1from apify_client import ApifyClient
2
3
4
5client = ApifyClient("<YOUR_API_TOKEN>")
6
7
8run_input = {
9 "query": "what is retrieval augmented generation",
10 "queries": [],
11 "startUrls": [],
12 "searchEngine": "auto",
13 "searchLanguage": "auto",
14 "searchCountry": "auto",
15 "maxResults": 3,
16 "maxItems": 0,
17 "outputFormats": ["markdown"],
18 "removeElementsCssSelector": "nav, footer, aside, script, style, .ad, .advertisement, [aria-hidden='true']",
19 "keepElementsCssSelector": [],
20 "htmlTransformer": "readable-text",
21 "maxContentLength": 0,
22 "includeMetadata": True,
23 "chunkMarkdown": False,
24 "chunkMaxChars": 2000,
25 "chunkOverlapChars": 200,
26 "maxCrawlDepth": 0,
27 "maxCrawlPages": 25,
28 "domainAllowlist": [],
29 "domainBlocklist": [],
30 "excludeContentSelectors": [],
31 "keywordsAnyOf": [],
32 "emitFailedResults": False,
33}
34
35
36run = client.actor("vexaloom-labs/rag-web-browser").call(run_input=run_input)
37
38
39print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
40for item in client.dataset(run.default_dataset_id).iterate_items():
41 print(item)
42
43
