{ "edges": [ { "to": "19c8577a-7191-45c5-b6b8-60a8be70ed5d", "from": "6a607aac-b15f-46d8-901c-a53d9099a08d", "toPort": "0", "fromPort": "0" }, { "to": "48997584-e40b-4a2d-bcaa-f1de24a6ad6b", "from": "19c8577a-7191-45c5-b6b8-60a8be70ed5d", "toPort": "0", "fromPort": "0" }, { "to": "b346f0d8-f5c1-4161-af9f-75059735f0f4", "from": "aeefe4c7-1603-4e57-b7a0-d373dcddf67d", "toPort": "0", "fromPort": "0" }, { "to": "7eaefa3f-577d-4809-895b-49d03e23495a", "from": "f0fbb4ac-164c-483c-b0eb-2bfedab2e2a0", "toPort": "0", "fromPort": "0" }, { "to": "aeefe4c7-1603-4e57-b7a0-d373dcddf67d", "from": "7eaefa3f-577d-4809-895b-49d03e23495a", "toPort": "0", "fromPort": "0" }, { "to": "19c8577a-7191-45c5-b6b8-60a8be70ed5d", "from": "fd96243b-229e-436d-91ea-b31438f175a8", "toPort": "0", "fromPort": "ai_languageModel" }, { "to": "ba782266-418a-4d19-90bf-255f0f955cb9", "from": "28dba153-23d5-4c12-a786-605cb236040d", "toPort": "0", "fromPort": "0" }, { "to": "28dba153-23d5-4c12-a786-605cb236040d", "from": "48997584-e40b-4a2d-bcaa-f1de24a6ad6b", "toPort": "0", "fromPort": "0" }, { "to": "6a607aac-b15f-46d8-901c-a53d9099a08d", "from": "b346f0d8-f5c1-4161-af9f-75059735f0f4", "toPort": "0", "fromPort": "0" } ], "nodes": [ { "id": "6a607aac-b15f-46d8-901c-a53d9099a08d", "name": "Filter", "type": "passthrough", "params": { "mode": "filter", "options": {}, "conditions": { "options": { "version": 3, "leftValue": "", "caseSensitive": true, "typeValidation": "strict" }, "combinator": "and", "conditions": [ { "id": "f3c26f60-2368-4461-9f2e-3072acb1f7b1", "operator": { "type": "number", "operation": "gt" }, "leftValue": "={{ $json.analysis.velocity_score }}", "rightValue": 1 } ] } }, "position": [ 1200, 176 ] }, { "id": "f0fbb4ac-164c-483c-b0eb-2bfedab2e2a0", "name": "Schedule Trigger", "type": "trigger", "params": { "kind": "n8n-nodes-base.scheduleTrigger", "rule": { "interval": [ { "field": "hours", "hoursInterval": 4 } ] } }, "position": [ 368, 176 ] }, { "id": "ba782266-418a-4d19-90bf-255f0f955cb9", "name": "Send a text message", "type": "integration", "params": { "text": "={{ $json.text }}", "chatId": "YourChatID", "service": "telegram", "operation": "read", "additionalFields": { "parse_mode": "HTML" } }, "position": [ 2096, 176 ] }, { "id": "fd96243b-229e-436d-91ea-b31438f175a8", "name": "OpenRouter Chat Model1", "type": "llm", "params": { "model": "google/gemini-2.5-flash-lite", "options": {}, "provider": "openai" }, "position": [ 1248, 416 ] }, { "id": "7eaefa3f-577d-4809-895b-49d03e23495a", "name": "Algolia parameters", "type": "code", "params": { "jsCode": "// @see https://hn.algolia.com/api\n\nconst now = DateTime.now();\n// Observation window (past X hours)\nconst lookBackHours = 24;\nconst startTime = Math.floor(now.minus({hours: lookBackHours}).toSeconds())\n\nconst numericFilters = [\n `created_at_i>${startTime}`,\n // Filter out noise/spam (posts with zero engagement)\n \"points>1\",\n];\n\nreturn [{\n json: {\n tags: \"(ask_hn,show_hn)\",\n startTime: now.minus({hours: lookBackHours}),\n numericFilters: numericFilters.join(\",\"),\n // Fetch maximum allowed items per page\n hitsPerPage: 1000\n }\n}]" }, "position": [ 576, 176 ] }, { "id": "b346f0d8-f5c1-4161-af9f-75059735f0f4", "name": "Recalculate popularity score", "type": "code", "params": { "jsCode": "const hits = items[0].json.hits;\n\nconst now = DateTime.now();\n\nconst cleanData = hits.map(item => {\n const postedTime = DateTime.fromISO(item.created_at);\n // Calculate post age (in hours)\n const ageInHours = now.diff(postedTime, 'hours').hours;\n const points = item.points || 0;\n const comments = item.num_comments || 0;\n\n // --- Strategy 1: Infancy Filtering (The Buffer) ---\n // If the post is less than 1 hour old and has fewer than 5 points, it is still \"baking\"\n // and lacks analysis value for now. Skip it.