Skip to content

Supercharge your AI agents with data from the web and beyond. Building the library for superintelligence. 🔥

Nick:

Nicolas1a1ac9fd60ba
diff --git a/apps/api/src/controllers/v0/crawl-status.ts b/apps/api/src/controllers/v0/crawl-status.ts
index 1b1ffdc5f..f809c7be3 100644
--- a/apps/api/src/controllers/v0/crawl-status.ts
+++ b/apps/api/src/controllers/v0/crawl-status.ts
@@ -50,11 +50,15 @@ export async function crawlStatusController(req: Request, res: Response) {
return res.status(403).json({ error: "Forbidden" });
}
- const jobIDs = await getCrawlJobs(req.params.jobId);
+ let jobIDs = await getCrawlJobs(req.params.jobId);
const jobs = (await getJobs(req.params.jobId, jobIDs)).sort((a, b) => a.timestamp - b.timestamp);
- const jobStatuses = await Promise.all(jobs.map(x => x.getState()));
- const jobStatus = sc.cancelled ? "failed" : jobStatuses.every(x => x === "completed") ? "completed" : jobs.some((x, i) => jobStatuses[i] === "failed" && x.failedReason !== "Concurrency limit hit") ? "failed" : "active";
+ let jobStatuses = await Promise.all(jobs.map(x => x.getState()));
+ // filter out failed jobs
+ jobIDs = jobIDs.filter(id => !jobStatuses.some(status => status[0] === id && status[1] === "failed"));
+ // filter the job statues
+ jobStatuses = jobStatuses.filter(x => x[1] !== "failed");
+ const jobStatus = sc.cancelled ? "failed" : jobStatuses.every(x => x === "completed") ? "completed" : "active";
const data = jobs.filter(x => x.failedReason !== "Concurreny limit hit").map(x => Array.isArray(x.returnvalue) ? x.returnvalue[0] : x.returnvalue);
diff --git a/apps/api/src/controllers/v1/crawl-status-ws.ts b/apps/api/src/controllers/v1/crawl-status-ws.ts
index 9832a948f..b67e559b2 100644
--- a/apps/api/src/controllers/v1/crawl-status-ws.ts
+++ b/apps/api/src/controllers/v1/crawl-status-ws.ts
@@ -94,11 +94,15 @@ async function crawlStatusWS(ws: WebSocket, req: RequestWithAuth<CrawlStatusPara
doneJobIDs = await getDoneJobsOrdered(req.params.jobId);
- const jobIDs = await getCrawlJobs(req.params.jobId);
+ let jobIDs = await getCrawlJobs(req.params.jobId);
let jobStatuses = await Promise.all(jobIDs.map(async x => [x, await getScrapeQueue().getJobState(x)] as const));
const throttledJobs = new Set(...await getThrottledJobs(req.auth.team_id));
jobStatuses = jobStatuses.filter(x => !throttledJobs.has(x[0])); // throttled jobs can have a failed status, but they are not actually failed
- const status: Exclude<CrawlStatusResponse, ErrorResponse>["status"] = sc.cancelled ? "cancelled" : jobStatuses.every(x => x[1] === "completed") ? "completed" : jobStatuses.some(x => x[1] === "failed") ? "failed" : "scraping";
+ // filter out failed jobs
+ jobIDs = jobIDs.filter(id => !jobStatuses.some(status => status[0] === id && status[1] === "failed"));
+ // filter the job statues
+ jobStatuses = jobStatuses.filter(x => x[1] !== "failed");
+ const status: Exclude<CrawlStatusResponse, ErrorResponse>["status"] = sc.cancelled ? "cancelled" : jobStatuses.every(x => x[1] === "completed") ? "completed" : "scraping";
const doneJobs = await getJobs(doneJobIDs);
const data = doneJobs.map(x => x.returnvalue);
diff --git a/apps/api/src/controllers/v1/crawl-status.ts b/apps/api/src/controllers/v1/crawl-status.ts
index 9c0026a08..63331c9c3 100644
