rypblog/helpers.py

import re
import json
from pathlib import Path
from datetime import datetime, timezone
from flask import render_template
from werkzeug.utils import secure_filename
import markdown
from bs4 import BeautifulSoup
import config

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)