\n if (ageInHours < 1.0 && points < 5) {\n return null; // Mark for removal\n }\n\n // --- Strategy 2: Recalculate Heat Score (Gravity Score) ---\n const rawScore = points + (comments * 2);\n const gravity = 1.8;\n const normalizedScore = (rawScore - 1) / Math.pow(ageInHours + 2, gravity);\n\n return {\n ...item,\n analysis: {\n age_hours: parseFloat(ageInHours.toFixed(2)),\n velocity_score: parseFloat(normalizedScore.toFixed(4)),\n }\n };\n})\n.filter(item => item !== null) // Filter out the new posts marked as null above\n.sort((a, b) => b.analysis.velocity_score - a.analysis.velocity_score);\n\nreturn cleanData.map(item => ({ json: item }));" }, "position": [ 992, 176 ] }, { "id": "19c8577a-7191-45c5-b6b8-60a8be70ed5d", "name": "Translate", "type": "llm", "params": { "text": "=title: {{ JSON.stringify($json.title) }}\nstory_text: {{ JSON.stringify($json.story_text) }}", "prompt": "=title: {{ JSON.stringify($json.title) }}\nstory_text: {{ JSON.stringify($json.story_text) }}", "options": { "systemMessage": "You are a professional news media translation API.\n\nYour Task:\nTranslate the 'title' and 'story_text' into Chinese.\nIf 'story_text' is empty, keep the original text.\n\nHTML Handling:\n1. Remove all HTML tags.\n2. If an HTML tag is a link (), convert it to Markdown format: [text](url).\n\nOutput Rules:\nReturn ONLY a valid JSON object. Do not include markdown formatting (like ```json) or any conversational text.\n\nExample:\nInput:\ntitle: \"hi\"\nstory_text: \"hello world\"\n\nOutput:\n{\n \"title\": \"嗨\",\n \"story_text\": \"你好世界\"\n}" }, "provider": "openai", "promptType": "define", "approximated": true, "systemMessage": "You are a professional news media translation API.\n\nYour Task:\nTranslate the 'title' and 'story_text' into Chinese.\nIf 'story_text' is empty, keep the original text.\n\nHTML Handling:\n1. Remove all HTML tags.\n2. If an HTML tag is a link (), convert it to Markdown format: [text](url).\n\nOutput Rules:\nReturn ONLY a valid JSON object. Do not include markdown formatting (like ```json) or any conversational text.\n\nExample:\nInput:\ntitle: \"hi\"\nstory_text: \"hello world\"\n\nOutput:\n{\n \"title\": \"嗨\",\n \"story_text\": \"你好世界\"\n}" }, "position": [ 1376, 176 ] }, { "id": "aeefe4c7-1603-4e57-b7a0-d373dcddf67d", "name": "Request Algolia", "type": "http", "params": { "url": "http://hn.algolia.com/api/v1/search_by_date", "options": {}, "sendQuery": true, "queryParameters": { "parameters": [ { "name": "tags", "value": "={{ $json.tags }}" }, { "name": "numericFilters", "value": "={{ $json.numericFilters }}" }, { "name": "hitsPerPage", "value": "={{ $json.hitsPerPage }}" } ] } }, "position": [ 784, 176 ] }, { "id": "48997584-e40b-4a2d-bcaa-f1de24a6ad6b", "name": "Merge translation results", "type": "code", "params": { "jsCode": "// Retrieve all original data from the 'Filter' node (since the AI Agent node replaces the original input data)\n// Note: Ensure the upstream node is named 'Filter'. If you renamed it, update the reference here accordingly.