@jackwener/opencli 1.6.2 → 1.6.3

package/src/clis/v2ex/latest.yaml

@@ -1,28 +0,0 @@
-site: v2ex
-name: latest
-description: V2EX 最新话题
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  limit:
-    type: int
-    default: 20
-    description: Number of topics
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/topics/latest.json
-
-  - map:
-      id: ${{ item.id }}
-      rank: ${{ index + 1 }}
-      title: ${{ item.title }}
-      node: ${{ item.node.title }}
-      replies: ${{ item.replies }}
-      url: ${{ item.url }}
-
-  - limit: ${{ args.limit }}
-
-columns: [id, rank, title, node, replies, url]

package/src/clis/v2ex/me.ts

@@ -1,104 +0,0 @@
-/**
- * V2EX Me (Profile/Balance) adapter.
- */
-import { CommandExecutionError } from '../../errors.js';
-import { cli, Strategy } from '../../registry.js';
-import type { IPage } from '../../types.js';
-
-cli({
-  site: 'v2ex',
-  name: 'me',
-  description: 'V2EX 获取个人资料 (余额/未读提醒)',
-  domain: 'www.v2ex.com',
-  strategy: Strategy.COOKIE,
-  browser: true,
-
-  args: [],
-  columns: ['username', 'balance', 'unread_notifications', 'daily_reward_ready'],
-  func: async (page: IPage | null) => {
-    if (!page) throw new CommandExecutionError('Browser page required');
-
-    if (process.env.OPENCLI_VERBOSE) {
-      console.error('[opencli:v2ex] Navigating to /');
-    }
-    await page.goto('https://www.v2ex.com/');
-    
-    // Cloudflare challenge bypass wait
-    for (let i = 0; i < 5; i++) {
-      await new Promise(r => setTimeout(r, 1500));
-      const title = await page.evaluate(`() => document.title`);
-      if (!title?.includes('Just a moment')) break;
-      if (process.env.OPENCLI_VERBOSE) console.error('[opencli:v2ex] Waiting for Cloudflare...');
-    }
-    
-    // Evaluate DOM to extract user profile
-    const data = await page.evaluate(`
-      async () => {
-        let username = 'Unknown';
-        const navLinks = Array.from(document.querySelectorAll('a.top')).map(a => a.textContent?.trim());
-        if (navLinks.length > 1 && navLinks[0] === '首页') {
-          username = navLinks[1] || 'Unknown';
-        }
-        
-        if (username === 'Unknown') {
-          // Fallback check just in case
-          const profileEl = document.querySelector('a[href^="/member/"]');
-          if (profileEl && profileEl.textContent && profileEl.textContent.trim().length > 0) {
-            username = profileEl.textContent.trim();
-          }
-        }
-
-        let balance = '0';
-        const balanceLink = document.querySelector('a.balance_area');
-        if (balanceLink) {
-          balance = Array.from(balanceLink.childNodes)
-                         .filter(n => n.nodeType === 3)
-                         .map(n => n.textContent?.trim())
-                         .join(' ')
-                         .trim();
-        }
-
-        let unread_notifications = '0';
-        const notesEl = document.querySelector('a[href="/notifications"]');
-        if (notesEl) {
-          const text = notesEl.textContent?.trim() || '';
-          const match = text.match(/(\\d+)\\s*未读提醒/);
-          if (match) {
-            unread_notifications = match[1];
-          }
-        }
-
-        let daily_reward_ready = false;
-        const dailyEl = document.querySelector('a[href^="/mission/daily"]');
-        if (dailyEl && dailyEl.textContent?.includes('领取今日的登录奖励')) {
-          daily_reward_ready = true;
-        }
-
-        if (username === 'Unknown') {
-          return { 
-            error: '请先登录 V2EX（可能是 Cookie 未配置或已失效）',
-            debug_title: document.title,
-            debug_body: document.body.innerText.substring(0, 200).replace(/\\n/g, ' ')
-          };
-        }
-
-        return {
-          username,
-          balance,
-          unread_notifications,
-          daily_reward_ready: daily_reward_ready ? '是' : '否'
-        };
-      }
-    `);
-
-    if (data.error) {
-      if (process.env.OPENCLI_VERBOSE) {
-        console.error(`[opencli:v2ex:debug] Page Title: ${data.debug_title}`);
-        console.error(`[opencli:v2ex:debug] Page Body: ${data.debug_body}`);
-      }
-      throw new CommandExecutionError(data.error);
-    }
-
-    return [data];
-  },
-});

