fmode-voc-projects/scripts/tools/hongchengdaojia/hongcheng-xhs-to-v3-report.js

#!/usr/bin/env node
/**
 * 洪城到家 · XHS 数据融入 VOC 洞察报告 v3
 *
 * 功能：
 *   1. 读取 docs/洪城到家/raw/xhs/*.json 的XHS数据
 *   2. 读取 hongcheng-voc-insight-report-v3.html
 *   3. 替换报告中的占位符为真实XHS数据
 *
 * 用法：
 *   node scripts/tools/xhs-to-v3-report.js
 */

const fs = require('fs');
const path = require('path');

const ROOT = path.resolve(__dirname, '..', '..');
const XHS_DIR = path.join(ROOT, 'docs', '洪城到家', 'raw', 'xhs');
const REPORT_PATH = path.join(ROOT, 'reports', 'hongcheng-voc-insight-report-v3.html');

function esc(s) { return String(s ?? '').replace(/[&<>"']/g, c => ({'&':'&amp;','<':'&lt;','>':'&gt;','"':'&quot;',"'":'&#39;'}[c])); }
function fmtLikes(n) { if (n == null) return '0'; if (n >= 10000) return (n/10000).toFixed(1).replace(/\.0$/,'')+'w'; if (n >= 1000) return (n/1000).toFixed(1).replace(/\.0$/,'')+'k'; return String(n); }
function truncate(s, n=120) { s = String(s ?? ''); return s.length > n ? s.slice(0,n-1)+'…' : s; }

function loadXHSData() {
    const files = fs.readdirSync(XHS_DIR).filter(f => f.endsWith('.json'));
    const allData = { notes: [], comments: [], byBrand: {} };

    for (const file of files) {
        const brand = file.replace('.json', '');
        const raw = JSON.parse(fs.readFileSync(path.join(XHS_DIR, file), 'utf8'));
        allData.byBrand[brand] = raw;

        for (const note of (raw.top_notes || [])) {
            note.brand = brand;
            allData.notes.push(note);
            for (const comment of (note.comments || [])) {
                comment.note_id = note.note_id;
                comment.brand = brand;
                allData.comments.push(comment);
            }
        }
    }
    return allData;
}

function analyzeXHS(xhsData) {
    const { notes, comments, byBrand } = xhsData;

    const brandStats = {};
    for (const [brand, data] of Object.entries(byBrand)) {
        const notes_count = data.notes_processed || 0;
        const comments_count = data.total_comments || 0;
        const total_likes = (data.top_notes || []).reduce((s, n) => s + (n.liked_count || 0), 0);
        brandStats[brand] = { notes_count, comments_count, total_likes };
    }

    const topNotes = [...notes].sort((a, b) => (b.liked_count || 0) - (a.liked_count || 0)).slice(0, 6);
    const topComments = [...comments].sort((a, b) => (b.comment_likes || 0) - (a.comment_likes || 0)).slice(0, 4);

    const sentimentStats = { positive: 0, negative: 0, neutral: 0, conflicted: 0 };
    for (const note of notes) {
        const s = note.sentiment || 'neutral';
        sentimentStats[s] = (sentimentStats[s] || 0) + 1;
    }

    const tagStats = {};
    for (const note of notes) {
        for (const tag of (note.inferred_tags || [])) {
            tagStats[tag] = (tagStats[tag] || 0) + 1;
        }
    }
    const topTags = Object.entries(tagStats).sort((a, b) => b[1] - a[1]).slice(0, 8);

    return { brandStats, topNotes, topComments, sentimentStats, tagStats, topTags, totalNotes: notes.length, totalComments: comments.length };
}

function vocCardXHS(item, type = 'note') {
    const senColor = {positive:'#00DC82',neutral:'#6B6B6B',negative:'#EF4444',conflicted:'#F97316'}[item.sentiment]||'#6B6B6B';
    const platformIcon = type === 'note' ? '红' : '评';
    const likes = type === 'note' ? (item.liked_count || 0) : (item.comment_likes || 0);
    const content = type === 'note' ? (item.content || item.title || '') : (item.content || '');

