events to the Redis connection
+redis.on("error", (error) => {
+  try {
+    if (error.message === "ECONNRESET") {
+      console.log("Connection to Redis Session Store timed out.");
+    } else if (error.message === "ECONNREFUSED") {
+      console.log("Connection to Redis Session Store refused!");
+    } else console.log(error);
+  } catch (error) {}
+});
+
+// Listen to 'reconnecting' event to Redis
+redis.on("reconnecting", (err) => {
+  try {
+    if (redis.status === "reconnecting")
+      console.log("Reconnecting to Redis Session Store...");
+    else console.log("Error reconnecting to Redis Session Store.");
+  } catch (error) {}
+});
+
+// Listen to the 'connect' event to Redis
+redis.on("connect", (err) => {
+  try {
+    if (!err) console.log("Connected to Redis Session Store!");
+  } catch (error) {}
+});
+
 /**
  * Set a value in Redis with an optional expiration time.
  * @param {string} key The key under which to store the value.
@@ -11,7 +38,7 @@ const redis = new Redis(process.env.REDIS_URL);
  */
 const setValue = async (key: string, value: string, expire?: number) => {
   if (expire) {
-    await redis.set(key, value, 'EX', expire);
+    await redis.set(key, value, "EX", expire);
   } else {
     await redis.set(key, value);
   }

apps/api/src/services/webhook.ts

@@ -1,17 +1,19 @@
 import { supabase_service } from "./supabase";
 
-export const callWebhook = async (teamId: string, data: any) => {
+export const callWebhook = async (teamId: string, jobId: string,data: any) => {
   try {
     const selfHostedUrl = process.env.SELF_HOSTED_WEBHOOK_URL;
+    const useDbAuthentication = process.env.USE_DB_AUTHENTICATION === 'true';
     let webhookUrl = selfHostedUrl;
 
-    if (!selfHostedUrl) {
+    // Only fetch the webhook URL from the database if the self-hosted webhook URL is not set
+    // and the USE_DB_AUTHENTICATION environment variable is set to true
+    if (!selfHostedUrl && useDbAuthentication) {
       const { data: webhooksData, error } = await supabase_service
         .from("webhooks")
         .select("url")
         .eq("team_id", teamId)
         .limit(1);
-
       if (error) {
         console.error(
           `Error fetching webhook URL for team ID: ${teamId}`,
@@ -45,6 +47,7 @@ export const callWebhook = async (teamId: string, data: any) => {
       },
       body: JSON.stringify({
         success: data.success,
+        jobId: jobId,
         data: dataToSend,
         error: data.error || undefined,
       }),

apps/python-sdk/firecrawl/__init__.py

@@ -1,3 +1,57 @@
+"""
+This is the Firecrawl package.
+
+This package provides a Python SDK for interacting with the Firecrawl API.
+It includes methods to scrape URLs, perform searches, initiate and monitor crawl jobs,
+and check the status of these jobs.
+
+For more information visit https://github.com/firecrawl/
+"""
+
+import logging
+import os
+
 from .firecrawl import FirecrawlApp
 
-__version__ = "0.0.14"
+__version__ = "0.0.15"
+
+# Define the logger for the Firecrawl project
+logger: logging.Logger = logging.getLogger("firecrawl")
+
+
+def _basic_config() -> None:
+    """Set up basic configuration for logging with a specific format and date format."""
+    try:
+        logging.basicConfig(
+            format="[%(asctime)s - %(name)s:%(lineno)d - %(levelname)s] %(message)s",
+            datefmt="%Y-%m-%d %H:%M:%S",
+        )
+    except Exception as e:
+        logger.error("Failed to configure logging: %s", e)
+
+
+def setup_logging() -> None:
+    """Set up logging based on the FIRECRAWL_LOGGING_LEVEL environment variable."""
+    env = os.environ.get(
+        "FIRECRAWL_LOGGING_LEVEL", "INFO"
+    ).upper()  # Default to 'INFO' level
+    _basic_config()
+
+    if env == "DEBUG":
+        logger.setLevel(logging.DEBUG)
+    elif env == "INFO":
+        logger.setLevel(logging.INFO)
+    elif env == "WARNING":
+        logger.setLevel(logging.WARNING)
+    elif env == "ERROR":
+        logger.setLevel(logging.ERROR)
+    elif env == "CRITICAL":
+        logger.setLevel(logging.CRITICAL)
+    else:
+        logger.setLevel(logging.INFO)
+        logger.warning("Unknown logging level: %s, defaulting to INFO", env)
+
+
+# Initialize logging configuration when the module is imported
+setup_logging()
+logger.debug("Debugging logger setup")

apps/python-sdk/firecrawl/firecrawl.py

@@ -9,13 +9,14 @@ and handles retries for certain HTTP status codes.
 Classes:
     - FirecrawlApp: Main class for interacting with the Firecrawl API.
 """
-
+import logging
 import os
 import time
 from typing import Any, Dict, Optional
 
 import requests
 
+logger : logging.Logger = logging.getLogger("firecrawl")
 
 class FirecrawlApp:
     """
@@ -28,8 +29,14 @@ class FirecrawlApp:
     def __init__(self, api_key: Optional[str] = None, api_url: Optional[str] = None) -> None:
         self.api_key = api_key or os.getenv('FIRECRAWL_API_KEY')
         if self.api_key is None:
+            logger.warning("No API key provided")
             raise ValueError('No API key provided')
+        else:
+            logger.debug("Initialized FirecrawlApp with API key: %s", self.api_key)
+
         self.api_url = api_url or os.getenv('FIRECRAWL_API_URL', 'https://api.firecrawl.dev')
+        if self.api_url != 'https://api.firecrawl.dev':
+            logger.debug("Initialized FirecrawlApp with API URL: %s", self.api_url)
 
     def scrape_url(self, url: str, params: Optional[Dict[str, Any]] = None) -> Any:
         """

examples/roastmywebsite-example-app/src/lib/LLM/llm.ts

@@ -1,4 +1,4 @@
-import OpenAI from "openai";
+import OpenAI from "openai/index.mjs";
 import { encoding_for_model } from "@dqbd/tiktoken";
 
 /**