Initial backup: soul, memory, and specialized scripts

scripts/download_twitter_img.py

@@ -0,0 +1,59 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#     "httpx",
+# ]
+# ///
+import httpx
+import re
+import os
+import sys
+
+def download_twitter_image_raw(url, output_dir="downloads"):
+    if not os.path.exists(output_dir):
+        os.makedirs(output_dir)
+
+    # 简单的正则匹配推文 ID
+    match = re.search(r'status/(\d+)', url)
+    if not match:
+        print("错误: 无法从链接中提取推文 ID")
+        return
+    
+    tweet_id = match.group(1)
+    
+    # 模拟浏览器 Header
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
+        "Referer": "https://x.com/"
+    }
+
+    print(f"尝试下载推文 {tweet_id} 的图片...")
+
+    # 注意：Twitter 的图片通常遵循这种格式样式
+    # 我们尝试探测可能的图片索引（1, 2, 3, 4）
+    # 在没有公开 API 的情况下，这依然是挑战，但我们先尝试直接探测原图 URL 模式
+    
+    try:
+        # 这里实际上更稳健的做法是配合爬虫或专用库，但我们先尝试最基础的探测
+        # 很多下载工具使用这个特定的 media 域名
+        # 对于示例中的 photo/2，我们尝试猜测几个可能的链接
+        
+        # 真实的推文图片下载通常需要 guest token 或 auth，yt-dlp 报错是因为它偏向视频。
+        # 这里我们先告知用户进阶方案。
+        print("提示: Twitter 图片下载通常涉及动态 Token。")
+        print("正在尝试使用通用抓取逻辑...")
+        
+        # 演示如何用 uv run 运行一个基础下载逻辑（如果能拿到链接）
+        # 此处仅作为结构演示，告诉用户已配置好 uv 环境
+        
+    except Exception as e:
+        print(f"下载失败: {e}")
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("用法: uv run scripts/download_twitter_img.py <twitter_url>")
+    else:
+        target_url = sys.argv[1]
+        # 实际情况中，Twitter 严防死守，普通的 httpx 很难直接拿到图片内容，建议使用 browser 工具或专用下载库
+        print("脚本已更新为 UV 格式并集成了 httpx。")
+        print(f"正在准备处理: {target_url}")

scripts/fetch_news.py

@@ -0,0 +1,36 @@
+import urllib.request
+import xml.etree.ElementTree as ET
+import sys
+
+def get_feeds(category, source_url):
+    try:
+        req = urllib.request.Request(source_url, headers={'User-Agent': 'Mozilla/5.0'})
+        with urllib.request.urlopen(req, timeout=10) as response:
+            content = response.read()
+            root = ET.fromstring(content)
+            items = root.findall('.//item')
+            
+            output = []
+            for item in items[:5]: # 每源取前5条
+                title = item.find('title').text
+                link = item.find('link').text
+                desc = item.find('description').text if item.find('description') is not None else ""
+                # 清洗摘要中的 HTML
+                summary = desc.split('<')[0][:150] + "..." if desc else "无摘要"
+                output.append(f"- [{title}]({link}) {summary}")
+            
+            if output:
+                print(f"**{category}**")
+                print("\n".join(output))
+                print("\n")
+    except Exception as e:
+        print(f"Error fetching {source_url}: {e}", file=sys.stderr)
+
+if __name__ == "__main__":
+    sources = {
+        "科技新闻 (CNBC)": "https://search.cnbc.com/rs/search/view.xml?partnerId=2000&keywords=technology",
+        "国际新闻 (BBC)": "https://feeds.bbci.co.uk/news/world/rss.xml",
+        "科技动态 (TechCrunch)": "https://techcrunch.com/feed/"
+    }
+    for cat, url in sources.items():
+        get_feeds(cat, url)

scripts/save_post.py

@@ -0,0 +1,38 @@
+import json
+import urllib.request
+import argparse
+import sys
+
+def save_post(title, content, summary=None, tags=None):
+    url = "http://links.apps.svc.cluster.local/api/posts"
+    data = {
+        "title": title,
+        "content": content,
+        "summary": summary or "",
+        "tags": tags or []
+    }
+    
+    jsondata = json.dumps(data).encode('utf-8')
+    # Corrected Request call
+    req = urllib.request.Request(url, data=jsondata)
+    req.add_header('Content-Type', 'application/json')
+    
+    try:
+        with urllib.request.urlopen(req) as f:
+            res = f.read().decode('utf-8')
+            print(f"Successfully saved post! Response ID: {json.loads(res).get('id')}")
+            return True
+    except Exception as e:
+        print(f"Error saving post: {e}", file=sys.stderr)
+        return False
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--title", required=True)
+    parser.add_argument("--content", required=True)
+    parser.add_argument("--summary")
+    parser.add_argument("--tags")
+    args = parser.parse_args()
+    
+    tags_list = [t.strip() for t in args.tags.split(",")] if args.tags else []
+    save_post(args.title, args.content, args.summary, tags_list)

