Scraper les données d’entreprise LinkedIn avec Python

1python -m venv .venv
2source .venv/bin/activate
3python -m pip install requests
4export DATABLIST_API_KEY="your-api-key"

1import json
2import os
3import time
4import uuid
5
6import requests
7
8
9accounts = [
10    {
11        "account_id": "account-n8n",
12        "linkedin_company_url": "https://www.linkedin.com/company/n8n/",
13    },
14    {
15        "account_id": "account-posthog",
16        "linkedin_company_url": "https://www.linkedin.com/company/posthog/",
17    },
18    {
19        "account_id": "account-back-market",
20        "linkedin_company_url": "https://www.linkedin.com/company/back-market/",
21    },
22    {
23        "account_id": "account-datablist",
24        "linkedin_company_url": "https://www.linkedin.com/company/datablist/",
25    },
26]
27
28
29def scrape_companies(accounts, api_key):
30    """Submit one task, wait for it to finish, and return every result."""
31    tasks_url = "https://api.datablist.com/v1/enrichments/tasks"
32    headers = {"X-API-Key": api_key}
33    inputs = [
34        {
35            "inputId": account["account_id"],
36            "data": {"companyLinkedInUrl": account["linkedin_company_url"]},
37        }
38        for account in accounts
39    ]
40
41    unique_id = str(uuid.uuid4())
42    print(f"Request ID: {unique_id}")
43    payload = {
44        "enrichmentId": "linkedin-company-enrichment",
45        "uniqueId": unique_id,
46        "settings": {},
47        "inputs": inputs,
48    }
49
50    response = requests.post(tasks_url, headers=headers, json=payload)
51    response.raise_for_status()
52    task = response.json()
53    task_id = task["@id"]
54    task_url = f"{tasks_url}/{task_id}"
55    print(f"Task ID: {task_id}")
56
57    while task["status"] not in ("completed", "failed", "stopped"):
58        time.sleep(3)
59        response = requests.get(task_url, headers=headers)
60        response.raise_for_status()
61        task = response.json()
62
63    if task["status"] != "completed":
64        raise RuntimeError(
65            f"Task {task_id} {task['status']}: {task.get('error')}"
66        )
67
68    results = []
69    results_url = f"{task_url}/results"
70    params = {"limit": 100}
71    while True:
72        response = requests.get(results_url, headers=headers, params=params)
73        response.raise_for_status()
74        page = response.json()
75        results.extend(page["results"])
76        next_cursor = page["nextCursor"]
77        if next_cursor is None:
78            break
79        params["cursor"] = next_cursor
80
81    return results
82
83
84api_key = os.environ.get("DATABLIST_API_KEY")
85if not api_key:
86    raise RuntimeError("Set DATABLIST_API_KEY before running this script.")
87
88company_results = scrape_companies(accounts, api_key)
89print(json.dumps(company_results, indent=2, ensure_ascii=False))

