简化 parse_input() 为纯调度器,通过遍历 readers 的 supports() 方法识别输入类型,移除 URL 特殊处理和文件检查逻辑。各 reader 的 parse() 方法负责完整验证(文件存在、格式有效性)。 新增功能: - 添加 chardet 编码自动检测,支持多种中文编码回退机制 - 创建统一配置类管理编码、下载超时、日志等级等配置项 - HTML reader 支持本地文件编码检测和 URL 统一处理 安全性改进: - 修复 safe_open_zip() 路径遍历漏洞,使用 pathlib 规范化路径 - 添加边界检查,search_markdown() 检查负数参数 其他改进: - 修复类型注解(argparse.Namespace) - 日志系统仅输出 ERROR 级别,避免干扰 Markdown 输出 - 更新 BaseReader 接口文档,明确 supports() 和 parse() 职责划分 - 同步 delta specs 到主 specs(document-reading、html-reader、configuration、encoding-detection)
90 lines
2.6 KiB
Python
90 lines
2.6 KiB
Python
"""HTML/URL 文件阅读器,支持多种解析方法。"""
|
|
|
|
import os
|
|
from typing import List, Optional, Tuple
|
|
|
|
from readers.base import BaseReader
|
|
from utils import is_url
|
|
import encoding_detection
|
|
|
|
from . import cleaner
|
|
from . import downloader
|
|
from . import trafilatura
|
|
from . import domscribe
|
|
from . import markitdown
|
|
from . import html2text
|
|
|
|
|
|
PARSERS = [
|
|
("trafilatura", lambda c, t: trafilatura.parse(c)),
|
|
("domscribe", lambda c, t: domscribe.parse(c)),
|
|
("MarkItDown", lambda c, t: markitdown.parse(c, t)),
|
|
("html2text", lambda c, t: html2text.parse(c)),
|
|
]
|
|
|
|
|
|
class HtmlReader(BaseReader):
|
|
"""HTML/URL 文件阅读器"""
|
|
|
|
@property
|
|
def supported_extensions(self) -> List[str]:
|
|
return [".html", ".htm"]
|
|
|
|
def supports(self, file_path: str) -> bool:
|
|
return is_url(file_path) or file_path.endswith(('.html', '.htm'))
|
|
|
|
def download_and_parse(self, url: str) -> Tuple[Optional[str], List[str]]:
|
|
"""下载 URL 并解析"""
|
|
all_failures = []
|
|
|
|
# 下载 HTML
|
|
html_content, download_failures = downloader.download_html(url)
|
|
all_failures.extend(download_failures)
|
|
|
|
if html_content is None:
|
|
return None, all_failures
|
|
|
|
# 清理 HTML
|
|
html_content = cleaner.clean_html_content(html_content)
|
|
|
|
# 解析 HTML
|
|
content, parse_failures = self._parse_html_content(html_content, None)
|
|
all_failures.extend(parse_failures)
|
|
|
|
return content, all_failures
|
|
|
|
def _parse_html_content(self, html_content: str, temp_file_path: Optional[str]) -> Tuple[Optional[str], List[str]]:
|
|
"""解析 HTML 内容"""
|
|
failures = []
|
|
content = None
|
|
|
|
for parser_name, parser_func in PARSERS:
|
|
content, error = parser_func(html_content, temp_file_path)
|
|
if content is not None:
|
|
return content, failures
|
|
else:
|
|
failures.append(f"- {parser_name}: {error}")
|
|
|
|
return None, failures
|
|
|
|
def parse(self, file_path: str) -> Tuple[Optional[str], List[str]]:
|
|
all_failures = []
|
|
|
|
# 判断输入类型
|
|
if is_url(file_path):
|
|
return self.download_and_parse(file_path)
|
|
|
|
# 读取本地 HTML 文件,使用编码检测
|
|
html_content, error = encoding_detection.read_text_file(file_path)
|
|
if error:
|
|
return None, [f"- {error}"]
|
|
|
|
# 清理 HTML
|
|
html_content = cleaner.clean_html_content(html_content)
|
|
|
|
# 解析 HTML
|
|
content, parse_failures = self._parse_html_content(html_content, file_path)
|
|
all_failures.extend(parse_failures)
|
|
|
|
return content, all_failures
|