How to scrape CNBC

# Scrape CNBC articles with the Ujeebu Scrape API.
# Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
# Uses the official SDK ujeebu-python 0.1.5: pip install ujeebu-python==0.1.5
# Run: UJEEBU_API_KEY=your_key python3 cnbc_detail_sdk.py

import json
import os
import sys
import time

import requests
from ujeebu_python import UjeebuClient

API_KEY = os.environ.get("UJEEBU_API_KEY")
URLS = [
    "https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html",
    "https://www.cnbc.com/2026/04/21/spacex-says-it-can-buy-cursor-later-this-year-for-60-billion-or-pay-10-billion-for-our-work-together.html",
]
SETTINGS = {
    "js": True,
}
EXTRACT_RULES = {
    "article": {
        "type": "fn",
        "fn": "const T = ['NewsArticle', 'Article', 'BlogPosting', 'ReportageNews', 'AnalysisNewsArticle', 'TechArticle', 'ScholarlyArticle', 'LiveBlogPosting', 'OpinionNewsArticle', 'Report']; const n = [...document.querySelectorAll('script[type=\"application/ld+json\"]')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x['@graph'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x['@type']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split('|').map(q => q.split('.').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== ''); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(', ') : (v && typeof v === 'object' ? (v.name ?? v['@id'] ?? null) : (typeof v === 'string' ? v.replace('https://schema.org/', '').replace('http://schema.org/', '') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g('description')), headline: all(g('headline|name')), image: one(g('image.url|image')), modified: all(g('dateModified')), published: all(g('datePublished')), publisher: all(g('publisher.name')), section: all(g('articleSection')) };",
    },
    "url": {
        "selector": "link[rel=\"canonical\"]",
        "type": "attr",
        "attribute": "href",
    },
}


def scrape(client, url):
    # with extract_rules, client.scrape sends a JSON POST and the API answers {"result": {...}}
    try:
        response = client.scrape(url, params={**SETTINGS, "extract_rules": EXTRACT_RULES})
    except requests.RequestException as error:
        sys.exit(f"Request failed for {url}: {error}")
    if response.status_code != 200:
        try:
            message = response.json().get("message", response.text)
        except ValueError:
            message = response.text
        sys.exit(f"HTTP {response.status_code} for {url}: {str(message)[:300]}")
    print(f"Ujb-Credits: {response.headers.get('Ujb-Credits')}", file=sys.stderr)
    return response.json()["result"]


if not API_KEY:
    sys.exit("Set the UJEEBU_API_KEY environment variable.")

client = UjeebuClient(api_key=API_KEY)

for i, url in enumerate(URLS):
    if i:
        time.sleep(2)  # be polite: one page at a time
    print(json.dumps(scrape(client, url), indent=2, ensure_ascii=False))

// Scrape CNBC articles with the Ujeebu Scrape API.
// Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
// Uses the official SDK @ujeebu-org/ujeebu-sdk 0.1.2: npm install @ujeebu-org/ujeebu-sdk@0.1.2
// Run: UJEEBU_API_KEY=your_key node cnbc_detail_sdk.mjs

import { UjeebuClient } from '@ujeebu-org/ujeebu-sdk';

const API_KEY = process.env.UJEEBU_API_KEY;
const URLS = [
  "https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html",
  "https://www.cnbc.com/2026/04/21/spacex-says-it-can-buy-cursor-later-this-year-for-60-billion-or-pay-10-billion-for-our-work-together.html",
];
const SETTINGS = {
  js: true,
};
const EXTRACT_RULES = {
  article: {
    type: "fn",
    fn: "const T = ['NewsArticle', 'Article', 'BlogPosting', 'ReportageNews', 'AnalysisNewsArticle', 'TechArticle', 'ScholarlyArticle', 'LiveBlogPosting', 'OpinionNewsArticle', 'Report']; const n = [...document.querySelectorAll('script[type=\"application/ld+json\"]')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x['@graph'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x['@type']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split('|').map(q => q.split('.').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== ''); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(', ') : (v && typeof v === 'object' ? (v.name ?? v['@id'] ?? null) : (typeof v === 'string' ? v.replace('https://schema.org/', '').replace('http://schema.org/', '') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g('description')), headline: all(g('headline|name')), image: one(g('image.url|image')), modified: all(g('dateModified')), published: all(g('datePublished')), publisher: all(g('publisher.name')), section: all(g('articleSection')) };",
  },
  url: {
    selector: "link[rel=\"canonical\"]",
    type: "attr",
    attribute: "href",
  },
};

