{ "name": "My workflow copy", "nodes": [ { "parameters": { "rule": { "interval": [ {} ] } }, "type": "n8n-nodes-base.scheduleTrigger", "typeVersion": 1.2, "position": [ -2592, -800 ], "id": "fc73f795-40f9-4fa0-8352-9a5db4af1072", "name": "Schedule Trigger" }, { "parameters": { "conditions": { "options": { "caseSensitive": true, "leftValue": "", "typeValidation": "strict", "version": 2 }, "conditions": [ { "id": "8b6187de-5add-4b59-a035-e1a751d5f89b", "leftValue": "={{ $('提取微博内容').item.json.postTime }}", "rightValue": "={{ $json.lastDate }}", "operator": { "type": "dateTime", "operation": "after" } } ], "combinator": "and" }, "options": {} }, "type": "n8n-nodes-base.if", "typeVersion": 2.2, "position": [ -272, -496 ], "id": "0933d7fe-a862-4dab-a583-68974eef36c9", "name": "If" }, { "parameters": { "options": {} }, "type": "n8n-nodes-base.dateTime", "typeVersion": 2, "position": [ -240, -992 ], "id": "1dbe4604-7563-41a4-85af-389b8a38ace6", "name": "CurrentDate" }, { "parameters": { "promptType": "define", "text": "=你是一位顶尖的数字舆情分析师，拥有社会学、传播学、消费者行为学和自然语言处理的复合背景。现在，请基于我提供的以下两项信息，执行一次深度、客观、结构化的舆情分析：\n\n### 工作流程\n\n1、先分析一下原始用户发帖内容：\"{{ $('微博内容拆分').item.json.content }}\"讨论的主体是不是目标产品名称：\"腾讯地图\"，如果是则继续执行下一步，如果不是，则本条情感分析结果直接标记为neutral\n2、判定整体情感极性（正面 / 中性 / 负面）、给出置信度（0–100%）、指出关键情感触发词或语句\n\n\n⚠️ 注意事项：\n\n不臆测未在文本中体现的信息\n避免主观情绪化语言，保持专业中立\n若发帖内容模糊或信息不足，请明确说明“信息不足，无法准确判断”，并指出缺失的关键要素\n\n请严格按照以下 JSON Schema 输出结果，不得包含任何额外文本、解释、Markdown 或注释。\n\n \"sentiment\": {\n \"polarity\": \"positive|neutral|negative\",\n \"confidence\": 0.0,\n \"trigger_phrases\": [\"string\"],\n \"summary\":[\"string\"]\n }", "options": {} }, "type": "@n8n/n8n-nodes-langchain.agent", "typeVersion": 3, "position": [ 224, -336 ], "id": "a56fce21-3cc0-4a5a-9a47-2d5d040ef666", "name": "AI Agent1" }, { "parameters": { "mode": "runOnceForEachItem", "language": "python", "pythonCode": "# Loop over input items and add a new field called 'myNewField' to the JSON of each one\nimport json;\n\nreturn json.loads(_input.item.json.output)\n" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ 672, -288 ], "id": "da8b0a05-66e8-45a3-bdb7-4f7fb256c3a7", "name": "Code in Python (Beta)" }, { "parameters": { "operation": "update", "table": { "__rl": true, "value": "user_sentiment_posts", "mode": "list", "cachedResultName": "user_sentiment_posts" }, "dataMode": "defineBelow", "columnToMatchOn": "post_id", "valueToMatchOn": "={{ $('微博内容拆分').item.json.weiboId }}", "valuesToSend": { "values": [ { "column": "sentiment_label", "value": "={{ $json.sentiment.polarity }}" }, { "column": "keywords", "value": "=\"{{ $json.sentiment.trigger_phrases }}\"" }, { "column": "summary", "value": "={{ $json.sentiment.summary }}" }, { "column": "confidence", "value": "={{ $json.sentiment.confidence }}" } ] }, "options": {} }, "type": "n8n-nodes-base.mySql", "typeVersion": 2.5, "position": [ 992, -80 ], "id": "5324dbcf-fa60-454e-8a05-b0b06a3abf69", "name": "Update rows in a table1", "credentials": { "mySql": { "id": "nyt5Hl6LlXn7qrOi", "name": "MySQL account" } } }, { "parameters": { "operation": "upsert", "table": { "__rl": true, "value": "user_sentiment_posts", "mode": "list", "cachedResultName": "user_sentiment_posts" }, "dataMode": "defineBelow", "columnToMatchOn": "post_id", "valueToMatchOn": "={{ $('提取微博内容').item.json.weiboId }}", "valuesToSend": { "values": [ { "column": "username", "value": "={{ $('提取微博内容').item.json.username }}" }, { "column": "platform", "value": "weibo" }, { "column": "post_url", "value": "={{ $('提取微博内容').item.json.postTimeLink }}" }, { "column": "content", "value": "={{ $('提取微博内容').item.json.content }}" }, { "column": "post_time", "value": "={{ $('提取微博内容').item.json.postTime }}" } ] }, "options": {} }, "type": "n8n-nodes-base.mySql", "typeVersion": 2.5, "position": [ 512, -672 ], "id": "5c672121-a812-4d01-990f-87aa9d72b967", "name": "Insert or update rows in a table1", "credentials": { "mySql": { "id": "nyt5Hl6LlXn7qrOi", "name": "MySQL account" } } }, { "parameters": { "model": "deepseek-r1:7b", "options": { "format": "json" } }, "type": "@n8n/n8n-nodes-langchain.lmChatOllama", "typeVersion": 1, "position": [ 192, -160 ], "id": "6cd6af68-3628-4e56-a3a3-c64620422a6c", "name": "Ollama Chat Model", "credentials": { "ollamaApi": { "id": "9nWs7XUJz2fyGXDw", "name": "Ollama account" } } }, { "parameters": { "jsCode": "return [\n { json: { page: 1 } },\n { json: { page: 2 } },\n { json: { page: 3 } },\n { json: { page: 4 } },\n { json: { page: 5 } }\n];" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ -2144, -752 ], "id": "f26b5c1e-4e51-427f-8a4f-370477279259", "name": "Code in JavaScript" }, { "parameters": { "fieldToSplitOut": "content", "include": "selectedOtherFields", "fieldsToInclude": "weiboId", "options": {} }, "type": "n8n-nodes-base.splitOut", "typeVersion": 1, "position": [ -704, -896 ], "id": "dbb9aea4-1b31-4b09-bf41-162f5eb4e1ad", "name": "微博内容拆分" }, { "parameters": { "operation": "scrape", "url": "", "additionalFields": { "customProperties": "={\n \"url\": \"https://s.weibo.com/realtime?q=腾讯地图&rd=realtime&tw=realtime&Refer=weibo_realtime&page= {{ $json.page }}\",\n \"formats\": [\n {\n \"type\": \"markdown\"\n }\n ],\n \"onlyMainContent\": true,\n \n \"maxAge\": 172800000,\n \"headers\": {\n \"Cookie\": \"XSRF-TOKEN=Vpllcum96mWkoZUnTDUPd3Lq; _s_tentry=passport.weibo.com; Apache=8041587221385.712.1763972929386; SINAGLOBAL=8041587221385.712.1763972929386; ULV=1763972929387:1:1:1:8041587221385.712.1763972929386:; wb_view_log_3093094175=1512*9822; webim_unReadCount=%7B%22time%22%3A1764559323255%2C%22dm_pub_total%22%3A0%2C%22chat_group_client%22%3A0%2C%22chat_group_notice%22%3A0%2C%22allcountNum%22%3A8%2C%22msgbox%22%3A0%7D; ALF=1767174963; SUB=_2A25EKRZhDeRhGeVO4lER-SrNzDmIHXVnRxeprDV8PUJbkNAbLVbikW1NTXwxtwuVrr8SU2IXt6B0GARRr09VdsZ0; SUBP=0033WrSXqPxfM725Ws9jqgMF55529P9D9WF_Da88lxSPdGiWKJiIfHaO5JpX5KMhUgL.Foe71Ke71KBpS0-2dJLoI7pzdcpjIPShIGHydJMt; WBPSESS=8CcMmWu8i823I0m0tutp2OQ7eG_tRsF3Zv_q6LNsZD8mircqph1Jx7VmyZDFN9Nia7NBhC3L7v-iGnMsbRlKFmVetQWxu_QYMSKrit3hiFpoeC5ZsD7B5RtfiX41vFfNBlr7zP5Xg2R69xGLjuwA3A==\",\n \"User-Agent\": \"Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/142.0.0.0 Safari/537.36\"\n },\n \"waitFor\": 0,\n \"mobile\": false,\n \"skipTlsVerification\": true,\n \"timeout\": 100000,\n \"parsers\": [\n \"pdf\"\n ],\n \"location\": {\n \"country\": \"US\",\n \"languages\": [\n \"en-US\"\n ]\n },\n \"removeBase64Images\": false,\n \"blockAds\": true,\n \"proxy\": \"auto\",\n \"storeInCache\": true,\n \"zeroDataRetention\": false\n}" }, "useCustomBody": true, "requestOptions": {} }, "type": "@mendable/n8n-nodes-firecrawl.firecrawl", "typeVersion": 1, "position": [ -1696, -784 ], "id": "b012bcdc-d1e1-41d7-beb6-c17a515e1c74", "name": "爬取网页信息", "credentials": { "firecrawlApi": { "id": "Gx82yvk4YGHTvQee", "name": "Firecrawl account" } } }, { "parameters": { "jsCode": "// Loop over input items and add a new field called 'myNewField' to the JSON of each one\n// 基准当前时间\n const offsetHours = 8; \n const now = new Date();\n const utc = now.getTime() + (now.getTimezoneOffset() * 60000);\n const NOW = new Date(utc + (offsetHours * 3600000));\n\n function parseWeiboTime(raw) {\n raw = raw.trim();\n\n // 1. 已是标准格式：2023-10-23 21:48\n if (/^\\d{4}-\\d{2}-\\d{2}\\s+\\d{1,2}:\\d{2}$/.test(raw)) {\n return raw + ':00';\n }\n\n // 2. X秒前\n const secMatch = raw.match(/^(\\d+)秒前$/);\n if (secMatch) {\n const secs = parseInt(secMatch[1], 10);\n const dt = new Date(NOW.getTime() - secs * 1000);\n return dt.toISOString().replace('T', ' ').substring(0, 19);\n }\n\n // 3. X分钟前 \n const minMatch = raw.match(/^(\\d+)分钟前$/);\n if (minMatch) {\n const mins = parseInt(minMatch[1], 10);\n const dt = new Date(NOW.getTime() - mins * 60 * 1000);\n return dt.toISOString().replace('T', ' ').substring(0, 19);\n }\n\n // 4. X小时前\n const hourMatch = raw.match(/^(\\d+)小时前$/);\n if (hourMatch) {\n const hours = parseInt(hourMatch[1], 10);\n const dt = new Date(NOW.getTime() - hours * 60 * 60 * 1000);\n return dt.toISOString().replace('T', ' ').substring(0, 19);\n }\n\n // 5. 