package/src/clis/v2ex/member.yaml

@@ -1,29 +0,0 @@
-site: v2ex
-name: member
-description: V2EX 用户资料
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  username:
-    positional: true
-    type: str
-    required: true
-    description: Username
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/members/show.json
-      params:
-        username: ${{ args.username }}
-
-  - map:
-      username: ${{ item.username }}
-      tagline: ${{ item.tagline }}
-      website: ${{ item.website }}
-      github: ${{ item.github }}
-      twitter: ${{ item.twitter }}
-      location: ${{ item.location }}
-
-columns: [username, tagline, website, github, twitter, location]

package/src/clis/v2ex/node.yaml

@@ -1,34 +0,0 @@
-site: v2ex
-name: node
-description: V2EX 节点话题列表
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  name:
-    positional: true
-    type: str
-    required: true
-    description: Node name (e.g. python, javascript, apple)
-  limit:
-    type: int
-    default: 10
-    description: Number of topics (API returns max 20)
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/topics/show.json
-      params:
-        node_name: ${{ args.name }}
-
-  - map:
-      rank: ${{ index + 1 }}
-      title: ${{ item.title }}
-      author: ${{ item.member.username }}
-      replies: ${{ item.replies }}
-      url: ${{ item.url }}
-
-  - limit: ${{ args.limit }}
-
-columns: [rank, title, author, replies, url]

package/src/clis/v2ex/nodes.yaml

@@ -1,31 +0,0 @@
-site: v2ex
-name: nodes
-description: V2EX 所有节点列表
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  limit:
-    type: int
-    default: 30
-    description: Number of nodes
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/nodes/all.json
-
-  - sort:
-      by: topics
-      order: desc
-
-  - map:
-      rank: ${{ index + 1 }}
-      name: ${{ item.name }}
-      title: ${{ item.title }}
-      topics: ${{ item.topics }}
-      stars: ${{ item.stars }}
-
-  - limit: ${{ args.limit }}
-
-columns: [rank, name, title, topics, stars]

package/src/clis/v2ex/notifications.ts

@@ -1,76 +0,0 @@
-/**
- * V2EX Notifications adapter.
- */
-import { CommandExecutionError } from '../../errors.js';
-import { cli, Strategy } from '../../registry.js';
-import type { IPage } from '../../types.js';
-
-cli({
-  site: 'v2ex',
-  name: 'notifications',
-  description: 'V2EX 获取提醒 (回复/由于)',
-  domain: 'www.v2ex.com',
-  strategy: Strategy.COOKIE,
-  browser: true,
-
-  args: [
-    { name: 'limit', type: 'int', default: 20, help: 'Number of notifications' }
-  ],
-  columns: ['type', 'content', 'time'],
-  func: async (page: IPage | null, kwargs) => {
-    if (!page) throw new CommandExecutionError('Browser page required');
-
-    if (process.env.OPENCLI_VERBOSE) {
-      console.error('[opencli:v2ex] Navigating to /notifications');
-    }
-    await page.goto('https://www.v2ex.com/notifications');
-    await new Promise(r => setTimeout(r, 1500)); // waitForLoadState doesn't always work robustly
-    
-    // Evaluate DOM to extract notifications
-    const data = await page.evaluate(`
-      async () => {
-        const items = Array.from(document.querySelectorAll('#Main .box .cell[id^="n_"]'));
-        return items.map(item => {
-          let type = '通知';
-          let time = '';
-          
-          // determine type based on text content
-          const text = item.textContent || '';
-          if (text.includes('回复了你')) type = '回复';
-          else if (text.includes('感谢了你')) type = '感谢';
-          else if (text.includes('收藏了你')) type = '收藏';
-          else if (text.includes('提及你')) type = '提及';
-
-          const timeEl = item.querySelector('.snow');
-          if (timeEl) {
-            time = timeEl.textContent?.trim() || '';
-          }
-
-          // payload contains the actual reply text if any
-          let payload = '';
-          const payloadEl = item.querySelector('.payload');
-          if (payloadEl) {
-            payload = payloadEl.textContent?.trim() || '';
-          }
-
-          // fallback to full text cleaning if no payload (e.g. for favorites/thanks)
-          let content = payload;
-          if (!content) {
-            content = text.replace(/\\s+/g, ' ').trim();
-            // strip out time from content if present
-            if (time && content.includes(time)) {
-              content = content.replace(time, '').trim();
-            }
-          }
-
-          return { type, content, time };
-        });
-      }
-    `);
-
-    if (!Array.isArray(data)) throw new CommandExecutionError('Failed to parse notifications data');
-
-    const limit = kwargs.limit || 20;
-    return data.slice(0, limit);
-  },
-});