async function scrape(client, url) {
  let res;
  try {
    // client.scrape POSTs JSON when extract_rules is set (scrapeWithRules would send it as a GET query, which the API ignores)
    res = await client.scrape(url, { ...SETTINGS, extract_rules: EXTRACT_RULES });
  } catch (err) {
    const status = err.response?.status ?? 'error';
    const message = err.response?.data?.message ?? err.message;
    console.error(`HTTP ${status} for ${url}: ${String(message).slice(0, 300)}`);
    process.exit(1);
  }
  console.error(`Ujb-Credits: ${res.headers['ujb-credits']}`);
  return res.data.result;
}

if (!API_KEY) {
  console.error('Set the UJEEBU_API_KEY environment variable.');
  process.exit(1);
}

const client = new UjeebuClient(API_KEY);
const pause = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

for (const [i, url] of URLS.entries()) {
  if (i) await pause(2000); // be polite: one page at a time
  console.log(JSON.stringify(await scrape(client, url), null, 2));
}

# Scrape CNBC articles with the Ujeebu Scrape API.
# Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
# Run: UJEEBU_API_KEY=your_key python3 cnbc_detail.py   (pip install requests)

import json
import os
import sys
import time

import requests

API_URL = "https://api.ujeebu.com/scrape"
API_KEY = os.environ.get("UJEEBU_API_KEY")
URLS = [
    "https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html",
    "https://www.cnbc.com/2026/04/21/spacex-says-it-can-buy-cursor-later-this-year-for-60-billion-or-pay-10-billion-for-our-work-together.html",
]
SETTINGS = {
    "js": True,
}
EXTRACT_RULES = {
    "article": {
        "type": "fn",
        "fn": "const T = ['NewsArticle', 'Article', 'BlogPosting', 'ReportageNews', 'AnalysisNewsArticle', 'TechArticle', 'ScholarlyArticle', 'LiveBlogPosting', 'OpinionNewsArticle', 'Report']; const n = [...document.querySelectorAll('script[type=\"application/ld+json\"]')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x['@graph'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x['@type']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split('|').map(q => q.split('.').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== ''); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(', ') : (v && typeof v === 'object' ? (v.name ?? v['@id'] ?? null) : (typeof v === 'string' ? v.replace('https://schema.org/', '').replace('http://schema.org/', '') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g('description')), headline: all(g('headline|name')), image: one(g('image.url|image')), modified: all(g('dateModified')), published: all(g('datePublished')), publisher: all(g('publisher.name')), section: all(g('articleSection')) };",
    },
    "url": {
        "selector": "link[rel=\"canonical\"]",
        "type": "attr",
        "attribute": "href",
    },
}


def scrape(url):
    response = requests.post(
        API_URL,
        headers={"ApiKey": API_KEY},
        json={"url": url, **SETTINGS, "extract_rules": EXTRACT_RULES},
        timeout=180,
    )
    if response.status_code != 200:
        try:
            message = response.json().get("message", response.text)
        except ValueError:
            message = response.text
        sys.exit(f"HTTP {response.status_code} for {url}: {str(message)[:300]}")
    print(f"Ujb-Credits: {response.headers.get('Ujb-Credits')}", file=sys.stderr)
    return response.json()["result"]


if not API_KEY:
    sys.exit("Set the UJEEBU_API_KEY environment variable.")

for i, url in enumerate(URLS):
    if i:
        time.sleep(2)  # be polite: one page at a time
    print(json.dumps(scrape(url), indent=2, ensure_ascii=False))

