const XLSX = require('xlsx');

const fs = require('fs');

const wb = XLSX.readFile('articles.xlsx');

const ws = wb.Sheets[wb.SheetNames[0]];

const data = XLSX.utils.sheet_to_json(ws);

const cols = Object.keys(data[0]);

const titleCol = cols.find(c => c.includes('\u6807\u9898'));

const contentCol = cols.find(c => c.includes('\u5185\u5bb9'));

const summaryCol = cols.find(c => c.includes('\u6458\u8981'));

const coverCol = cols.find(c => c.includes('\u5c01\u9762'));

function formatContent(raw) {

if (!raw || raw.length < 20) return raw || '';

let text = raw;

// Add breaks before section headers

text = text.replace(/([。！？～\n])(一、|二、|三、|四、|五、|六、)/g, '$1\n\n$2');

text = text.replace(/([。！？～\n])(第一步|第二步|第三步|第四步|第五步|第六步)/g, '$1\n\n$2');

// Add breaks before 纪实/思考/反思 markers

text = text.replace(/([。！？～\n])(纪实\d|思考\d|反思\d|思考：|反思：)/g, '$1\n\n$2');

text = text.replace(/([。！？～\n])(纪实 ?\d|思考 ?\d)/g, '$1\n\n$2');

// Add breaks before emoji section markers

text = text.replace(/([。！？～\n])([\u{1F300}-\u{1F9FF}\u{2600}-\u{26FF}\u{2700}-\u{27BF}✅❌✔️])/gu, '$1\n\n$2');

// Add breaks before numbered markers like "1️⃣", "2️⃣"

text = text.replace(/([。！？～\n])([1-9]️⃣)/g, '$1\n\n$2');

// Add breaks before "方式一", "方式二", "方向一" etc

text = text.replace(/([。！？～\n])(方式[一二三四]|方向[一二三四])/g, '$1\n\n$2');

// Add breaks before "死局一", "调整1" etc

text = text.replace(/([。！？～\n])(死局[一二三四五]|调整[1-9])/g, '$1\n\n$2');

// Add breaks before lines starting with "👉"

text = text.replace(/([。！？～\n])(👉)/g, '$1\n\n$2');

// Add breaks before "今日小结", "写在最后", "最后" section headers

text = text.replace(/([。！？～])(今日小结|写在最后|核心总结|Day\d+小结|end$)/g, '$1\n\n$2');

// Add breaks before "先说说", "再说说", "其他" transition phrases at start of new topics

text = text.replace(/([。！？～])(先说说|再说说|除此之外|不过|当然|但|所以从今天开始|从今天开始)/g, '$1\n\n$2');

// Break long paragraphs: if a sentence ends with 。and next starts a new idea (>100 chars since last break)

const lines = text.split('\n');

const result = [];

for (const line of lines) {

if (line.length > 200) {

// Split at sentence boundaries for very long paragraphs

let parts = line.split(/(?<=[。！？])/);

let currentPara = '';

for (const part of parts) {

currentPara += part;

if (currentPara.length > 150) {

result.push(currentPara.trim());

currentPara = '';

}

}

if (currentPara.trim()) result.push(currentPara.trim());

} else {

result.push(line);

}

}

text = result.join('\n');

// Clean up excessive newlines

text = text.replace(/\n{3,}/g, '\n\n');

return text.trim();

}

function extractDayNumber(title) {

const m = title.match(/[Dd]ay\s*(\d+)/i);

if (m) return parseInt(m[1]);

const m2 = title.match(/周报\s*[|｜]?\s*[Nn]o\.?\s*(\d+)/i);

if (m2) return 'week' + m2[1];

return null;

}

function extractShortTitle(title) {

let t = title

.replace(/^(一人公司|独舟一人公司|方鑫一人公司)(创业日志|创业周报)?[·・\s]*[Dd]ay\s*\d+\s*[|｜:：]?\s*/i, '')

.replace(/^(开发日志)[Dd]ay\s*\d+\s*[——|｜:：]+\s*/i, '')

.trim();

if (!t || t === title) {

t = title.replace(/^.*?[Dd]ay\s*\d+\s*[|｜:：·・]\s*/i, '').trim();

}

if (!t) t = title;

return t;

