feat: 替换 HTML 解析实现

jkit/article.py

@@ -1,6 +1,5 @@
 from datetime import datetime
 from enum import Enum
-from re import sub as re_sub
 from typing import (
     TYPE_CHECKING,
     Any,
@@ -13,8 +12,6 @@
 )
 
 from httpx import HTTPStatusError
-from lxml.html import HtmlElement
-from lxml.html import fromstring as parse_html
 
 from jkit._base import (
     DATA_OBJECT_CONFIG,
@@ -30,6 +27,7 @@
     normalize_percentage,
 )
 from jkit.config import CONFIG
+from jkit.constants import BLANK_LINES_REGEX, HTML_TAG_REGEX
 from jkit.exceptions import ResourceUnavailableError
 from jkit.identifier_check import is_article_slug
 from jkit.identifier_convert import article_slug_to_url, article_url_to_slug
@@ -109,9 +107,8 @@ class ArticleInfo(DataObject, **DATA_OBJECT_CONFIG):
 
     @property
     def text_content(self) -> str:
-        html_obj: HtmlElement = parse_html(self.html_content)
-        result = "".join(html_obj.itertext())  # type: ignore
-        return re_sub(r"\s{3,}", "", result)  # 去除多余的空行
+        result = HTML_TAG_REGEX.sub("", self.html_content)
+        return BLANK_LINES_REGEX.sub("\n", result)
 
 
 class ArticleAudioInfo(DataObject, **DATA_OBJECT_CONFIG):

jkit/constants.py

@@ -1,6 +1,9 @@
 from re import compile as regex_compile
 
-USER_NAME_REGEX = regex_compile(r"^[\w]*$")
+MAX_ID = 10**9
+
+HTML_TAG_REGEX = regex_compile("<.*?>")
+BLANK_LINES_REGEX = regex_compile("\n{2,}")
 
 JIANSHU_URL_REGEX = regex_compile(r"^https://www\.jianshu\.com/[a-zA-Z0-9/]*/?$")
 USER_UPLOADED_URL_REGEX = regex_compile(r"^https?:\/\/.*/?$")
@@ -14,9 +17,10 @@
 
 USER_SLUG_REGEX = regex_compile(r"^[a-zA-Z0-9]{6,12}$")
 ARTICLE_SLUG_REGEX = regex_compile(r"^[a-zA-Z0-9]{12}$")
-NOTEBOOK_ID_MIN = 10000000
-NOTEBOOK_ID_MAX = 99999999
 COLLECTION_SLUG_REGEX = regex_compile(r"^[a-zA-Z0-9]{6,12}$")
 ISLAND_SLUG_REGEX = regex_compile(r"^[a-zA-Z0-9]{16}$")
 
-MAX_ID = 10**9
+NOTEBOOK_ID_MIN = 10000000
+NOTEBOOK_ID_MAX = 99999999
+
+USER_NAME_REGEX = regex_compile(r"^[\w]*$")