astrbot_plugin_private_comp.../qzone_recent_parser.py

219 行
7.6 KiB
Python

2026-07-24 00:25:43 +08:00
# -*- coding: utf-8 -*-
from __future__ import annotations
import re
from html import unescape
from html.parser import HTMLParser
from types import SimpleNamespace
from typing import Any
class _QzoneFeedHtmlParser(HTMLParser):
_VOID_TAGS = {
"area",
"base",
"br",
"col",
"embed",
"hr",
"img",
"input",
"link",
"meta",
"param",
"source",
"track",
"wbr",
}
def __init__(self) -> None:
super().__init__(convert_charrefs=True)
self.text_parts: list[str] = []
self.repost_parts: list[str] = []
self.image_urls: list[str] = []
self.image_items: list[dict[str, str]] = []
self._class_stack: list[set[str]] = []
self._tag_stack: list[str] = []
@staticmethod
def _classes(attrs: list[tuple[str, str | None]]) -> set[str]:
for key, value in attrs:
if str(key or "").lower() == "class":
return {item.strip() for item in str(value or "").split() if item.strip()}
return set()
@staticmethod
def _attr(attrs: list[tuple[str, str | None]], name: str) -> str:
lowered = str(name or "").lower()
for key, value in attrs:
if str(key or "").lower() == lowered:
return str(value or "")
return ""
@classmethod
def _first_attr(cls, attrs: list[tuple[str, str | None]], *names: str) -> str:
for name in names:
value = cls._attr(attrs, name).strip()
if value:
return value
return ""
@staticmethod
def _is_content_image_url(value: str) -> bool:
source = str(value or "").strip().lower()
return bool(source) and not source.startswith("http://qzonestyle.gtimg.cn")
def _record_image(self, attrs: list[tuple[str, str | None]]) -> None:
preview_url = self._first_attr(attrs, "src", "data-src", "data-lazy-src", "data-original-src")
full_url = self._first_attr(
attrs,
"data-original",
"data-origin",
"data-origin-url",
"data-original-url",
"data-originalurl",
"data-big-url",
"data-bigurl",
"data-full-url",
"data-fullurl",
"data-raw-url",
"data-rawurl",
"origsrc",
"original",
"data-url",
)
if not self._is_content_image_url(full_url):
full_url = ""
if not self._is_content_image_url(preview_url):
preview_url = ""
source = full_url or preview_url
if not source:
return
item = {"preview_url": preview_url or source, "full_url": source}
if item not in self.image_items:
self.image_items.append(item)
if source not in self.image_urls:
self.image_urls.append(source)
def handle_starttag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
normalized_tag = str(tag or "").lower()
classes = self._classes(attrs)
if normalized_tag != "img":
if normalized_tag not in self._VOID_TAGS:
self._tag_stack.append(normalized_tag)
self._class_stack.append(classes)
elif classes:
self._class_stack.append(classes)
self._class_stack.pop()
return
active_classes = self._class_stack + [classes]
if any({"img-box", "video-img"} & item for item in active_classes):
self._record_image(attrs)
def handle_startendtag(self, tag: str, attrs: list[tuple[str, str | None]]) -> None:
normalized_tag = str(tag or "").lower()
if normalized_tag == "img":
self.handle_starttag(tag, attrs)
return
classes = self._classes(attrs)
if classes:
self._class_stack.append(classes)
self._class_stack.pop()
def handle_endtag(self, tag: str) -> None:
normalized_tag = str(tag or "").lower()
for index in range(len(self._tag_stack) - 1, -1, -1):
if self._tag_stack[index] == normalized_tag:
del self._tag_stack[index:]
del self._class_stack[index:]
return
if self._tag_stack and self._class_stack:
self._tag_stack.pop()
self._class_stack.pop()
def handle_data(self, data: str) -> None:
text = unescape(str(data or "")).strip()
if not text:
return
if any("f-info" in item for item in self._class_stack):
self.text_parts.append(text)
if any("txt-box" in item for item in self._class_stack):
self.repost_parts.append(text)
def _parse_feed_html(html_content: str) -> tuple[str, str, list[str]]:
text, repost, image_urls, _items = _parse_feed_html_details(html_content)
return text, repost, image_urls
def _parse_feed_html_details(html_content: str) -> tuple[str, str, list[str], list[dict[str, str]]]:
parser = _QzoneFeedHtmlParser()
parser.feed(str(html_content or ""))
text = "".join(parser.text_parts).strip()
repost = "".join(parser.repost_parts).strip()
if ":" in repost:
repost = repost.split(":", 1)[1].strip()
return text, repost, parser.image_urls, parser.image_items
def _html_attr_value(html_content: str, *names: str) -> str:
source = str(html_content or "")
for name in names:
escaped = re.escape(str(name or ""))
match = re.search(rf"""{escaped}\s*=\s*["']([^"']+)["']""", source, flags=re.IGNORECASE)
if match:
return unescape(match.group(1)).strip()
return ""
def parse_recent_feeds(data: dict[str, Any]) -> list[Any]:
feeds: Any = []
if isinstance(data, dict):
first = data.get("data")
if isinstance(first, dict):
second = first.get("data")
feeds = second if isinstance(second, list) else []
elif isinstance(first, list):
feeds = first
elif isinstance(data.get("msglist"), list):
feeds = data.get("msglist")
if not isinstance(feeds, list) or not feeds:
return []
posts: list[Any] = []
for feed in feeds:
if not feed:
continue
appid = str(feed.get("appid", ""))
if appid != "311":
continue
uin = int(feed.get("uin") or 0)
tid = str(feed.get("key") or "")
if not uin or not tid:
continue
html_content = str(feed.get("html") or "")
if not html_content:
continue
text, rt_con, image_urls, image_items = _parse_feed_html_details(html_content)
posts.append(
SimpleNamespace(
tid=tid,
uin=uin,
name=str(feed.get("nickname") or feed.get("name") or uin),
text=text,
rt_con=rt_con,
images=image_urls,
image_items=image_items,
comments=[],
create_time=int(feed.get("abstime") or 0),
appid=str(feed.get("appid") or "311"),
typeid=str(feed.get("typeid") or feed.get("type") or "0"),
abstime=int(feed.get("abstime") or 0),
fid=str(feed.get("key") or tid),
unikey=str(feed.get("unikey") or feed.get("likeKey") or feed.get("like_key") or _html_attr_value(html_content, "data-unikey", "unikey")),
curkey=str(feed.get("curkey") or feed.get("curlikekey") or feed.get("likeKey") or feed.get("like_key") or _html_attr_value(html_content, "data-curkey", "curkey")),
raw=feed,
status="approved",
)
)
return posts