    return `<div style="background:linear-gradient(180deg,rgba(255,255,255,0.03) 0%,rgba(255,255,255,0.01) 100%),var(--bg-1);border:1px solid var(--border);border-radius:var(--radius-md);padding:16px 18px;backdrop-filter:blur(8px);margin-bottom:12px;">
        <div style="display:flex;justify-content:space-between;align-items:center;gap:10px;margin-bottom:10px;flex-wrap:wrap;">
            <div style="display:flex;align-items:center;gap:8px;">
                <span style="width:20px;height:20px;border-radius:50%;background:#FF2442;color:#fff;font-size:0.55rem;font-weight:700;display:flex;align-items:center;justify-content:center;">${platformIcon}</span>
                <span style="font-weight:600;font-size:0.8rem;">${esc(item.author || item.user_nickname || '匿名')}</span>
                ${item.ip_location ? `<span style="font-size:0.68rem;color:var(--text-3);">· ${esc(item.ip_location)}</span>` : ''}
            </div>
            <div style="display:flex;align-items:center;gap:8px;">
                <span style="font-size:0.68rem;font-weight:600;color:${senColor};">${item.sentiment === 'positive' ? '▲' : item.sentiment === 'negative' ? '▼' : '●'} ${item.sentiment}</span>
                <span style="font-size:0.68rem;color:var(--text-3);font-family:var(--font-mono);">♥${fmtLikes(likes)}</span>
            </div>
        </div>
        <div style="font-size:0.85rem;line-height:1.6;color:var(--text-1);">"${esc(truncate(content, 150))}"</div>
        ${item.inferred_tags && item.inferred_tags.length ? `<div style="margin-top:8px;display:flex;flex-wrap:wrap;gap:4px;">${item.inferred_tags.slice(0,3).map(t=>`<span style="padding:2px 6px;background:${senColor}15;color:${senColor};border-radius:3px;font-size:0.6rem;font-weight:500;">${esc(t)}</span>`).join('')}</div>` : ''}
    </div>`;
}

function brandTableRow(brand, stats) {
    return `<tr>
        <td style="padding:10px 12px;border-bottom:1px solid var(--border);font-weight:600;font-size:0.85rem;">${esc(brand)}</td>
        <td style="padding:10px 12px;border-bottom:1px solid var(--border);text-align:right;font-variant-numeric:tabular-nums;">${stats.notes_count}</td>
        <td style="padding:10px 12px;border-bottom:1px solid var(--border);text-align:right;font-variant-numeric:tabular-nums;">${stats.comments_count}</td>
        <td style="padding:10px 12px;border-bottom:1px solid var(--border);text-align:right;font-variant-numeric:tabular-nums;">♥ ${fmtLikes(stats.total_likes)}</td>
    </tr>`;
}

function generateXHSSection(stats) {
    const { brandStats, topNotes, totalNotes, totalComments, sentimentStats, topTags } = stats;
    const selfBrand = brandStats['洪诚到家月嫂'] || { notes_count: 0, comments_count: 0, total_likes: 0 };
    const otherBrands = Object.entries(brandStats).filter(([k]) => k !== '洪诚到家月嫂');
    const totalLikes = Object.values(brandStats).reduce((s, b) => s + b.total_likes, 0);

    const sentimentHtml = Object.entries(sentimentStats).map(([s, cnt]) => {
        const colors = {positive:'#00DC82',negative:'#EF4444',neutral:'#6B6B6B',conflicted:'#F97316'};
        const labels = {positive:'正向',negative:'负向',neutral:'中立',conflicted:'矛盾'};
        const pct = totalNotes ? Math.round((cnt / totalNotes) * 100) : 0;
        return `<div style="display:flex;align-items:center;gap:10px;margin-bottom:8px;">
            <span style="width:50px;font-size:0.75rem;color:var(--text-2);">${labels[s]}</span>
            <div style="flex:1;height:8px;background:var(--bg-3);border-radius:4px;overflow:hidden;">
                <div style="width:${pct}%;height:100%;background:${colors[s]};border-radius:4px;"></div>
            </div>
            <span style="width:40px;font-size:0.72rem;color:var(--text-3);text-align:right;">${pct}%</span>
        </div>`;
    }).join('');

