rypblog/app.py

import os
import re
import json
import time
import uuid
import shutil
from datetime import datetime, timezone
from pathlib import Path

from flask import (
    Flask,
    render_template,
    request,
    redirect,
    url_for,
    send_from_directory,
    abort,
    jsonify,
)
from werkzeug.utils import secure_filename
import markdown
from bs4 import BeautifulSoup

import config

app = Flask(__name__)
app.config.from_object(config)

# 确保必要的目录存在
for dir_path in [
    config.UPLOAD_FOLDER,
    config.GENERATED_FOLDER,
    config.POSTS_DATA_FOLDER,
    config.INDEX_FILE.parent,
]:
    dir_path.mkdir(parents=True, exist_ok=True)

# 如果索引文件不存在，创建空列表
if not config.INDEX_FILE.exists():
    with open(config.INDEX_FILE, "w", encoding="utf-8") as f:
        json.dump([], f, ensure_ascii=False, indent=2)


def allowed_file(filename: str) -> bool:
    """检查文件扩展名是否允许"""
    return (
        "." in filename
        and filename.rsplit(".", 1)[1].lower() in config.ALLOWED_EXTENSIONS
    )


def load_index() -> list:
    """加载文章索引"""
    with open(config.INDEX_FILE, "r", encoding="utf-8") as f:
        return json.load(f)


def save_index(index: list):
    """保存文章索引"""
    with open(config.INDEX_FILE, "w", encoding="utf-8") as f:
        json.dump(index, f, ensure_ascii=False, indent=2)


def fix_image_paths(md_content: str, post_id: str) -> str:
    """将 Markdown 中的本地图片路径替换为可访问的 URL"""
    pattern = r'!\[([^\]]*)\]\(([^)]+)\)'

    def repl(match):
        alt, src = match.groups()
        # 仅当 src 不含 '/' 或 'http' 且扩展名为图片时视为本地文件
        if (
            not src.startswith(("http://", "https://", "/", "#"))
            and "." in src
        ):
            new_src = f"/static/uploads/posts/{post_id}/{src}"
            return f'![{alt}]({new_src})'
        return match.group(0)

    return re.sub(pattern, repl, md_content)


def extract_summary(html_body: str, max_chars: int = 200) -> str:
    """从 HTML 中提取纯文本摘要（返回完整文本）"""
    soup = BeautifulSoup(html_body, "html.parser")
    text = soup.get_text()
    # 去除多余空白
    text = re.sub(r"\s+", " ", text).strip()
    return text


def extract_thumbnail(html_body: str) -> str:
    """从 HTML 中提取第一张图片的 URL"""
    soup = BeautifulSoup(html_body, "html.parser")
    img = soup.find("img")
    if img and img.get("src"):
        return img["src"]
    return ""


def render_markdown(md_content: str) -> str:
    """将 Markdown 渲染为 HTML"""
    md = markdown.Markdown(
        extensions=[
            "extra",
            "codehilite",
            "tables",
            "fenced_code",
        ]
    )
    return md.convert(md_content)


def generate_static_page(post_id: str, title: str, html_body: str, date: str, thumbnail: str):
    """生成独立的静态 HTML 文件"""
    rendered = render_template(
        "post_template.html",
        title=title,
        content=html_body,
        date=date,
        thumbnail=thumbnail,
    )
    output_path = config.GENERATED_FOLDER / f"{post_id}.html"
    with open(output_path, "w", encoding="utf-8") as f:
        f.write(rendered)


# ==================== 路由 ====================


@app.route("/")
def index():
    """首页：瀑布流文章列表（展示完整博文）"""
    posts = load_index()
    # 按日期倒序排列
    posts.sort(key=lambda p: p.get("date", ""), reverse=True)

    # 为每篇文章获取内容
    for post in posts:
        post_id = post.get("id", "")
        content = ""

        # 优先从原始 Markdown 文件渲染内容（即使 generated 文件不存在也能显示）
        md_path = config.POSTS_DATA_FOLDER / post_id / "content.md"
        if md_path.exists():
            try:
                with open(md_path, "r", encoding="utf-8") as f:
                    md_content = f.read()
                fixed_md = fix_image_paths(md_content, post_id)
                html_body = render_markdown(fixed_md)
                content = html_body
            except Exception:
                content = ""
        else:
            # 回退到生成的静态 HTML 文件
            generated_path = config.GENERATED_FOLDER / f"{post_id}.html"
            if generated_path.exists():
                try:
                    with open(generated_path, "r", encoding="utf-8") as f:
                        html_content = f.read()
                    soup = BeautifulSoup(html_content, "html.parser")
                    card_summary = soup.find("div", class_="card-summary")
                    if card_summary:
                        content = card_summary.decode_contents()
                except Exception:
                    content = ""

        post["content"] = content

    return render_template("index.html", posts=posts)


