1import { ApifyClient } from 'apify-client';
2
3
4
5const client = new ApifyClient({
6 token: '<YOUR_API_TOKEN>',
7});
8
9
10const input = {
11 "sources": [
12 "https://nvlpubs.nist.gov/nistpubs/CSWP/NIST.CSWP.04162018.pdf",
13 "https://raw.githubusercontent.com/mwilliamson/python-mammoth/f3b7b9fc73fdbffe6ebac77e6b3ac6c418ee33d9/tests/test-data/tables.docx",
14 "https://raw.githubusercontent.com/tesseract-ocr/test/232ff181c66516116ec0e84c4963f70de15050fd/testing/phototest.tif"
15 ],
16 "ocrLanguages": [
17 "eng"
18 ],
19 "maxPagesPerDocument": 5
20};
21
22
23const run = await client.actor("tinlark/document-to-markdown").call(input);
24
25
26console.log('Results from dataset');
27console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
28const { items } = await client.dataset(run.defaultDatasetId).listItems();
29items.forEach((item) => {
30 console.dir(item);
31});
32
33