\nconst originalItems = $('Filter').all();\n\nreturn items.map((item, index) => {\n // 1. Get the output text from the AI Agent\n const aiOutputString = item.json.output || \"\";\n\n // 2. Sanitize the data (remove Markdown formatting like ```json ... ``` if added by the AI)\n const cleanJsonString = aiOutputString.replace(/```json/g, '').replace(/```/g, '').trim();\n\n let translatedData = {};\n\n // 3. Attempt to parse the JSON string\n try {\n translatedData = JSON.parse(cleanJsonString);\n } catch (error) {\n // If parsing fails, retain the raw text for debugging purposes\n translatedData = {\n translation_error: true,\n raw_ai_response: aiOutputString\n };\n }\n\n // 4. Merge original data with the translated data\n const originalData = originalItems[index] ? originalItems[index].json : {};\n\n return {\n json: {\n ...originalData, // Preserve original fields (e.g., url, objectID)\n ...translatedData // Add new translated fields (e.g., title_cn, story_cn)\n }\n };\n});" }, "position": [ 1680, 176 ] }, { "id": "28dba153-23d5-4c12-a786-605cb236040d", "name": "Combine message templates", "type": "code", "params": { "jsCode": "const items = $input.all();\n\n// --- Helper Functions ---\n\n// HTML escape (prevents < > & errors)\nfunction escapeHtml(text) {\n if (!text) return \"\";\n return text\n .replace(/&/g, \"&\")\n .replace(//g, \">\");\n}\n\n// Clean and truncate text\nfunction cleanAndTruncate(story_text = \"\") {\n if (!story_text) return \"\";\n \n // 1. Convert HTML tags to newlines or empty strings\n let cleaned = story_text\n .replace(/

/g, \"\\n\") // p tag to newline\n .replace(/
/g, \"\\n\") // br tag to newline\n .replace(/<[^>]+>/g, \"\") // Remove all other HTML tags\n .replace(/"/g, '\"')\n .replace(/'/g, \"'\")\n .replace(/&/g, \"&\");\n\n // 2. Truncation logic (limit to 120 characters)\n if (cleaned.length > 120) {\n cleaned = cleaned.substring(0, 120) + \"...\";\n }\n\n // 3. Escape output\n return escapeHtml(cleaned);\n}\n\n// --- Main Logic ---\n\nconst count = items.length;\nlet message = \"\";\n\n// 1. Header Information\nmessage += `🚨 Hacker News 24H Trending Alert\\n`;\nmessage += `📉 Top stories captured: ${count}\\n`; \nmessage += `\\n`; // Extra empty line to separate the header\n\nconst contents = [];\n\nfor (const item of items) {\n const data = item.json;\n let content = \"\";\n\n // A. Title (with link)\n const safeTitle = escapeHtml(data.title);\n const hnLink = `https://news.ycombinator.com/item?id=${data.objectID}`;\n content += `🔥 ${safeTitle}\\n`;\n\n // B. Data row (aligned using Code style)\n const score = parseFloat(data.analysis.velocity_score || 0).toFixed(2);\n const age = parseFloat(data.analysis.age_hours || 0).toFixed(1);\n \n content += `📈 ${score} | 💬 ${data.num_comments} | ⭐️ ${data.points}\\n`;\n content += `🕒 Posted ${age} hours ago\\n`;\n\n // C. Content preview (only shown if story_text exists and is not empty)\n if (data.story_text && data.story_text.trim().length > 0) {\n content += `

${cleanAndTruncate(data.story_text)}

`;\n }\n \n contents.push(content);\n}\n\n// --- Combine Output ---\nconst separator = \"\\n\";\n\nreturn {\n json: {\n // If no stories, do not join, just output header\n text: message + (contents.length > 0 ? contents.join(separator) : \"\")\n }\n};" }, "position": [ 1872, 176 ] } ] }

Send 24-hour Hacker News trends to Telegram with OpenRouter AI translation