AI/firecrawl
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/mendableai/firecrawl synced 2025-08-14 05:05:55 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #994 from mendableai/feat/added-id-to-ws-sdks

Browse Source 
feat-SDK/added crawl id to ws
This commit is contained in:
Nicolas 2024-12-19 16:08:08 -03:00 committed by GitHub
commit d67db99791
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 18 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

15
apps/js-sdk/firecrawl/src/index.ts
View File

@ -935,9 +935,11 @@ export class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
private ws: WebSocket; private ws: WebSocket;
public data: FirecrawlDocument<undefined>[]; public data: FirecrawlDocument<undefined>[];
public status: CrawlStatusResponse["status"]; public status: CrawlStatusResponse["status"];
public id: string;
constructor(id: string, app: FirecrawlApp) { constructor(id: string, app: FirecrawlApp) {
super(); super();
this.id = id;
this.ws = new WebSocket(`${app.apiUrl}/v1/crawl/${id}`, app.apiKey); this.ws = new WebSocket(`${app.apiUrl}/v1/crawl/${id}`, app.apiKey);
this.status = "scraping"; this.status = "scraping";
this.data = []; this.data = [];
@ -968,6 +970,7 @@ export class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
detail: { detail: {
status: this.status, status: this.status,
data: this.data, data: this.data,
id: this.id,
}, },
})); }));
} else if (msg.type === "error") { } else if (msg.type === "error") {
@ -977,6 +980,7 @@ export class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
status: this.status, status: this.status,
data: this.data, data: this.data,
error: msg.error, error: msg.error,
id: this.id,
}, },
})); }));
} else if (msg.type === "catchup") { } else if (msg.type === "catchup") {
@ -984,12 +988,18 @@ export class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
this.data.push(...(msg.data.data ?? [])); this.data.push(...(msg.data.data ?? []));
for (const doc of this.data) { for (const doc of this.data) {
this.dispatchTypedEvent("document", new CustomEvent("document", { this.dispatchTypedEvent("document", new CustomEvent("document", {
detail: doc, detail: {
...doc,
id: this.id,
},
})); }));
} }
} else if (msg.type === "document") { } else if (msg.type === "document") {
this.dispatchTypedEvent("document", new CustomEvent("document", { this.dispatchTypedEvent("document", new CustomEvent("document", {
detail: msg.data, detail: {
...msg.data,
id: this.id,
},
})); }));
} }
} }
@ -1016,6 +1026,7 @@ export class CrawlWatcher extends TypedEventTarget<CrawlWatcherEvents> {
status: this.status, status: this.status,
data: this.data, data: this.data,
error: "WebSocket error", error: "WebSocket error",
id: this.id,
}, },
})); }));
}).bind(this); }).bind(this);

2
apps/python-sdk/firecrawl/__init__.py
View File

@ -13,7 +13,7 @@ import os
from .firecrawl import FirecrawlApp # noqa from .firecrawl import FirecrawlApp # noqa
__version__ = "1.6.4" __version__ = "1.6.5"
# Define the logger for the Firecrawl project # Define the logger for the Firecrawl project
logger: logging.Logger = logging.getLogger("firecrawl") logger: logging.Logger = logging.getLogger("firecrawl")

8
apps/python-sdk/firecrawl/firecrawl.py
View File

@ -704,15 +704,15 @@ class CrawlWatcher:
async def _handle_message(self, msg: Dict[str, Any]): async def _handle_message(self, msg: Dict[str, Any]):
if msg['type'] == 'done': if msg['type'] == 'done':
self.status = 'completed' self.status = 'completed'
self.dispatch_event('done', {'status': self.status, 'data': self.data}) self.dispatch_event('done', {'status': self.status, 'data': self.data, 'id': self.id})
elif msg['type'] == 'error': elif msg['type'] == 'error':
self.status = 'failed' self.status = 'failed'
self.dispatch_event('error', {'status': self.status, 'data': self.data, 'error': msg['error']}) self.dispatch_event('error', {'status': self.status, 'data': self.data, 'error': msg['error'], 'id': self.id})
elif msg['type'] == 'catchup': elif msg['type'] == 'catchup':
self.status = msg['data']['status'] self.status = msg['data']['status']
self.data.extend(msg['data'].get('data', [])) self.data.extend(msg['data'].get('data', []))
for doc in self.data: for doc in self.data:
self.dispatch_event('document', doc) self.dispatch_event('document', {'data': doc, 'id': self.id})
elif msg['type'] == 'document': elif msg['type'] == 'document':
self.data.append(msg['data']) self.data.append(msg['data'])
self.dispatch_event('document', msg['data']) self.dispatch_event('document', {'data': msg['data'], 'id': self.id})