// Scrape CNBC articles with the Ujeebu Scrape API.
// Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
// Run: UJEEBU_API_KEY=your_key node cnbc_detail.mjs   (Node.js 18 or later)

const API_URL = 'https://api.ujeebu.com/scrape';
const API_KEY = process.env.UJEEBU_API_KEY;
const URLS = [
  "https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html",
  "https://www.cnbc.com/2026/04/21/spacex-says-it-can-buy-cursor-later-this-year-for-60-billion-or-pay-10-billion-for-our-work-together.html",
];
const SETTINGS = {
  js: true,
};
const EXTRACT_RULES = {
  article: {
    type: "fn",
    fn: "const T = ['NewsArticle', 'Article', 'BlogPosting', 'ReportageNews', 'AnalysisNewsArticle', 'TechArticle', 'ScholarlyArticle', 'LiveBlogPosting', 'OpinionNewsArticle', 'Report']; const n = [...document.querySelectorAll('script[type=\"application/ld+json\"]')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x['@graph'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x['@type']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split('|').map(q => q.split('.').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== ''); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(', ') : (v && typeof v === 'object' ? (v.name ?? v['@id'] ?? null) : (typeof v === 'string' ? v.replace('https://schema.org/', '').replace('http://schema.org/', '') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g('description')), headline: all(g('headline|name')), image: one(g('image.url|image')), modified: all(g('dateModified')), published: all(g('datePublished')), publisher: all(g('publisher.name')), section: all(g('articleSection')) };",
  },
  url: {
    selector: "link[rel=\"canonical\"]",
    type: "attr",
    attribute: "href",
  },
};

async function scrape(url) {
  const res = await fetch(API_URL, {
    method: 'POST',
    headers: { ApiKey: API_KEY, 'Content-Type': 'application/json' },
    body: JSON.stringify({ url, ...SETTINGS, extract_rules: EXTRACT_RULES }),
    signal: AbortSignal.timeout(180000),
  });
  const text = await res.text();
  if (res.status !== 200) {
    let message = text;
    try {
      message = JSON.parse(text).message ?? text;
    } catch {}
    console.error(`HTTP ${res.status} for ${url}: ${String(message).slice(0, 300)}`);
    process.exit(1);
  }
  console.error(`Ujb-Credits: ${res.headers.get('ujb-credits')}`);
  return JSON.parse(text).result;
}

if (!API_KEY) {
  console.error('Set the UJEEBU_API_KEY environment variable.');
  process.exit(1);
}

const pause = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

for (const [i, url] of URLS.entries()) {
  if (i) await pause(2000); // be polite: one page at a time
  console.log(JSON.stringify(await scrape(url), null, 2));
}

#!/usr/bin/env bash
# Scrape CNBC articles with the Ujeebu Scrape API.
# Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
# Run: UJEEBU_API_KEY=your_key bash cnbc_detail.sh   (curl 7.76 or later)
set -euo pipefail

curl -sS --fail-with-body --max-time 180 -D headers.txt \
  -X POST 'https://api.ujeebu.com/scrape' \
  -H "ApiKey: ${UJEEBU_API_KEY:?Set the UJEEBU_API_KEY environment variable.}" \
  -H 'Content-Type: application/json' \
  --data-binary @- <<'JSON'
{
  "url": "https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html",
  "js": true,
  "extract_rules": {
    "article": {
      "type": "fn",
      "fn": "const T = ['NewsArticle', 'Article', 'BlogPosting', 'ReportageNews', 'AnalysisNewsArticle', 'TechArticle', 'ScholarlyArticle', 'LiveBlogPosting', 'OpinionNewsArticle', 'Report']; const n = [...document.querySelectorAll('script[type=\"application/ld+json\"]')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x['@graph'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x['@type']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split('|').map(q => q.split('.').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== ''); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(', ') : (v && typeof v === 'object' ? (v.name ?? v['@id'] ?? null) : (typeof v === 'string' ? v.replace('https://schema.org/', '').replace('http://schema.org/', '') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g('description')), headline: all(g('headline|name')), image: one(g('image.url|image')), modified: all(g('dateModified')), published: all(g('datePublished')), publisher: all(g('publisher.name')), section: all(g('articleSection')) };"
    },
    "url": {
      "selector": "link[rel=\"canonical\"]",
      "type": "attr",
      "attribute": "href"
    }
  }
}
JSON
echo
grep -i '^ujb-credits' headers.txt >&2

