{ "name": "Job Pipeline - Phase 1", "nodes": [ { "parameters": {}, "type": "n8n-nodes-base.manualTrigger", "typeVersion": 1, "position": [ -512, -16 ], "id": "c9f3e062-4f5c-439f-95a7-ed2f5b69d1ed", "name": "When clicking \u2018Execute workflow\u2019" }, { "parameters": { "method": "POST", "url": "http://jobspy:8000/scrape", "sendBody": true, "specifyBody": "json", "jsonBody": "={{ JSON.stringify($json) }}", "options": { "timeout": 300000 } }, "type": "n8n-nodes-base.httpRequest", "typeVersion": 4.4, "position": [ 0, -16 ], "id": "609a3975-30ec-49bc-bd1d-41b815c2efe0", "name": "HTTP Request" }, { "parameters": { "jsCode": "// Aggregate gives us a single item with `responses[]` (8 entries from HTTP loop).\n// Flatten all jobs, dedupe on job_url, emit one item per unique job in\n// the column shape Google Sheets expects.\n\nconst responses = $input.first().json.responses || [];\n\nconst allJobs = [];\nfor (const resp of responses) {\n const jobs = resp.jobs || [];\n for (const job of jobs) {\n allJobs.push({\n ...job,\n scraped_at: resp.scraped_at,\n role_family: job.role_family || resp.role_family,\n search_term: job.search_term || resp.search_term,\n geo_searched: job.geo_searched || resp.location,\n });\n }\n}\n\n// Dedupe on job_url \u2014 the same job can appear in multiple role/geo searches.\nconst seen = new Set();\nconst unique = [];\nfor (const job of allJobs) {\n const key = job.job_url;\n if (!key || seen.has(key)) continue;\n seen.add(key);\n unique.push(job);\n}\n\n// Emit one item per unique job in the shape Google Sheets expects.\nreturn unique.map(job => ({\n json: {\n scraped_at: job.scraped_at,\n role_family: job.role_family || '',\n site: job.site || '',\n title: job.title || '',\n company: job.company || '',\n location: job.location || '',\n date_posted: job.date_posted || '',\n is_remote: job.is_remote === true ? 'TRUE' : 'FALSE',\n job_url: job.job_url || '',\n min_amount: job.min_amount || '',\n max_amount: job.max_amount || '',\n currency: job.currency || '',\n search_term: job.search_term || '',\n },\n}));" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 496, -16 ], "id": "1c08f75a-6bc7-4f2a-a4ee-d8c389837e8d", "name": "Code in JavaScript" }, { "parameters": { "jsCode": "// Emit one item per (role_family \u00d7 geo) combination.\n// n8n will then loop the HTTP Request node once per item.\n\nconst ROLE_FAMILIES = {\n BA: \"Business Analyst\",\n Data: \"Data Analyst\",\n CRM_Salesforce: \"Salesforce Administrator\",\n Systems: \"Systems Analyst\",\n};\n\nconst GEOS = [\n { location: \"Dublin, Ireland\", country_indeed: \"ireland\" },\n { location: \"London, United Kingdom\", country_indeed: \"uk\" },\n];\n\nconst items = [];\nfor (const [family, term] of Object.entries(ROLE_FAMILIES)) {\n for (const geo of GEOS) {\n items.push({\n json: {\n role_family: family,\n search_term: term,\n location: geo.location,\n country_indeed: geo.country_indeed,\n sites: [\"indeed\", \"linkedin\"],\n results_wanted: 15,\n hours_old: 168,\n save_csv: true,\n },\n });\n }\n}\nreturn items;" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ -256, -16 ], "id": "716a4abe-a556-4a9d-bca9-5026ae59e227", "name": "Build search configs" }, { "parameters": { "aggregate": "aggregateAllItemData", "destinationFieldName": "responses", "options": {} }, "type": "n8n-nodes-base.aggregate", "typeVersion": 1, "position": [ 256, -16 ], "id": "34496538-61fa-44b3-aa2a-e1d9c8e5c1d9", "name": "Aggregate" }, { "parameters": { "operation": "appendOrUpdate", "documentId": { "__rl": true, "value": "https://docs.google.com/spreadsheets/d/1Z3qbKjYuKA1EFids9QYppzjuaiMf9quxN8hw_chZhfM/edit?gid=0#gid=0", "mode": "url" }, "sheetName": { "__rl": true, "value": "gid=0", "mode": "list", "cachedResultName": "Sheet1", "cachedResultUrl": "https://docs.google.com/spreadsheets/d/1Z3qbKjYuKA1EFids9QYppzjuaiMf9quxN8hw_chZhfM/edit#gid=0" }, "columns": { "mappingMode": "autoMapInputData", "value": {}, "matchingColumns": [ "job_url" ], "schema": [ { "id": "scraped_at", "displayName": "scraped_at", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "role_family", "displayName": "role_family", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "site", "displayName": "site", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "title", "displayName": "title", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "company", "displayName": "company", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true, "removed": false }, { "id": "location", "displayName": "location", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true, "removed": false }, { "id": "date_posted", "displayName": "date_posted", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "is_remote", "displayName": "is_remote", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "job_url", "displayName": "job_url", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true, "removed": false }, { "id": "min_amount", "displayName": "min_amount", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "max_amount", "displayName": "max_amount", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "currency", "displayName": "currency", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true }, { "id": "search_term", "displayName": "search_term", "required": false, "defaultMatch": false, "display": true, "type": "string", "canBeUsedToMatch": true } ], "attemptToConvertTypes": false, "convertFieldsToString": false }, "options": {} }, "type": "n8n-nodes-base.googleSheets", "typeVersion": 4.7, "position": [ 704, -16 ], "id": "de02ed86-cc4f-4e8d-8227-020469342c4d", "name": "Append or update row in sheet", "credentials": { "googleSheetsOAuth2Api": { "name": "" } } } ], "connections": { "When clicking \u2018Execute workflow\u2019": { "main": [ [ { "node": "Build search configs", "type": "main", "index": 0 } ] ] }, "HTTP Request": { "main": [ [ { "node": "Aggregate", "type": "main", "index": 0 } ] ] }, "Code in JavaScript": { "main": [ [ { "node": "Append or update row in sheet", "type": "main", "index": 0 } ] ] }, "Build search configs": { "main": [ [ { "node": "HTTP Request", "type": "main", "index": 0 } ] ] }, "Aggregate": { "main": [ [ { "node": "Code in JavaScript", "type": "main", "index": 0 } ] ] } }, "active": false, "settings": { "executionOrder": "v1", "binaryMode": "separate" }, "versionId": "fb52e649-425f-4d7c-b424-b6fd5d28f531", "meta": { "templateCredsSetupCompleted": true }, "id": "f2xuFZxItrETLZlo", "tags": [] }