今天HH:mm\n const todayMatch = raw.match(/^今天(\\d{1,2}):(\\d{2})$/);\n if (todayMatch) {\n const h = parseInt(todayMatch[1], 10);\n const m = parseInt(todayMatch[2], 10);\n const dt = new Date(NOW);\n dt.setHours(h + 8, m, 0, 0);\n return dt.toISOString().replace('T', ' ').substring(0, 19);\n }\n\n // 6. MM月DD日 HH:mm\n const monthDayMatch = raw.match(/^(\\d{1,2})月(\\d{1,2})日\\s+(\\d{1,2}):(\\d{2})$/);\n if (monthDayMatch) {\n const month = parseInt(monthDayMatch[1], 10) - 1; // JS 月份从 0 开始\n const day = parseInt(monthDayMatch[2], 10);\n const h = parseInt(monthDayMatch[3], 10) + 8;\n const m = parseInt(monthDayMatch[4], 10);\n\n let year = NOW.getFullYear();\n const candidate = new Date(year, month, day, h, m, 0);\n\n // 如果解析出的日期比 NOW 还晚（比如 12月2日 在 12月1日看就是未来），则减一年\n if (candidate > NOW) {\n year -= 1;\n }\n\n const dt = new Date(year, month, day, h, m, 0);\n const pad = n => n.toString().padStart(2, '0');\n return `${dt.getFullYear()}-${pad(dt.getMonth() + 1)}-${pad(dt.getDate())} ${pad(dt.getHours())}:${pad(dt.getMinutes())}:00`;\n }\n\n // 5. 无法识别，返回原始 + 补00秒（保守处理）\n if (/^\\d{4}-\\d{2}-\\d{2}\\s+\\d{1,2}:\\d{2}:\\d{2}$/.test(raw)) {\n return raw;\n }\n // 兜底：尝试直接补 :00\n if (raw.includes(':') && !raw.includes(':00')) {\n return raw + ':00';\n }\n return raw; // 保留原样\n }\nconst posts = [];\n\nfor (const item of $input.all()) {\n // === 主提取逻辑 ===\n const blocks = item.json.data.markdown.split(/\\n-\\s*转发\\s*\\n/);\n \n const inlineImageRegex = /\\[\\]\\((https?:\\/\\/[^\\s\\)]+\\.(?:jpg|jpeg|png|gif))\\)/gi;\n\n for (const block of blocks) {\n const userPattern = /\\[!\\[.*?\\]\\((https?:\\/\\/[^\\s\\)]+)\\)\\]\\((https?:\\/\\/[^\\s\\)]+)\\)\\s*\\n\\s*\\[([^\\]]+)\\]\\((https?:\\/\\/[^\\s\\)]+)\\)(?:\\s*\\n\\s*\\[!\\[.*?\\]\\(.*?\\)\\])?\\s*\\n\\s*\\[([^\\]]+)\\]\\((https?:\\/\\/[^\\s\\)]+)\\)/;\n const userMatch = block.match(userPattern);\n if (!userMatch) continue;\n\n const username = userMatch[3];\n const userLink = userMatch[4];\n const rawPostTime = userMatch[5];\n const postTimeLink = userMatch[6];\n\n // ✅ 标准化时间\n const postTime = parseWeiboTime(rawPostTime);\n\n let rawContent = block.slice(userMatch.index + userMatch[0].length).trim();\n const cutOffIndex = rawContent.search(/\\n-\\s*(\\[\\]\\(https?:\\/\\/.*?\\.(?:jpg|png|gif)\\)|\\d+|转发|评论|赞)/i);\n if (cutOffIndex !== -1) {\n rawContent = rawContent.slice(0, cutOffIndex).trim();\n }\n\n let content = rawContent\n .replace(/\\[展开 _c_\\]\\([^)]*\\)/g, '')\n .replace(/收起 _d_/g, '')\n .replace(/\\s+/g, ' ')\n .trim();\n\n content = content.replace(/^来自\\s*\\[[^\\]]+\\]\\([^)]*\\)\\s*/, '').trim();\n\n // 提取并移除内联图片\n const inlineImages = [];\n let imgMatch;\n while ((imgMatch = inlineImageRegex.exec(content)) !== null) {\n inlineImages.push(imgMatch[1]);\n }\n content = content.replace(inlineImageRegex, '').replace(/\\s+/g, ' ').trim();\n\n // 提取列表图片\n const listImages = [];\n const listImageRegex = /-\\s*\\[\\]\\((https?:\\/\\/[^\\s\\)]+\\.(?:jpg|jpeg|png|gif))\\)/gi;\n let listMatch;\n while ((listMatch = listImageRegex.exec(block)) !