package/src/clis/v2ex/replies.yaml

@@ -1,32 +0,0 @@
-site: v2ex
-name: replies
-description: V2EX 主题回复列表
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  id:
-    positional: true
-    type: str
-    required: true
-    description: Topic ID
-  limit:
-    type: int
-    default: 20
-    description: Number of replies
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/replies/show.json
-      params:
-        topic_id: ${{ args.id }}
-
-  - map:
-      floor: ${{ index + 1 }}
-      author: ${{ item.member.username }}
-      content: ${{ item.content }}
-
-  - limit: ${{ args.limit }}
-
-columns: [floor, author, content]

package/src/clis/v2ex/topic.yaml

@@ -1,33 +0,0 @@
-site: v2ex
-name: topic
-description: V2EX 主题详情和回复
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  id:
-    positional: true
-    type: str
-    required: true
-    description: Topic ID
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/topics/show.json
-      params:
-        id: ${{ args.id }}
-
-  - map:
-      id: ${{ item.id }}
-      title: ${{ item.title }}
-      content: ${{ item.content }}
-      member: ${{ item.member.username }}
-      created: ${{ item.created }}
-      node: ${{ item.node.title }}
-      replies: ${{ item.replies }}
-      url: ${{ item.url }}
-
-  - limit: 1
-
-columns: [id, title, content, member, created, node, replies, url]

package/src/clis/v2ex/user.yaml

@@ -1,34 +0,0 @@
-site: v2ex
-name: user
-description: V2EX 用户发帖列表
-domain: www.v2ex.com
-strategy: public
-browser: false
-
-args:
-  username:
-    positional: true
-    type: str
-    required: true
-    description: Username
-  limit:
-    type: int
-    default: 10
-    description: Number of topics (API returns max 20)
-
-pipeline:
-  - fetch:
-      url: https://www.v2ex.com/api/topics/show.json
-      params:
-        username: ${{ args.username }}
-
-  - map:
-      rank: ${{ index + 1 }}
-      title: ${{ item.title }}
-      node: ${{ item.node.title }}
-      replies: ${{ item.replies }}
-      url: ${{ item.url }}
-
-  - limit: ${{ args.limit }}
-
-columns: [rank, title, node, replies, url]