@app.route("/upload", methods=["GET", "POST"])
def upload():
    """上传文章"""
    if request.method == "GET":
        return render_template("upload.html")

    # POST 处理
    title = request.form.get("title", "").strip()
    if not title:
        return "标题不能为空", 400

    markdown_file = request.files.get("markdown_file")
    if not markdown_file or markdown_file.filename == "":
        return "请选择 Markdown 文件", 400

    if not allowed_file(markdown_file.filename):
        return "不支持的文件类型，请上传 .md 文件", 400

    # 生成唯一 ID
    post_id = str(int(time.time() * 1000))

    # 创建目录
    upload_dir = config.UPLOAD_FOLDER / post_id
    upload_dir.mkdir(parents=True, exist_ok=True)

    posts_data_dir = config.POSTS_DATA_FOLDER / post_id
    posts_data_dir.mkdir(parents=True, exist_ok=True)

    # 保存 Markdown 文件
    md_path = posts_data_dir / "content.md"
    markdown_file.save(str(md_path))

    # 处理图片上传
    images = request.files.getlist("images")
    for img in images:
        if img and img.filename:
            filename = secure_filename(img.filename)
            if filename:
                img.save(str(upload_dir / filename))

    # 读取 Markdown 内容
    with open(md_path, "r", encoding="utf-8") as f:
        md_content = f.read()

    # 修正图片路径
    fixed_md = fix_image_paths(md_content, post_id)

    # 渲染 HTML
    html_body = render_markdown(fixed_md)

    # 提取摘要
    summary = extract_summary(html_body)

    # 提取缩略图
    thumbnail = extract_thumbnail(html_body)

    # 生成日期
    date_iso = datetime.now(timezone.utc).isoformat()

    # 生成静态页面
    generate_static_page(post_id, title, html_body, date_iso, thumbnail)

    # 更新索引
    index = load_index()
    index.append(
        {
            "id": post_id,
            "title": title,
            "date": date_iso,
            "summary": summary,
            "thumbnail": thumbnail,
        }
    )
    # 按日期倒序排序
    index.sort(key=lambda p: p.get("date", ""), reverse=True)
    save_index(index)

    return redirect(url_for("index"))


@app.route("/post/<post_id>")
def view_post(post_id: str):
    """查看文章详情"""
    # 安全验证：只允许字母数字和下划线
    if not re.match(r"^[a-zA-Z0-9_]+$", post_id):
        abort(404)

    try:
        return send_from_directory(
            config.GENERATED_FOLDER,
            f"{post_id}.html",
        )
    except FileNotFoundError:
        abort(404)


@app.route("/admin")
def admin():
    """管理页面"""
    posts = load_index()
    return render_template("admin.html", posts=posts)


@app.route("/admin/delete/<post_id>", methods=["POST"])
def delete_post(post_id: str):
    """删除文章"""
    # 安全验证
    if not re.match(r"^[a-zA-Z0-9_]+$", post_id):
        abort(404)

    # 删除生成的静态文件
    generated_file = config.GENERATED_FOLDER / f"{post_id}.html"
    if generated_file.exists():
        generated_file.unlink()

    # 删除 posts_data 目录
    posts_data_dir = config.POSTS_DATA_FOLDER / post_id
    if posts_data_dir.exists():
        shutil.rmtree(posts_data_dir)

    # 删除上传的图片目录
    upload_dir = config.UPLOAD_FOLDER / post_id
    if upload_dir.exists():
        shutil.rmtree(upload_dir)

    # 从索引中移除
    index = load_index()
    index = [p for p in index if p["id"] != post_id]
    save_index(index)

    return redirect(url_for("admin"))


@app.errorhandler(404)
def not_found(e):
    return render_template("404.html"), 404


if __name__ == "__main__":
    app.run(host="0.0.0.0", port=5000, debug=True)