--- a/apps/api/src/controllers/v1/crawl-status.ts
+++ b/apps/api/src/controllers/v1/crawl-status.ts
@@ -57,11 +57,15 @@ export async function crawlStatusController(req: RequestWithAuth<CrawlStatusPara
const start = typeof req.query.skip === "string" ? parseInt(req.query.skip, 10) : 0;
const end = typeof req.query.limit === "string" ? (start + parseInt(req.query.limit, 10) - 1) : undefined;
- const jobIDs = await getCrawlJobs(req.params.jobId);
+ let jobIDs = await getCrawlJobs(req.params.jobId);
let jobStatuses = await Promise.all(jobIDs.map(async x => [x, await getScrapeQueue().getJobState(x)] as const));
const throttledJobs = new Set(...await getThrottledJobs(req.auth.team_id));
jobStatuses = jobStatuses.filter(x => !throttledJobs.has(x[0])); // throttled jobs can have a failed status, but they are not actually failed
- const status: Exclude<CrawlStatusResponse, ErrorResponse>["status"] = sc.cancelled ? "cancelled" : jobStatuses.every(x => x[1] === "completed") ? "completed" : jobStatuses.some(x => x[1] === "failed") ? "failed" : "scraping";
+ // filter out failed jobs
+ jobIDs = jobIDs.filter(id => !jobStatuses.some(status => status[0] === id && status[1] === "failed"));
+ // filter the job statues
+ jobStatuses = jobStatuses.filter(x => x[1] !== "failed");
+ const status: Exclude<CrawlStatusResponse, ErrorResponse>["status"] = sc.cancelled ? "cancelled" : jobStatuses.every(x => x[1] === "completed") ? "completed" : "scraping";
const doneJobsLength = await getDoneJobsOrderedLength(req.params.jobId);
const doneJobsOrder = await getDoneJobsOrdered(req.params.jobId, start, end ?? -1);
diff --git a/apps/api/src/main/runWebScraper.ts b/apps/api/src/main/runWebScraper.ts
index 571122f95..6e642c65a 100644
--- a/apps/api/src/main/runWebScraper.ts
+++ b/apps/api/src/main/runWebScraper.ts
@@ -59,6 +59,7 @@ export async function startWebScraperPipeline({
is_scrape: job.data.is_scrape ?? false,
})) as { success: boolean; message: string; docs: Document[] };
}
+
export async function runWebScraper({
url,
mode,
diff --git a/apps/api/src/services/queue-worker.ts b/apps/api/src/services/queue-worker.ts
index 532e8feeb..1ccf486ec 100644
--- a/apps/api/src/services/queue-worker.ts
+++ b/apps/api/src/services/queue-worker.ts
@@ -395,6 +395,7 @@ async function processJob(job: Job, token: string) {
pageOptions: sc.pageOptions,
origin: job.data.origin,
crawl_id: job.data.crawl_id,
+ webhook: job.data.webhook,
v1: job.data.v1,
},
{},
@@ -468,9 +469,8 @@ async function processJob(job: Job, token: string) {
}
} else {
const jobIDs = await getCrawlJobs(job.data.crawl_id);
- const jobStatuses = await Promise.all(jobIDs.map((x) => getScrapeQueue().getJobState(x)));
const jobStatus =
- sc.cancelled || jobStatuses.some((x) => x === "failed")
+ sc.cancelled
? "failed"
: "completed";
@@ -554,16 +554,16 @@ async function processJob(job: Job, token: string) {
job.data.v1
);
}
- if (job.data.v1) {
- callWebhook(
- job.data.team_id,
- job.id as string,
- [],
- job.data.webhook,
- job.data.v1,
- "crawl.failed"
- );
- }
+ // if (job.data.v1) {
+ // callWebhook(
+ // job.data.team_id,
+ // job.id as string,
+ // [],
+ // job.data.webhook,
+ // job.data.v1,
+ // "crawl.failed"
+ // );
+ // }
if (job.data.crawl_id) {
await logJob({