== null) {\n listImages.push(listMatch[1]);\n }\n\n const imageUrls = [...new Set([...inlineImages, ...listImages])];\n\n const hashtags = [...new Set(\n content.match(/#([^#\\s]+)#?/g)?.map(tag => tag.replace(/^#|#$| $/g, '')) || []\n )];\n\n // ✅ 新增：从微博链接提取唯一ID\n function extractWeiboId(url) {\n try {\n const cleanUrl = url.split('?')[0]; // 移除查询参数\n const parts = cleanUrl.split('/');\n return parts[parts.length - 1] || null;\n } catch (e) {\n return null;\n }\n }\n\n const weiboId = extractWeiboId(postTimeLink); // ✅ 提取唯一ID\n\n posts.push({\n weiboId,\n username,\n userLink,\n postTime, // ← 已标准化为 YYYY-MM-DD HH:mm:ss\n postTimeLink,\n content,\n imageUrls,\n hashtags\n });\n }\n\n}\n\nreturn posts;" }, "type": "n8n-nodes-base.code", "typeVersion": 2, "position": [ -1216, -832 ], "id": "ba68cac4-9a9f-4941-95b7-64f56e4fd07c", "name": "提取微博内容" }, { "parameters": { "operation": "subtractFromDate", "magnitude": "={{ $json.currentDate }}", "duration": 1, "outputFieldName": "lastDate", "options": {} }, "type": "n8n-nodes-base.dateTime", "typeVersion": 2, "position": [ -32, -992 ], "id": "74c29dfd-9b1a-4b9d-b8f5-c3a60f68df7d", "name": "LastDate" } ], "pinData": {}, "connections": { "Schedule Trigger": { "main": [ [ { "node": "Code in JavaScript", "type": "main", "index": 0 } ] ] }, "If": { "main": [ [ { "node": "AI Agent1", "type": "main", "index": 0 }, { "node": "Insert or update rows in a table1", "type": "main", "index": 0 } ], [] ] }, "CurrentDate": { "main": [ [ { "node": "LastDate", "type": "main", "index": 0 } ] ] }, "AI Agent1": { "main": [ [ { "node": "Code in Python (Beta)", "type": "main", "index": 0 } ] ] }, "Code in Python (Beta)": { "main": [ [ { "node": "Update rows in a table1", "type": "main", "index": 0 } ] ] }, "Update rows in a table1": { "main": [ [] ] }, "Insert or update rows in a table1": { "main": [ [] ] }, "Ollama Chat Model": { "ai_languageModel": [ [ { "node": "AI Agent1", "type": "ai_languageModel", "index": 0 } ] ] }, "Code in JavaScript": { "main": [ [ { "node": "爬取网页信息", "type": "main", "index": 0 } ] ] }, "微博内容拆分": { "main": [ [ { "node": "CurrentDate", "type": "main", "index": 0 } ] ] }, "爬取网页信息": { "main": [ [ { "node": "提取微博内容", "type": "main", "index": 0 } ] ] }, "提取微博内容": { "main": [ [ { "node": "微博内容拆分", "type": "main", "index": 0 } ] ] }, "LastDate": { "main": [ [ { "node": "If", "type": "main", "index": 0 } ] ] } }, "active": false, "settings": { "executionOrder": "v1", "timezone": "Asia/Shanghai", "callerPolicy": "workflowsFromSameOwner", "availableInMCP": false }, "versionId": "5513217d-a0d9-4116-8aa2-45813bf3aef9", "meta": { "templateCredsSetupCompleted": true, "instanceId": "e2f1810a3d343b4cb91647ff32c3720de7e17c264aa766bc8c994b1cbb80b056" }, "id": "Otf1gnbMitsLtmK2", "tags": [] }