1python scrape_linkedin_companies.py

{
  "taskId": "69516a973dc34481920f41ade748e917",
  "results": [
    {
      "inputId": "account-n8n",
      "status": "success",
      "outputs": {
        "companyName": "n8n",
        "companyHQPostalCode": "10115",
        "companyDescription": "n8n is a workflow automation platform that uniquely combines AI capabilities with business process automation. The platform enables connection to any app or API while maintaining the flexibility of code with the speed of no-code.\n\nReleased under a fair-code license, n8n can be self-hosted and is supported by a vibrant community of developers and builders. Users can start simple and layer complexity as needed - utilizing the visual builder for quick wins, and adding custom Javascript or Python code where more control is required, enabling the connection of anything to everything.\n\nn8n is privately held, with funding from Sequoia, Felicis Ventures, Firstminute Capital, Harpoon Ventures, and others. Individual investors include Eventbrite's Kevin Hartz and Supercell's IIkka Paananen. n8n was founded in 2019 and is headquartered in Berlin.",
        "companyTagline": "Flexible AI workflow automation for technical teams.",
        "companyIndustry": "Software Development",
        "companySpecialities": "AI, SOAR, Workflow Automation, and AI agents",
        "companyWebsite": "https://n8n.io",
        "companyHQCity": "Berlin",
        "linkedInCompanyId": "28491094",
        "companyType": "Privately Held",
        "companyHQCountry": "DE",
        "companySize": 1389,
        "companyLogoUrl": "https://media.licdn.com/dms/image/v2/D4D0BAQEgEphZqatXwA/company-logo_200_200/B4DaFY.Sa2JcAE-/0/1792494086070/n8n_logo?e=2147483647&v=beta&t=NTL4C6f5QAd_o9nXd15PO8pOMSSEoms_tJdDHEo6eLk",
        "companyHQAddress": "Novalisstraße 10",
        "companyFoundedYear": 2019,
        "companyFollowerCount": 442229
      },
      "creditsUsed": "25",
      "error": null
    },
    {
      "inputId": "account-posthog",
      "status": "success",
      "outputs": {
        "companyName": "PostHog",
        "companyHQPostalCode": null,
        "companyDescription": "At PostHog, we're working to increase the number of successful products in the world.\n\nUntil now, tools for building products have been fragmented. Product analytics, heatmaps, session recording, web analytics, feature flags, and A/B testing are all helpful, but no one wants to buy, send data to, and integrate multiple products.\n\nPostHog offers these tools (and more) in an integrated, open source platform which can be hosted in either the US or EU. Both versions are SOC2 certified, GDPR-ready, and HIPAA compliant. \n\nWe started PostHog during Y Combinator's W20 cohort and had the most successful B2B software launch on Hacker News since 2012 - with a product that was just 4 weeks old.\n\nWith over 100,000 users, we're default alive, growing 97% through word of mouth, and we are in the top 0.01% most popular repos on GitHub.",
        "companyTagline": "Your product's context layer. Get started with npx @posthog/wizard",
        "companyIndustry": "Software Development",
        "companySpecialities": "analytics, open source, product analytics, product, data, engineering, self-driving product, session replay, context warehouse, AI, product context, and error tracking",
        "companyWebsite": "https://posthog.com/",
        "companyHQCity": "San Francisco",
        "linkedInCompanyId": "37415928",
        "companyType": "Privately Held",
        "companyHQCountry": "US",
        "companySize": 234,
        "companyLogoUrl": "https://media.licdn.com/dms/image/v2/D560BAQGQUyE_CXUxIw/company-logo_200_200/B56aCMWEgaJAAE-/0/1789060991251/posthog_logo?e=2147483647&v=beta&t=WfTGRSqp5UfYL33sBXx3tZ1zmFUECfUPf7JtEGPAvbA",
        "companyHQAddress": null,
        "companyFoundedYear": 2020,
        "companyFollowerCount": 59696
      },
      "creditsUsed": "25",
      "error": null
    },
    {
      "inputId": "account-back-market",
      "status": "success",
      "outputs": {
        "companyName": "Back Market",
        "companyHQPostalCode": null,
        "companyDescription": "The world is waking up to what fast tech costs. Back Market is where the revolution shops. \nWe make refurbished tech the smart, confident choice by building trust in quality, repair and circularity. New is not always better. Smarter is.\n\nAs a certified B Corp, our mission is to create a world that does more with what we already have by prolonging the lifespan of electronic devices through circularity and repair. We have grown by double digits year over year and now serve 18 million customers across 17 markets, avoiding more than 2 million tons of CO2 emissions*\n*Back Market calculation based on ADEME 2022 methodology.\n\nFounded in Paris in 2014 by Thibaud Hug de Larauze, Quentin Le Brouster and Vianney Vaute, Back Market is a leading global marketplace where people can buy professionally refurbished technology or trade in their devices.\n\nAre you ready to join us?",
        "companyTagline": "Where the world shops refurbished tech.",
        "companyIndustry": "Internet Marketplace Platforms",
        "companySpecialities": "refurbished tech, marketplace, sustainability, circularity, Right to Repair, sustainable solutions, refurbished marketplace, circular economy, and sustainable alternatives",
        "companyWebsite": "http://www.backmarket.com",
        "companyHQCity": "Paris",
        "linkedInCompanyId": "4793443",
        "companyType": "Privately Held",
        "companyHQCountry": "FR",
        "companySize": 1054,
        "companyLogoUrl": "https://media.licdn.com/dms/image/v2/D4E0BAQHRv6UClt5D2g/company-logo_200_200/B4EZnoUG3tKgAI-/0/1760539209613/back_market_logo?e=2147483647&v=beta&t=Pywpeu1Qs3h0Cl8kgvYurmBfXbImm5QIIc-aQog1MqM",
        "companyHQAddress": null,
        "companyFoundedYear": 2014,
        "companyFollowerCount": 101878
      },
      "creditsUsed": "25",
      "error": null
    },
    {
      "inputId": "account-datablist",
      "status": "success",
      "outputs": {
        "companyName": "Datablist",
        "companyHQPostalCode": null,
        "companyDescription": "When people hear \"data,\" they think of something complicated. We will change this.\n\nDatablist is an AI-powered spreadsheet that helps you: get, manipulate, or clean data and create workflows that run automatically while you focus on what matters most.\n\nIf you want to separate Datablist.com into its main parts, you can say we have:\n\n- Deduplicating, grouping, and labeling data\n- Cleaning and editing data\n- Managing data and automated workflows\n- 50+ B2B data sources and scrapers\n- 3 AI agents:\n    - Editing assistant\n    - AI Research agent\n    - Deduplication assistant\n\nThe top 3 reasons why companies choose Datablist.com are:\n\n1. Simplicity: you don't need to have technical knowledge or experience with formulas to get everything out of it.\n2. Flexibility that allows you to do (almost) everything with your data and AI while keeping everything under your control.\n3. Scalability that no other solution can match. Up to 2 million rows—forget Excel.\n\nGo on try it for free!\n\nhttps://www.Datablist.com",
        "companyTagline": "Let your tasks become workflows: automated workflows.",
        "companyIndustry": "Information Technology & Services",
        "companyWebsite": "https://www.datablist.com",
        "companyHQCity": "Paris",
        "linkedInCompanyId": "69842428",
        "companyType": "Privately Held",
        "companyHQCountry": "FR",
        "companySize": 3,
        "companyLogoUrl": "https://media.licdn.com/dms/image/v2/C4E0BAQFNFlMb8KEPRQ/company-logo_200_200/company-logo_200_200/0/1630622803412?e=2147483647&v=beta&t=UVkk7BOerChiWgS9Y06zOpT08SlK4kpAWGDYiVvUEVU",
        "companyFollowerCount": 134,
        "companyFoundedYear": 2020,
        "companyHQAddress": null
      },
      "creditsUsed": "25",
      "error": null
    }
  ],
  "nextCursor": null
}

