fix(crawl): submit sitemapped jobs in bulk

apps/api/src/controllers/crawl.ts

@@ -10,7 +10,8 @@ import { createIdempotencyKey } from "../../src/services/idempotency/create";
 import { defaultCrawlPageOptions, defaultCrawlerOptions, defaultOrigin } from "../../src/lib/default-values";
 import { v4 as uuidv4 } from "uuid";
 import { Logger } from "../../src/lib/logger";
-import { addCrawlJob, crawlToCrawler, lockURL, saveCrawl, StoredCrawl } from "../../src/lib/crawl-redis";
+import { addCrawlJob, addCrawlJobs, crawlToCrawler, lockURL, lockURLs, saveCrawl, StoredCrawl } from "../../src/lib/crawl-redis";
+import { getScrapeQueue } from "../../src/services/queue-service";
 
 export async function crawlController(req: Request, res: Response) {
   try {
@@ -115,20 +116,31 @@ export async function crawlController(req: Request, res: Response) {
     const sitemap = sc.crawlerOptions?.ignoreSitemap ? null : await crawler.tryGetSitemap();
 
     if (sitemap !== null) {
-      for (const url of sitemap.map(x => x.url)) {
+      const jobs = sitemap.map(x => {
-        await lockURL(id, sc, url);
+        const url = x.url;
-        const job = await addScrapeJob({
+        const uuid = uuidv4();
-          url,
+        return {
-          mode: "single_urls",
+          name: uuid,
-          crawlerOptions: crawlerOptions,
+          data: {
-          team_id: team_id,
+            url,
-          pageOptions: pageOptions,
+            mode: "single_urls",
-          origin: req.body.origin ?? defaultOrigin,
+            crawlerOptions: crawlerOptions,
-          crawl_id: id,
+            team_id: team_id,
-          sitemapped: true,
+            pageOptions: pageOptions,
-        });
+            origin: req.body.origin ?? defaultOrigin,
-        await addCrawlJob(id, job.id);
+            crawl_id: id,
-      }
+            sitemapped: true,
+          },
+          opts: {
+            jobId: uuid,
+            priority: 2,
+          }
+        };
+      })
+
+      await lockURLs(id, jobs.map(x => x.data.url));
+      await addCrawlJobs(id, jobs.map(x => x.opts.jobId));
+      await getScrapeQueue().addBulk(jobs);
     } else {
       await lockURL(id, sc, url);
       const job = await addScrapeJob({

apps/api/src/lib/crawl-redis.ts

@@ -30,6 +30,11 @@ export async function addCrawlJob(id: string, job_id: string) {
     await redisConnection.expire("crawl:" + id + ":jobs", 24 * 60 * 60, "NX");
 }
 
+export async function addCrawlJobs(id: string, job_ids: string[]) {
+    await redisConnection.sadd("crawl:" + id + ":jobs", ...job_ids);
+    await redisConnection.expire("crawl:" + id + ":jobs", 24 * 60 * 60, "NX");
+}
+
 export async function addCrawlJobDone(id: string, job_id: string) {
     await redisConnection.sadd("crawl:" + id + ":jobs_done", job_id);
     await redisConnection.expire("crawl:" + id + ":jobs_done", 24 * 60 * 60, "NX");
@@ -54,6 +59,13 @@ export async function lockURL(id: string, sc: StoredCrawl, url: string): Promise
     return res;
 }
 
+/// NOTE: does not check limit. only use if limit is checked beforehand e.g. with sitemap
+export async function lockURLs(id: string, urls: string[]): Promise<boolean> {
+    const res = (await redisConnection.sadd("crawl:" + id + ":visited", ...urls)) !== 0
+    await redisConnection.expire("crawl:" + id + ":visited", 24 * 60 * 60, "NX");
+    return res;
+}
+
 export function crawlToCrawler(id: string, sc: StoredCrawl): WebCrawler {
     const crawler = new WebCrawler({
         jobId: id,