<?php
// Scrape CNBC articles with the Ujeebu Scrape API.
// Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
// Run: UJEEBU_API_KEY=your_key php cnbc_detail.php   (PHP 7.4 or later with curl)

$apiKey = getenv('UJEEBU_API_KEY') ?: exit("Set the UJEEBU_API_KEY environment variable.\n");
$urls = [
    'https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html',
    'https://www.cnbc.com/2026/04/21/spacex-says-it-can-buy-cursor-later-this-year-for-60-billion-or-pay-10-billion-for-our-work-together.html',
];
$settings = [
    'js' => true,
];
$extractRules = [
    'article' => [
        'type' => 'fn',
        'fn' => 'const T = [\'NewsArticle\', \'Article\', \'BlogPosting\', \'ReportageNews\', \'AnalysisNewsArticle\', \'TechArticle\', \'ScholarlyArticle\', \'LiveBlogPosting\', \'OpinionNewsArticle\', \'Report\']; const n = [...document.querySelectorAll(\'script[type="application/ld+json"]\')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x[\'@graph\'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x[\'@type\']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split(\'|\').map(q => q.split(\'.\').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== \'\'); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(\', \') : (v && typeof v === \'object\' ? (v.name ?? v[\'@id\'] ?? null) : (typeof v === \'string\' ? v.replace(\'https://schema.org/\', \'\').replace(\'http://schema.org/\', \'\') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g(\'description\')), headline: all(g(\'headline|name\')), image: one(g(\'image.url|image\')), modified: all(g(\'dateModified\')), published: all(g(\'datePublished\')), publisher: all(g(\'publisher.name\')), section: all(g(\'articleSection\')) };',
    ],
    'url' => [
        'selector' => 'link[rel="canonical"]',
        'type' => 'attr',
        'attribute' => 'href',
    ],
];

function scrape(string $url, string $apiKey, array $settings, array $extractRules)
{
    $ch = curl_init('https://api.ujeebu.com/scrape');
    $headers = [];
    curl_setopt_array($ch, [
        CURLOPT_POST => true,
        CURLOPT_RETURNTRANSFER => true,
        CURLOPT_TIMEOUT => 180,
        CURLOPT_HTTPHEADER => ['ApiKey: ' . $apiKey, 'Content-Type: application/json'],
        CURLOPT_POSTFIELDS => json_encode(['url' => $url] + $settings + ['extract_rules' => $extractRules]),
        CURLOPT_HEADERFUNCTION => function ($ch, $line) use (&$headers) {
            if (strpos($line, ':') !== false) {
                [$k, $v] = explode(':', $line, 2);
                $headers[strtolower(trim($k))] = trim($v);
            }
            return strlen($line);
        },
    ]);
    $body = curl_exec($ch);
    if ($body === false) {
        fwrite(STDERR, 'Request failed: ' . curl_error($ch) . "\n");
        exit(1);
    }
    $status = curl_getinfo($ch, CURLINFO_RESPONSE_CODE);
    if ($status !== 200) {
        $message = json_decode($body, true)['message'] ?? $body;
        fwrite(STDERR, "HTTP $status for $url: " . substr((string) $message, 0, 300) . "\n");
        exit(1);
    }
    fwrite(STDERR, 'Ujb-Credits: ' . ($headers['ujb-credits'] ?? '?') . "\n");
    return json_decode($body, true)['result'];
}

foreach ($urls as $i => $url) {
    if ($i > 0) {
        sleep(2); // be polite: one page at a time
    }
    echo json_encode(scrape($url, $apiKey, $settings, $extractRules), JSON_PRETTY_PRINT | JSON_UNESCAPED_SLASHES | JSON_UNESCAPED_UNICODE), "\n";
}