package/src/clis/web/read.ts

@@ -1,210 +0,0 @@
-/**
- * Generic web page reader — fetch any URL and export as Markdown.
- *
- * Uses browser-side DOM heuristics to extract the main content:
- *   1. <article> element
- *   2. [role="main"] element
- *   3. <main> element
- *   4. Largest text-dense block as fallback
- *
- * Pipes through the shared article-download pipeline (Turndown + image download).
- *
- * Usage:
- *   opencli web read --url "https://www.anthropic.com/research/..." --output ./articles
- *   opencli web read --url "https://..." --download-images false
- */
-
-import { cli, Strategy } from '../../registry.js';
-import { downloadArticle } from '../../download/article-download.js';
-
-cli({
-  site: 'web',
-  name: 'read',
-  description: 'Fetch any web page and export as Markdown',
-  strategy: Strategy.COOKIE,
-  navigateBefore: false, // we handle navigation ourselves
-  args: [
-    { name: 'url', required: true, help: 'Any web page URL' },
-    { name: 'output', default: './web-articles', help: 'Output directory' },
-    { name: 'download-images', type: 'boolean', default: true, help: 'Download images locally' },
-    { name: 'wait', type: 'int', default: 3, help: 'Seconds to wait after page load' },
-  ],
-  columns: ['title', 'author', 'publish_time', 'status', 'size'],
-  func: async (page, kwargs) => {
-    const url = kwargs.url;
-    const waitSeconds = kwargs.wait ?? 3;
-
-    // Navigate to the target URL
-    await page.goto(url);
-    await page.wait(waitSeconds);
-
-    // Extract article content using browser-side heuristics
-    const data = await page.evaluate(`
-      (() => {
-        const result = {
-          title: '',
-          author: '',
-          publishTime: '',
-          contentHtml: '',
-          imageUrls: []
-        };
-
-        // --- Title extraction ---
-        // Priority: og:title > <title> > first <h1>
-        const ogTitle = document.querySelector('meta[property="og:title"]');
-        if (ogTitle) {
-          result.title = ogTitle.getAttribute('content')?.trim() || '';
-        }
-        if (!result.title) {
-          result.title = document.title?.trim() || '';
-        }
-        if (!result.title) {
-          const h1 = document.querySelector('h1');
-          result.title = h1?.textContent?.trim() || 'untitled';
-        }
-        // Strip site suffix (e.g. " | Anthropic", " - Blog")
-        result.title = result.title.replace(/\\s*[|\\-–—]\\s*[^|\\-–—]{1,30}$/, '').trim();
-
-        // --- Author extraction ---
-        const authorMeta = document.querySelector(
-          'meta[name="author"], meta[property="article:author"], meta[name="twitter:creator"]'
-        );
-        result.author = authorMeta?.getAttribute('content')?.trim() || '';
-
-        // --- Publish time extraction ---
-        const timeMeta = document.querySelector(
-          'meta[property="article:published_time"], meta[name="date"], meta[name="publishdate"], time[datetime]'
-        );
-        if (timeMeta) {
-          result.publishTime = timeMeta.getAttribute('content')
-            || timeMeta.getAttribute('datetime')
-            || timeMeta.textContent?.trim()
-            || '';
-        }
-
-        // --- Content extraction ---
-        // Strategy: try semantic elements first, then fall back to largest text block
-        let contentEl = null;
-
-        // 1. <article>
-        const articles = document.querySelectorAll('article');
-        if (articles.length === 1) {
-          contentEl = articles[0];
-        } else if (articles.length > 1) {
-          // Pick the largest article by text length
-          let maxLen = 0;
-          articles.forEach(a => {
-            const len = a.textContent?.length || 0;
-            if (len > maxLen) { maxLen = len; contentEl = a; }
-          });
-        }
-
-        // 2. [role="main"]
-        if (!contentEl) {
-          contentEl = document.querySelector('[role="main"]');
-        }
-
-        // 3. <main>
-        if (!contentEl) {
-          contentEl = document.querySelector('main');
-        }
-
-        // 4. Largest text-dense block fallback
-        if (!contentEl) {
-          const candidates = document.querySelectorAll(
-            'div[class*="content"], div[class*="article"], div[class*="post"], ' +
-            'div[class*="entry"], div[class*="body"], div[id*="content"], ' +
-            'div[id*="article"], div[id*="post"], section'
-          );
-          let maxLen = 0;
-          candidates.forEach(c => {
-            const len = c.textContent?.length || 0;
-            if (len > maxLen) { maxLen = len; contentEl = c; }
-          });
-        }
-
-        // 5. Last resort: document.body
-        if (!contentEl || (contentEl.textContent?.length || 0) < 200) {
-          contentEl = document.body;
-        }
-
-        // Clean up noise elements before extraction
-        const clone = contentEl.cloneNode(true);
-        const noise = 'nav, header, footer, aside, .sidebar, .nav, .menu, .footer, ' +
-          '.header, .comments, .comment, .ad, .ads, .advertisement, .social-share, ' +
-          '.related-posts, .newsletter, .cookie-banner, script, style, noscript, iframe';
-        clone.querySelectorAll(noise).forEach(el => el.remove());
-
-        // Deduplicate: some sites (e.g. Anthropic) render each paragraph twice
-        // (a visible version + a line-broken animation version with missing spaces).
-        // Compare by stripping ALL whitespace so "Hello world" matches "Helloworld".
-        const stripWS = (s) => (s || '').replace(/\\s+/g, '');
-        const dedup = (parent) => {
-          const children = Array.from(parent.children || []);
-          for (let i = children.length - 1; i >= 1; i--) {
-            const curRaw = children[i].textContent || '';
-            const prevRaw = children[i - 1].textContent || '';
-            const cur = stripWS(curRaw);
-            const prev = stripWS(prevRaw);
-            if (cur.length < 20 || prev.length < 20) continue;
-            // Exact match after whitespace strip, or >90% overlap
-            if (cur === prev) {
-              // Keep the one with more proper spacing (more spaces = better formatted)
-              const curSpaces = (curRaw.match(/ /g) || []).length;
-              const prevSpaces = (prevRaw.match(/ /g) || []).length;
-              if (curSpaces >= prevSpaces) children[i - 1].remove();
-              else children[i].remove();
-            } else if (prev.includes(cur) && cur.length / prev.length > 0.8) {
-              children[i].remove();
-            } else if (cur.includes(prev) && prev.length / cur.length > 0.8) {
-              children[i - 1].remove();
-            }
-          }
-        };
-        dedup(clone);
-        clone.querySelectorAll('section, div').forEach(el => {
-          if (el.children && el.children.length > 2) dedup(el);
-        });
-
-        result.contentHtml = clone.innerHTML;
-
-        // --- Image extraction ---
-        const seen = new Set();
-        clone.querySelectorAll('img').forEach(img => {
-          const src = img.getAttribute('data-src')
-            || img.getAttribute('data-original')
-            || img.getAttribute('src');
-          if (src && !src.startsWith('data:') && !seen.has(src)) {
-            seen.add(src);
-            result.imageUrls.push(src);
-          }
-        });
-
-        return result;
-      })()
-    `);
-
-    // Determine Referer from URL for image downloads
-    let referer = '';
-    try {
-      const parsed = new URL(url);
-      referer = parsed.origin + '/';
-    } catch { /* ignore */ }
-
-    return downloadArticle(
-      {
-        title: data?.title || 'untitled',
-        author: data?.author,
-        publishTime: data?.publishTime,
-        sourceUrl: url,
-        contentHtml: data?.contentHtml || '',
-        imageUrls: data?.imageUrls,
-      },
-      {
-        output: kwargs.output,
-        downloadImages: kwargs['download-images'],
-        imageHeaders: referer ? { Referer: referer } : undefined,
-      },
-    );
-  },
-});