    const tagHtml = topTags.map(([tag, cnt]) =>
        `<span style="display:inline-flex;align-items:center;gap:6px;padding:4px 10px;background:var(--bg-2);border:1px solid var(--border);border-radius:4px;font-size:0.72rem;margin:3px;">
            <span style="color:var(--text-2);">${esc(tag)}</span>
            <span style="color:var(--amber);font-weight:600;">${cnt}</span>
        </span>`
    ).join('');

    const notesHtml = topNotes.slice(0, 4).map(n => vocCardXHS(n, 'note')).join('');

    const brandsHtml = brandTableRow('洪诚到家', selfBrand) +
        otherBrands.map(([b, s]) => brandTableRow(b, s)).join('');

    return `
    <div style="background:linear-gradient(180deg,rgba(255,36,66,0.06) 0%,rgba(255,36,66,0.02) 100%);border:1px solid rgba(255,36,66,0.2);border-radius:var(--radius-md);padding:24px;margin-bottom:24px;">
        <div style="display:flex;align-items:center;gap:12px;margin-bottom:20px;">
            <span style="width:32px;height:32px;border-radius:8px;background:#FF2442;color:#fff;font-size:0.7rem;font-weight:700;display:flex;align-items:center;justify-content:center;">红</span>
            <div>
                <div style="font-weight:700;font-size:1rem;color:var(--text-1);">小红书（XHS）真实数据</div>
                <div style="font-size:0.72rem;color:var(--text-3);">基于${totalNotes}篇笔记 · ${totalComments}条评论</div>
            </div>
        </div>

        <div style="display:grid;grid-template-columns:repeat(4,1fr);gap:12px;margin-bottom:24px;">
            <div style="background:var(--bg-1);border:1px solid var(--border);border-radius:var(--radius-sm);padding:14px;text-align:center;">
                <div style="font-family:var(--font-head);font-size:1.6rem;font-weight:800;color:#FF2442;">${totalNotes}</div>
                <div style="font-size:0.68rem;color:var(--text-3);margin-top:4px;">采集笔记</div>
            </div>
            <div style="background:var(--bg-1);border:1px solid var(--border);border-radius:var(--radius-sm);padding:14px;text-align:center;">
                <div style="font-family:var(--font-head);font-size:1.6rem;font-weight:800;color:#FF2442;">${totalComments}</div>
                <div style="font-size:0.68rem;color:var(--text-3);margin-top:4px;">采集评论</div>
            </div>
            <div style="background:var(--bg-1);border:1px solid var(--border);border-radius:var(--radius-sm);padding:14px;text-align:center;">
                <div style="font-family:var(--font-head);font-size:1.6rem;font-weight:800;color:#FF2442;">${Object.keys(brandStats).length}</div>
                <div style="font-size:0.68rem;color:var(--text-3);margin-top:4px;">品牌覆盖</div>
            </div>
            <div style="background:var(--bg-1);border:1px solid var(--border);border-radius:var(--radius-sm);padding:14px;text-align:center;">
                <div style="font-family:var(--font-head);font-size:1.6rem;font-weight:800;color:#FF2442;">♥ ${fmtLikes(totalLikes)}</div>
                <div style="font-size:0.68rem;color:var(--text-3);margin-top:4px;">总点赞数</div>
            </div>
        </div>