// Scrape CNBC articles with the Ujeebu Scrape API.
// Settings and extract rules validated on real pages on 2026-10-13 (Ujeebu site guide).
// Uses the official SDK github.com/ujeebu/ujeebu-go v0.1.8: go mod init scraper && go get github.com/ujeebu/ujeebu-go@v0.1.8
// Run: UJEEBU_API_KEY=your_key go run cnbc_detail.go

package main

import (
	"encoding/json"
	"errors"
	"fmt"
	"os"
	"time"

	ujeebu "github.com/ujeebu/ujeebu-go"
)

var urls = []string{
	"https://www.cnbc.com/2025/08/21/wealthy-investors-etfs-capital-gains-taxes-.html",
	"https://www.cnbc.com/2026/04/21/spacex-says-it-can-buy-cursor-later-this-year-for-60-billion-or-pay-10-billion-for-our-work-together.html",
}

// The extract rules of the recipe, as the JSON the API takes.
const extractRulesJSON = `{
    "article": {
        "type": "fn",
        "fn": "const T = ['NewsArticle', 'Article', 'BlogPosting', 'ReportageNews', 'AnalysisNewsArticle', 'TechArticle', 'ScholarlyArticle', 'LiveBlogPosting', 'OpinionNewsArticle', 'Report']; const n = [...document.querySelectorAll('script[type=\"application/ld+json\"]')].flatMap(s => { try { return [].concat(JSON.parse(s.textContent)).flatMap(x => x['@graph'] || [x]); } catch (e) { return []; } }).find(x => x && [].concat(x['@type']).some(t => T.includes(t))); if (!n) return null; const g = p => p.split('|').map(q => q.split('.').reduce((v, k) => v == null ? v : (Array.isArray(v) ? v[0] : v)?.[k], n)).find(v => v != null && v !== ''); const all = v => Array.isArray(v) ? v.map(all).filter(x => x != null).join(', ') : (v && typeof v === 'object' ? (v.name ?? v['@id'] ?? null) : (typeof v === 'string' ? v.replace('https://schema.org/', '').replace('http://schema.org/', '') : v ?? null)); const one = v => all(Array.isArray(v) ? v[0] : v); return { description: all(g('description')), headline: all(g('headline|name')), image: one(g('image.url|image')), modified: all(g('dateModified')), published: all(g('datePublished')), publisher: all(g('publisher.name')), section: all(g('articleSection')) };"
    },
    "url": {
        "selector": "link[rel=\"canonical\"]",
        "type": "attr",
        "attribute": "href"
    }
}`

var extractRules map[string]any

func fail(format string, args ...any) {
	fmt.Fprintf(os.Stderr, format+"\n", args...)
	os.Exit(1)
}

func short(s string) string {
	if len(s) > 300 {
		return s[:300]
	}
	return s
}

func params(target string) ujeebu.ScrapeParams {
	p := ujeebu.ScrapeParams{URL: target}
	p.JS = true
	p.ExtractRules = extractRules
	return p
}

func scrape(client *ujeebu.Client, target string) map[string]any {
	// the int is the credits the request cost (the Ujb-Credits header), not an HTTP status
	resp, credits, err := client.Scrape(params(target))
	if err != nil {
		var apiErr *ujeebu.APIError
		if errors.As(err, &apiErr) {
			fail("HTTP %d for %s: %s", apiErr.StatusCode, target, short(apiErr.Message))
		}
		fail("Request failed for %s: %v", target, err)
	}
	fmt.Fprintf(os.Stderr, "Ujb-Credits: %d\n", credits)
	result, _ := resp.Result.(map[string]any)
	return result
}

func printJSON(v any) {
	enc := json.NewEncoder(os.Stdout)
	enc.SetEscapeHTML(false)
	enc.SetIndent("", "  ")
	if err := enc.Encode(v); err != nil {
		fail("Cannot print the result: %v", err)
	}
}

func main() {
	apiKey := os.Getenv("UJEEBU_API_KEY")
	if apiKey == "" {
		fail("Set the UJEEBU_API_KEY environment variable.")
	}
	if err := json.Unmarshal([]byte(extractRulesJSON), &extractRules); err != nil {
		fail("Bad extract rules: %v", err)
	}
	client, err := ujeebu.NewClient(apiKey, ujeebu.WithTimeout(180*time.Second))
	if err != nil {
		fail("%v", err)
	}

	for i, target := range urls {
		if i > 0 {
			time.Sleep(2 * time.Second) // be polite: one page at a time
		}
		printJSON(scrape(client, target))
	}
}