package/src/clis/weibo/comments.ts

@@ -1,54 +0,0 @@
-/**
- * Weibo comments — get comments on a post.
- */
-import { cli, Strategy } from '../../registry.js';
-
-cli({
-  site: 'weibo',
-  name: 'comments',
-  description: 'Get comments on a Weibo post',
-  domain: 'weibo.com',
-  strategy: Strategy.COOKIE,
-  args: [
-    { name: 'id', required: true, positional: true, help: 'Post ID (numeric idstr)' },
-    { name: 'limit', type: 'int', default: 20, help: 'Number of comments (max 50)' },
-  ],
-  columns: ['rank', 'author', 'text', 'likes', 'replies', 'time'],
-  func: async (page, kwargs) => {
-    const count = Math.min(kwargs.limit || 20, 50);
-    await page.goto('https://weibo.com');
-    await page.wait(2);
-
-    const id = String(kwargs.id);
-    const data = await page.evaluate(`
-      (async () => {
-        const id = ${JSON.stringify(id)};
-        const count = ${count};
-        const strip = (html) => (html || '').replace(/<[^>]+>/g, '').replace(/&nbsp;/g, ' ').replace(/&lt;/g, '<').replace(/&gt;/g, '>').replace(/&amp;/g, '&').trim();
-
-        const url = '/ajax/statuses/buildComments?flow=0&is_reload=1&id=' + id + '&is_show_bulletin=2&is_mix=0&count=' + count;
-        const resp = await fetch(url, {credentials: 'include'});
-        if (!resp.ok) return {error: 'HTTP ' + resp.status};
-        const data = await resp.json();
-        if (!data.ok) return {error: 'API error: ' + (data.msg || 'unknown')};
-
-        return (data.data || []).map((c, i) => {
-          const item = {
-            rank: i + 1,
-            author: c.user?.screen_name || '',
-            text: strip(c.text || ''),
-            likes: c.like_count || 0,
-            replies: c.total_number || 0,
-            time: c.created_at || '',
-          };
-          if (c.reply_comment) {
-            item.reply_to = (c.reply_comment.user?.screen_name || '') + ': ' + strip(c.reply_comment.text || '').substring(0, 80);
-          }
-          return item;
-        });
-      })()
-    `);
-    if (!Array.isArray(data)) return [];
-    return data;
-  },
-});