lovelyCARDINAL/WikiBots

import { Buffer } from 'buffer';
import { env } from 'process';
import { Octokit } from '@octokit/core';
import { MediaWikiApi } from 'wiki-saikou';
import config from '../utils/config.js';
import readData from '../utils/readData.js';
import splitAndJoin from '../utils/splitAndJoin.js';

const zhapi = new MediaWikiApi(config.zh.api, {
        headers: { 'user-agent': config.useragent },
    }),
    cmapi = new MediaWikiApi(config.cm.api, {
        headers: { 'user-agent': config.useragent },
    });

const octokit = new Octokit({ auth: env.GITHUB_TOKEN });

async function getDetails(title) {
    const { data: { query: { logevents } } } = await cmapi.post({
        list: 'logevents',
        leprop: 'type|details',
        ledir: 'older',
        letitle: title,
        lelimit: 'max',
    }, {
        retry: 15,
    });
    if (!logevents.length) {
        return '未曾上传';
    }
    for (const { type, params } of logevents) {
        if (type === 'delete') {
            return `被删除<span class='sysop-show'>（[https://commons.moegirl.org.cn/Special:恢复被删页面/${title.replaceAll(' ', '_')} 恢复]）</span>`;
        }
        if (type === 'move') {
            return `被移动至[[:${params.target_title}]]`;
        }
    }
    return '未知';
}

(async () => {
    console.log(`Start time: ${new Date().toISOString()}`);

    await Promise.all([
        zhapi.login(
            config.zh.ibot.name,
            config.zh.ibot.password,
            undefined,
            { retry: 25, noCache: true },
        ).then(console.log),
        cmapi.login(
            config.cm.ibot.name,
            config.cm.ibot.password,
            undefined,
            { retry: 25, noCache: true },
        ).then(console.log),
    ]);

    const lastData = JSON.parse(await readData('brokenFiles.json'));
    
    const pageData = {};
    const imgData = {};
    const eol = Symbol();
    let imcontinue = undefined;
    while (imcontinue !== eol) {
        const { data } = await zhapi.post({
            prop: 'images',
            generator: 'categorymembers',
            imlimit: 'max',
            gcmtitle: 'Category:含有受损文件链接的页面',
            gcmnamespace: '0|10|12|4',
            gcmlimit: 'max',
            gcmsort: 'timestamp',
            gcmdir: 'older',
            imcontinue,
        }, {
            retry: 15,
        });
        imcontinue = data.continue ? data.continue.imcontinue : eol;

        const pagelist = Object.values(data.query.pages).filter((page) => page.images && !/Sandbox|沙盒|页面格式/i.test(page.title));

        const imageTitles = pagelist.flatMap(({ images }) => images.map(({ title }) => title));
        const imageTitlesGroup = splitAndJoin(imageTitles, 500);
        await Promise.all(imageTitlesGroup.map(async(titles) => {
            const { data: { query: { pages } } } = await cmapi.post({
                prop: 'revisions',
                titles,
                rvprop: '',
            }, {
                retry: 15,
            });
            await Promise.all(pages.map(async ({ title, missing, known }) => {
                if (missing) {
                    imgData[title] = lastData?.[title] || (known ? `页面丢失<span class='sysop-show'>（[https://commons.moegirl.org.cn/Special:恢复被删页面/${title.replaceAll(' ', '_')} 恢复]）</span>` : await getDetails(title));
                }
            }));
        }));
        
        for (const page of pagelist) {
            const { pageid, title, ns, images } = page;
            pageData[pageid] ||= { title, ns, images: {} };
            const imagelist = images.map(({ title }) => title);
            for (const title of imagelist) {
                if (imgData?.[title]) {
                    pageData[pageid].images[title] = imgData[title];
                }
            }
        }
    }

    try {
        const { data: { sha } } = await octokit.request('GET /repos/{owner}/{repo}/contents/{path}', {
            owner: 'lovelyCARDINAL',
            repo: 'WikiBots',
            path: 'data/brokenFiles.json',
        });
        await octokit.request('PUT /repos/{owner}/{repo}/contents/{path}', {
            owner: 'lovelyCARDINAL',
            repo: 'WikiBots',
            path: 'data/brokenFiles.json',
            message: 'auto: update broken files data',
            content: Buffer.from(JSON.stringify(imgData, null, '\t'), 'utf-8').toString('base64'),
            sha,
        });
        console.log('SUCCESS!');
    } catch (error) {
        console.error('ERROR:', error.message);
    }

    let text = '* 本页面为[[U:星海-interfacebot|机器人]]生成的[[:Category:含有受损文件链接的页面|受损文件]]详细信息，完成修复的<b>任何用户</b>都可以<b class="plainlinks">[{{fullurl:{{FULLPAGENAME}}|action=edit}} 编辑下方表格]</b>。\n* 生成时间：{{subst:#time:Y年n月j日 (D) H:i (T)}}｜{{subst:#time:Y年n月j日 (D) H:i (T)|||1}} \n\n{| class="wikitable sortable plainlinks" style="word-break:break-all" width=100%\n|-\n! 页面名 || 名字空间 || 文件名 || 文件状态 \n|-\n';
    for (const { title, ns, images } of Object.values(pageData)) {
        const namespace = `data-sort-value="${ns}"|${ns === 0 ? '（主）' : `{{ns:${ns}}}`}`;
        const rowspan = Object.keys(images).length;
        if (!rowspan) {
            continue;
        }
        const results = Object.keys(images).map((title) => `|[[cm:${title}|${title}]]||${images[title]}\n|-`).join('\n');
        text += rowspan === 1
            ? `|[[${title}]]\n|${namespace}\n${results}\n`
            : `|rowspan=${rowspan}|[[${title}]]\n|rowspan=${rowspan} ${namespace}\n${results}\n`;
    }
    text += '|}\n[[Category:萌娘百科数据报告]][[Category:积压工作]]';

    await zhapi.postWithToken('csrf', {
        action: 'edit',
        pageid: '555599',
        text,
        summary: '更新数据报告',
        bot: true,
        notminor: true,
        tags: 'Bot',
        watchlist: 'nochange',
    }, {
        retry: 50,
        noCache: true,
    }).then(({ data }) => console.log(JSON.stringify(data)));

    console.log(`End time: ${new Date().toISOString()}`);
})();