n8n nodes for scraping linkedin using openai to google sheets

n8n nodes for scraping linkedin using openai to google sheets.js

{
  "meta": {
    "instanceId": "5970330cf84a96bb271f539d4bac6134d8132cbad51d8237733e8f17982e0bdc"
  },
  "nodes": [
    {
      "parameters": {},
      "id": "bb2d4060-f43f-4393-800b-55bf0f44ddb0",
      "name": "Execute Workflow Trigger",
      "type": "n8n-nodes-base.executeWorkflowTrigger",
      "typeVersion": 1,
      "position": [
        820,
        320
      ]
    },
    {
      "parameters": {
        "url": "https://www.google.com/search",
        "sendQuery": true,
        "queryParameters": {
          "parameters": [
            {
              "name": "q",
              "value": "=site:linkedin.com/in/ {{ $json.message.content.jobTitle }} + {{ $json.message.content.industry }} + {{ $json.message.content.location }}"
            }
          ]
        },
        "sendHeaders": true,
        "headerParameters": {
          "parameters": [
            {
              "name": "user-agent",
              "value": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/117.0.5938.132 Safari/537.36"
            },
            {}
          ]
        },
        "options": {}
      },
      "id": "082e178f-69f7-4253-a75a-9a70b7a819eb",
      "name": "HTTP Request",
      "type": "n8n-nodes-base.httpRequest",
      "typeVersion": 4.2,
      "position": [
        1380,
        320
      ]
    },
    {
      "parameters": {
        "modelId": {
          "__rl": true,
          "value": "gpt-3.5-turbo",
          "mode": "list",
          "cachedResultName": "GPT-3.5-TURBO"
        },
        "messages": {
          "values": [
            {
              "content": "=Parse the JSON \"query\" and output the following parameters separately:  \n\njobTitle \ncompany\nIndustry \nlocation\n",
              "role": "system"
            },
            {
              "content": "={{JSON.stringify ($json.query) }}"
            }
          ]
        },
        "jsonOutput": true,
        "options": {}
      },
      "id": "323b4e32-526f-4ffe-b5e1-ecf6bca39e8a",
      "name": "OpenAI1",
      "type": "@n8n/n8n-nodes-langchain.openAi",
      "typeVersion": 1.5,
      "position": [
        1000,
        320
      ],
      "credentials": {
        "openAiApi": {
          "id": "yN1DQVBvF8QW7j77",
          "name": "OpenAi account"
        }
      }
    },
    {
      "parameters": {
        "jsCode": "// 'items' is the input provided by n8n, containing data from the previous node (HTTP Request in this case)\n\n// Assuming the HTTP Request data is available in items[0].json.data\n// Modify this if your actual structure is different\nconst httpRequestData = items[0].json.data;\n\n// Regular expression to match LinkedIn URLs\nconst linkedinUrlPattern = /(https?:\\/\\/[a-zA-Z0-9.-]*linkedin\\.com\\/in\\/[a-zA-Z0-9-]+)/g;\n\n// Function to extract unique LinkedIn URLs from the request data\nfunction extractUniqueUrls(data) {\n    const matches = [...data.matchAll(linkedinUrlPattern)]; // Use matchAll to get all matches\n    const urls = matches.map(match => match[0]); // Extract full URL (group 0)\n    const uniqueUrls = [...new Set(urls)]; // Remove duplicates using Set\n    return uniqueUrls;\n}\n\n// Extract unique URLs from the HTTP request data\nconst linkedinUrls = extractUniqueUrls(httpRequestData);\n\n// Prepare the output in n8n's expected format: an array of objects with unique URLs\nconst output = linkedinUrls.map(url => {\n    return { json: { url } };  // each URL becomes an object with a `json` key\n});\n\n// Return the output array to n8n\nreturn output;\n"
      },
      "id": "bb6b77a0-94ec-4270-b3d8-e1419a462ede",
      "name": "Code",
      "type": "n8n-nodes-base.code",
      "typeVersion": 2,
      "position": [
        1600,
        320
      ]
    },
    {
      "parameters": {
        "operation": "append",
        "documentId": {
          "__rl": true,
          "value": "1GHALx9nA0MTB7gP16rxj8R6afH-tFb-Qvo3DK34M_uI",
          "mode": "id"
        },
        "sheetName": {
          "__rl": true,
          "value": "Sheet1",
          "mode": "name"
        },
        "columns": {
          "mappingMode": "autoMapInputData",
          "value": {},
          "matchingColumns": [],
          "schema": [
            {
              "id": "URLs",
              "displayName": "URLs",
              "required": false,
              "defaultMatch": false,
              "display": true,
              "type": "string",
              "canBeUsedToMatch": true,
              "removed": false
            },
            {
              "id": "url",
              "displayName": "url",
              "required": false,
              "defaultMatch": false,
              "display": true,
              "type": "string",
              "canBeUsedToMatch": true,
              "removed": false
            }
          ]
        },
        "options": {}
      },
      "id": "d85eed9f-5b35-4d32-b11d-a50be855da0a",
      "name": "Google Sheets",
      "type": "n8n-nodes-base.googleSheets",
      "typeVersion": 4.5,
      "position": [
        1820,
        320
      ],
      "credentials": {
        "googleSheetsOAuth2Api": {
          "id": "Eomtk076fGT5DGhI",
          "name": "Google Sheets account"
        }
      }
    },
    {
      "parameters": {
        "assignments": {
          "assignments": [
            {
              "id": "e9931689-079e-40ae-9d2d-aca3255afa78",
              "name": "response",
              "value": "done",
              "type": "string"
            }
          ]
        },
        "options": {}
      },
      "id": "d9f4c124-6fb7-4dd0-b72e-42c086733243",
      "name": "Edit Fields",
      "type": "n8n-nodes-base.set",
      "typeVersion": 3.4,
      "position": [
        2040,
        320
      ]
    }
  ],
  "connections": {
    "Execute Workflow Trigger": {
      "main": [
        [
          {
            "node": "OpenAI1",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "HTTP Request": {
      "main": [
        [
          {
            "node": "Code",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "OpenAI1": {
      "main": [
        [
          {
            "node": "HTTP Request",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Code": {
      "main": [
        [
          {
            "node": "Google Sheets",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Google Sheets": {
      "main": [
        [
          {
            "node": "Edit Fields",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  },
  "pinData": {
    "Execute Workflow Trigger": [
      {
        "query": "jobTitle=&industry=&location=&"
      }
    ]
  }
}