        <div style="display:grid;grid-template-columns:1fr 1fr;gap:20px;">
            <div>
                <div style="font-size:0.78rem;font-weight:600;color:var(--text-2);margin-bottom:12px;letter-spacing:0.05em;">品牌声量排行</div>
                <table style="width:100%;border-collapse:collapse;font-size:0.8rem;">
                    <thead>
                        <tr style="border-bottom:1px solid var(--border-strong);">
                            <th style="text-align:left;padding:8px 12px;font-weight:600;color:var(--text-3);font-size:0.68rem;">品牌</th>
                            <th style="text-align:right;padding:8px 12px;font-weight:600;color:var(--text-3);font-size:0.68rem;">笔记</th>
                            <th style="text-align:right;padding:8px 12px;font-weight:600;color:var(--text-3);font-size:0.68rem;">评论</th>
                            <th style="text-align:right;padding:8px 12px;font-weight:600;color:var(--text-3);font-size:0.68rem;">点赞</th>
                        </tr>
                    </thead>
                    <tbody>${brandsHtml}</tbody>
                </table>
            </div>

            <div>
                <div style="font-size:0.78rem;font-weight:600;color:var(--text-2);margin-bottom:12px;letter-spacing:0.05em;">情感分布</div>
                ${sentimentHtml}
                <div style="margin-top:16px;">
                    <div style="font-size:0.78rem;font-weight:600;color:var(--text-2);margin-bottom:10px;letter-spacing:0.05em;">热门标签</div>
                    ${tagHtml}
                </div>
            </div>
        </div>
    </div>

    <div style="margin-bottom:24px;">
        <div style="font-size:0.78rem;font-weight:600;color:var(--text-2);margin-bottom:14px;letter-spacing:0.05em;">高赞笔记精选</div>
        <div style="display:grid;grid-template-columns:repeat(2,1fr);gap:12px;">${notesHtml}</div>
    </div>`;
}

function main() {
    console.log('📊 加载XHS数据...');
    const xhsData = loadXHSData();
    console.log(`   加载 ${xhsData.notes.length} 篇笔记, ${xhsData.comments.length} 条评论`);

    console.log('🔧 分析XHS数据...');
    const stats = analyzeXHS(xhsData);

    console.log('📝 生成XHS数据区块...');
    const xhsSectionHtml = generateXHSSection(stats);

    console.log('📄 读取v3报告...');
    let reportHtml = fs.readFileSync(REPORT_PATH, 'utf8');

    const placeholderPattern = /<div style="background:linear-gradient\(180deg,rgba\(255,36,66,0\.06[^>]*>[\s\S]*?<div style="margin-bottom:24px;">[\s\S]*?<\/div>\n    <\/div>/;

    if (placeholderPattern.test(reportHtml)) {
        reportHtml = reportHtml.replace(placeholderPattern, xhsSectionHtml);
        console.log('✅ 替换占位符完成');
    } else {
        const insertMarker = '<!-- XHS数据填充区域 -->';
        if (reportHtml.includes(insertMarker)) {
            reportHtml = reportHtml.replace(insertMarker, insertMarker + '\n' + xhsSectionHtml);
            console.log('✅ 插入XHS数据区块完成');
        } else {
            console.log('⚠️ 未找到占位符，将在报告末尾添加XHS数据区块');
            reportHtml = reportHtml.replace('</body>', xhsSectionHtml + '\n</body>');
        }
    }

    const outPath = REPORT_PATH;
    fs.writeFileSync(outPath, reportHtml, 'utf8');
    console.log(`✅ 报告已更新: ${outPath}`);

    console.log('\n📊 XHS数据摘要:');
    console.log(`   笔记总数: ${stats.totalNotes}`);
    console.log(`   评论总数: ${stats.totalComments}`);
    console.log(`   品牌数量: ${Object.keys(stats.brandStats).length}`);
    console.log(`   总点赞: ♥ ${fmtLikes(Object.values(stats.brandStats).reduce((s, b) => s + b.total_likes, 0))}`);
    console.log(`   情感分布:`, stats.sentimentStats);
    console.log(`   热门标签:`, stats.topTags.slice(0, 5).map(([t, c]) => `${t}(${c})`).join(', '));
}

if (require.main === module) {
    try {
        main();
    } catch (e) {
        console.error('fatal:', e);
        process.exit(1);
    }
}

module.exports = { loadXHSData, analyzeXHS, generateXHSSection };