1accounts_by_id = {account["account_id"]: account for account in accounts}
2qualified_accounts = []
3other_accounts = []
4
5for result in company_results:
6    account = accounts_by_id.get(result["inputId"])
7    qualification = "needs_review"
8
9    if result["status"] == "success":
10        outputs = result["outputs"]
11        industry = outputs.get("companyIndustry")
12        company_size = outputs.get("companySize")
13        country = outputs.get("companyHQCountry")
14
15        if industry and isinstance(company_size, (int, float)) and country:
16            industry_matches = any(
17                keyword in industry.lower()
18                for keyword in ("software", "technology")
19            )
20            size_matches = 50 <= company_size <= 500
21            country_matches = country in {"US", "DE"}
22            if industry_matches and size_matches and country_matches:
23                qualification = "qualified"
24            else:
25                qualification = "not_qualified"
26
27    source_url = None
28    if account is not None:
29        source_url = account["linkedin_company_url"]
30
31    reviewed_account = dict(result)
32    reviewed_account["sourceLinkedInUrl"] = source_url
33    reviewed_account["qualification"] = qualification
34
35    if qualification == "qualified":
36        qualified_accounts.append(reviewed_account)
37    else:
38        other_accounts.append(reviewed_account)
39
40print(json.dumps(qualified_accounts, indent=2, ensure_ascii=False))

1curl "https://api.datablist.com/v1/enrichments/tasks?limit=100" \
2  -H "X-API-Key: $DATABLIST_API_KEY"