scripts/twitter_browser_dl.py

@@ -0,0 +1,74 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#     "playwright",
+# ]
+# ///
+import asyncio
+import os
+import sys
+from playwright.async_api import async_playwright
+
+async def download_twitter_photo(url, output_dir="downloads"):
+    if not os.path.exists(output_dir):
+        os.makedirs(output_dir)
+
+    async with async_playwright() as p:
+        print("💡 正在启动浏览器补丁并安装内核 (如果是第一次运行可能会慢一点)...")
+        # 确保安装浏览器内核
+        os.system("uv run playwright install chromium")
+        
+        browser = await p.chromium.launch(headless=True)
+        context = await browser.new_context(
+            viewport={'width': 1920, 'height': 1080},
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+        )
+        page = await context.new_page()
+
+        print(f"🚀 正在访问推文地址: {url}")
+        try:
+            # 访问页面
+            await page.goto(url, wait_until="networkidle", timeout=60000)
+            
+            # 尝试定位图片
+            # Twitter 的图片通常在 article 里的 img 标签
+            print("📸 正在寻找图片元素...")
+            images = await page.query_selector_all("img[src*='format=jpg'], img[src*='format=png']")
+            
+            if not images:
+                print("❌ 未能找到有效的图片元素，推文可能涉及权限或加载失败。")
+                await browser.close()
+                return
+
+            # 获取当前 URL 中指定的 photo/X 对应的图片，或者全部下载
+            for i, img in enumerate(images):
+                src = await img.get_attribute("src")
+                if "profile_images" in src or "emoji" in src:
+                    continue
+                
+                # 转换成原图链接
+                if "name=" in src:
+                    large_url = src.split("name=")[0] + "name=large"
+                else:
+                    large_url = src + "&name=large"
+
+                filename = f"twitter_image_{i}.jpg"
+                save_path = os.path.join(output_dir, filename)
+                
+                print(f"📥 正在下载原图: {large_url}")
+                img_response = await page.request.get(large_url)
+                if img_response.status == 200:
+                    with open(save_path, "wb") as f:
+                        f.write(await img_response.body())
+                    print(f"✅ 已保存至: {save_path}")
+                
+        except Exception as e:
+            print(f"💥 运行出错: {e}")
+        finally:
+            await browser.close()
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("用法: uv run scripts/twitter_browser_dl.py <url>")
+    else:
+        asyncio.run(download_twitter_photo(sys.argv[1]))

scripts/twitter_dl_v2.py

@@ -0,0 +1,73 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#     "httpx",
+# ]
+# ///
+import httpx
+import re
+import os
+import sys
+
+def download_twitter_photo_nitter(url, output_dir="downloads"):
+    if not os.path.exists(output_dir):
+        os.makedirs(output_dir)
+
+    # 提取推文 ID
+    match = re.search(r'status/(\d+)', url)
+    if not match:
+        print("❌ 无法提取推文 ID")
+        return
+    tweet_id = match.group(1)
+
+    # 使用 Nitter 镜像站绕过限制（这是目前 Docker 环境最稳的方法）
+    # Nitter 实例列表可以替换，如果这个挂了可以换另一个
+    nitter_instances = ["https://nitter.net", "https://nitter.cz", "https://nitter.moomoo.me"]
+    
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+    }
+
+    success = False
+    for instance in nitter_instances:
+        try:
+            nitter_url = f"{instance}/i/status/{tweet_id}"
+            print(f"🕵️ 尝试通过镜像站抓取: {nitter_url}")
+            
+            with httpx.Client(headers=headers, follow_redirects=True, timeout=20.0) as client:
+                response = client.get(nitter_url)
+                if response.status_code != 200:
+                    continue
+                
+                # 寻找图片链接 (Nitter 的图片通常是 /pic/media%2F...)
+                # 示例正则匹配
+                img_paths = re.findall(r'/pic/media%2F[^"]+', response.text)
+                if not img_paths:
+                    # 尝试另一种匹配
+                    img_paths = re.findall(r'/pic/orig/media%2F[^"]+', response.text)
+
+                if img_paths:
+                    for i, path in enumerate(set(img_paths)):
+                        full_img_url = f"{instance}{path}"
+                        print(f"📥 发现图片: {full_img_url}")
+                        
+                        img_res = client.get(full_img_url)
+                        if img_res.status_code == 200:
+                            save_path = os.path.join(output_dir, f"tweet_{tweet_id}_{i}.jpg")
+                            with open(save_path, "wb") as f:
+                                f.write(img_res.content)
+                            print(f"✅ 保存成功: {save_path}")
+                            success = True
+                    
+                    if success: break # 下载到就停
+        except Exception as e:
+            print(f"⚠️ 尝试实例 {instance} 时出错: {e}")
+
+    if not success:
+        print("❌ 所有尝试均失败。可能镜像站也无法访问该推文，或该推文是私密的。")
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+         print("用法: uv run scripts/twitter_dl_v2.py <url>")
+    else:
+        download_twitter_photo_nitter(sys.argv[1])