}

function extractSummary(content, maxLen = 80) {

if (!content) return '';

const lines = content.split('\n').filter(l => l.trim());

const textLines = lines.slice(1).filter(l => l.trim().length > 10);

if (textLines.length > 0) {

let s = textLines[0].trim();

if (s.length > maxLen) s = s.substring(0, maxLen) + '…';

return s;

}

return '';

}

function autoTag(title, content) {

const tags = [];

const text = (title || '') + ' ' + (content || '').substring(0, 500);

if (/DeskFlow/i.test(text)) tags.push('DeskFlow');

if (/AI/i.test(text) && tags.length < 3) tags.push('AI');

if (/自媒体|视频|口播|抖音|小红书/.test(text) && tags.length < 3) tags.push('自媒体');

if (/服务器|部署|上线/.test(text) && tags.length < 3) tags.push('技术');

if (/思考|感悟|心态|方向|孤独|成长/.test(text) && tags.length < 3) tags.push('思考');

if (/工具|效率|自动化/.test(text) && tags.length < 3) tags.push('工具');

if (/合规|封禁/.test(text) && tags.length < 3) tags.push('合规');

if (/学习|读书/.test(text) && tags.length < 3) tags.push('学习');

if (/开发|编程|代码/.test(text) && tags.length < 3) tags.push('开发');

if (/周报/.test(title)) tags.push('周报');

if (tags.length === 0) tags.push('日志');

return tags.slice(0, 3);

}

function esc(str) {

if (!str) return '';

return str.replace(/\\/g, '\\\\').replace(/`/g, '\\`').replace(/\$/g, '\\$');

}

function escQ(str) {

if (!str) return '';

return str.replace(/\\/g, '\\\\').replace(/'/g, "\\'");

}

// Process

const entries = [];

const seen = new Set();

data.forEach(row => {

const title = row[titleCol] || '';

const rawContent = row[contentCol] || '';

const excelSummary = row[summaryCol] || '';

const cover = row[coverCol] || '';

let dayId = extractDayNumber(title);

if (!dayId) return;

let isWeekReport = typeof dayId === 'string' && dayId.startsWith('week');

let weekNum = isWeekReport ? parseInt(dayId.replace('week', '')) : null;

let key = isWeekReport ? dayId : 'day' + dayId;

if (seen.has(key)) return;

seen.add(key);

const content = formatContent(rawContent);

const shortTitle = extractShortTitle(title);

const summary = extractSummary(content) || excelSummary || '点击查看详情…';

const tags = autoTag(title, rawContent);

entries.push({

isWeekReport, weekNum,

dayNum: isWeekReport ? null : dayId,

dayLabel: isWeekReport ? `'W${weekNum}'` : `${dayId}`,

title: shortTitle || title,

summary: summary.length > 80 ? summary.substring(0, 80) + '…' : summary,

content,

tags,

cover,

});

});

entries.sort((a, b) => {

if (a.isWeekReport && !b.isWeekReport) return -1;

if (!a.isWeekReport && b.isWeekReport) return 1;

if (a.isWeekReport && b.isWeekReport) return b.weekNum - a.weekNum;

return b.dayNum - a.dayNum;

});

// Generate entries array

let entriesTs = '';

entries.forEach(e => {

const tagsStr = e.tags.map(t => `'${t}'`).join(', ');

entriesTs += ` {\n`;

entriesTs += ` day: ${e.dayLabel},\n`;

entriesTs += ` title: '${escQ(e.title)}',\n`;

entriesTs += ` summary: '${escQ(e.summary)}',\n`;

entriesTs += ` content: \`${esc(e.content)}\`,\n`;

entriesTs += ` tags: [${tagsStr}],\n`;

if (e.cover) {

entriesTs += ` cover: '${escQ(e.cover)}',\n`;

}

entriesTs += ` },\n`;

});

fs.writeFileSync('journal_entries.txt', entriesTs, 'utf-8');

console.log(`Generated ${entries.length} formatted entries`);

console.log(`File size: ${(entriesTs.length / 1024).toFixed(1)} KB`);

// Show a sample of formatted content

const sample = entries.find(e => e.dayNum === 1);

if (sample) {

console.log('\n--- Sample (Day 1) first 500 chars ---');

console.log(sample.content.substring(0, 500));

}