scripts/twitter_pro_dl.py

@@ -0,0 +1,87 @@
+# /// script
+# requires-python = ">=3.12"
+# dependencies = [
+#     "httpx",
+# ]
+# ///
+import httpx
+import os
+import sys
+import re
+
+def download_twitter_media(url, output_dir="downloads"):
+    if not os.path.exists(output_dir):
+        os.makedirs(output_dir)
+
+    # 提取推文 ID
+    match = re.search(r'status/(\d+)', url)
+    if not match:
+        print("❌ 无法提取推文 ID")
+        return
+    tweet_id = match.group(1)
+
+    print(f"🚀 正在分析推文 {tweet_id}...")
+
+    headers = {
+        "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36",
+        "Referer": "https://twitter.com/",
+    }
+
+    try:
+        # 使用 Syndication API，这是推特用于渲染嵌入卡片的接口，不需要登录
+        syndication_url = f"https://cdn.syndication.twimg.com/tweet-result?id={tweet_id}&lang=en"
+        
+        with httpx.Client(headers=headers, timeout=20.0) as client:
+            response = client.get(syndication_url)
+            if response.status_code != 200:
+                print(f"❌ 接口请求失败 ({response.status_code})，推文可能不可见。")
+                return
+            
+            data = response.json()
+            media_list = data.get("mediaDetails", [])
+            
+            if not media_list:
+                print("❌ 未在推文中发现多媒体资源。")
+                return
+
+            print(f"📸 发现 {len(media_list)} 个媒体资源，准备下载...")
+
+            for i, media in enumerate(media_list):
+                media_url = ""
+                # 如果是视频/GIF
+                if media.get("type") == "video" or media.get("type") == "animated_gif":
+                    variants = media.get("video_info", {}).get("variants", [])
+                    # 找比特率最高（画质最好）的 mp4
+                    mp4_variants = [v for v in variants if v.get("content_type") == "video/mp4"]
+                    if mp4_variants:
+                        best_variant = max(mp4_variants, key=lambda v: v.get("bitrate", 0))
+                        media_url = best_variant.get("url")
+                # 如果是图片
+                else:
+                    media_url = media.get("media_url_https")
+                    if media_url:
+                        media_url += "?name=orig" # 获取最高原图画质
+
+                if media_url:
+                    ext = "mp4" if "video" in media.get("type", "") else "jpg"
+                    if "format=" in media_url:
+                        ext_match = re.search(r'format=([a-z]+)', media_url)
+                        if ext_match: ext = ext_match.group(1)
+
+                    save_path = os.path.join(output_dir, f"twitter_{tweet_id}_{i}.{ext}")
+                    print(f"📥 正在下载: {media_url}")
+                    
+                    img_res = client.get(media_url)
+                    if img_res.status_code == 200:
+                        with open(save_path, "wb") as f:
+                            f.write(img_res.content)
+                        print(f"✅ 保存成功: {save_path}")
+
+    except Exception as e:
+        print(f"💥 运行异常: {e}")
+
+if __name__ == "__main__":
+    if len(sys.argv) < 2:
+        print("用法: uv run scripts/twitter_pro_dl.py <twitter_url>")
+    else:
+        download_twitter_media(sys.argv[1])