286 lines
10 KiB
Python
286 lines
10 KiB
Python
"""
|
||
CBA新闻采集模块 - 从 Google News RSS 获取CBA新闻并写入 la_article 表
|
||
在 main.py 中通过 cba_news 命令调用
|
||
|
||
数据源（RSS）:
|
||
1. Google News 中文简体 - CBA 篮球（主）
|
||
2. Google News 中文简体 - CBA 联赛（补充）
|
||
|
||
重复判断: 以 source_url 为唯一键，重复则更新标题/摘要
|
||
"""
|
||
import json
|
||
import re
|
||
import time
|
||
import xml.etree.ElementTree as ET
|
||
from datetime import datetime
|
||
from email.utils import parsedate_to_datetime
|
||
|
||
import pymysql
|
||
import requests
|
||
from article_publish_helper import has_publishable_article_content, resolve_crawled_article_is_show
|
||
|
||
# ─── 配置 ───
|
||
ARTICLE_CID = 18 # la_article_cate 中"CBA"的 id
|
||
|
||
RSS_FEEDS = [
|
||
{
|
||
"name": "CBA篮球(中文)",
|
||
"url": "https://news.google.com/rss/search?q=CBA+%E7%AF%AE%E7%90%83&hl=zh-CN&gl=CN&ceid=CN:zh-Hans",
|
||
"author_default": "Google News",
|
||
},
|
||
{
|
||
"name": "CBA联赛(中文)",
|
||
"url": "https://news.google.com/rss/search?q=CBA+%E8%81%94%E8%B5%9B&hl=zh-CN&gl=CN&ceid=CN:zh-Hans",
|
||
"author_default": "Google News",
|
||
},
|
||
]
|
||
|
||
NS = {
|
||
"content": "http://purl.org/rss/1.0/modules/content/",
|
||
"media": "http://search.yahoo.com/mrss/",
|
||
"dc": "http://purl.org/dc/elements/1.1/",
|
||
}
|
||
|
||
|
||
def strip_html(text: str) -> str:
|
||
import html as html_mod
|
||
text = re.sub(r"<[^>]+>", "", text or "")
|
||
text = html_mod.unescape(text)
|
||
text = html_mod.unescape(text)
|
||
text = text.replace("\xa0", " ")
|
||
return text.strip()
|
||
|
||
|
||
def parse_rfc2822(date_str: str) -> tuple[int, str]:
|
||
"""解析 RSS pubDate (RFC 2822)，返回 (unix时间戳, 'Y-m-d H:i:s'格式字符串)"""
|
||
if not date_str:
|
||
now = int(time.time())
|
||
return now, datetime.fromtimestamp(now).strftime("%Y-%m-%d %H:%M:%S")
|
||
try:
|
||
dt = parsedate_to_datetime(date_str)
|
||
return int(dt.timestamp()), dt.strftime("%Y-%m-%d %H:%M:%S")
|
||
except Exception:
|
||
now = int(time.time())
|
||
return now, datetime.fromtimestamp(now).strftime("%Y-%m-%d %H:%M:%S")
|
||
|
||
|
||
# ─── RSS 采集 ───
|
||
def fetch_rss(session: requests.Session, feed: dict) -> list[dict]:
|
||
"""从单个 RSS 源拉取新闻，返回标准化列表"""
|
||
articles = []
|
||
url = feed["url"]
|
||
name = feed["name"]
|
||
|
||
try:
|
||
resp = session.get(url, timeout=30)
|
||
except Exception as e:
|
||
print(f" [WARN] {name} 请求失败: {e}")
|
||
return articles
|
||
|
||
if resp.status_code != 200:
|
||
print(f" [WARN] {name} HTTP {resp.status_code}")
|
||
return articles
|
||
|
||
try:
|
||
root = ET.fromstring(resp.content)
|
||
except ET.ParseError as e:
|
||
print(f" [WARN] {name} XML 解析失败: {e}")
|
||
return articles
|
||
|
||
items = root.findall(".//item")
|
||
for item in items:
|
||
title = (item.findtext("title", "") or "").strip()
|
||
link = (item.findtext("link", "") or "").strip()
|
||
if not title or not link:
|
||
continue
|
||
|
||
pub_date = item.findtext("pubDate", "")
|
||
desc_raw = item.findtext("description", "") or ""
|
||
desc = strip_html(desc_raw)[:255]
|
||
|
||
# Google News 标题格式: "标题 - 来源名"
|
||
author = feed.get("author_default", name)
|
||
if " - " in title:
|
||
parts = title.rsplit(" - ", 1)
|
||
if len(parts) == 2:
|
||
author = parts[1].strip()
|
||
title = parts[0].strip()
|
||
|
||
# 尝试提取图片
|
||
image = ""
|
||
media_el = item.find("media:content", NS)
|
||
if media_el is not None:
|
||
image = media_el.get("url", "")
|
||
if not image:
|
||
media_thumb = item.find("media:thumbnail", NS)
|
||
if media_thumb is not None:
|
||
image = media_thumb.get("url", "")
|
||
if not image:
|
||
img_match = re.search(r'<img[^>]+src=["\']([^"\']+)["\']', desc_raw)
|
||
if img_match:
|
||
image = img_match.group(1)
|
||
|
||
pub_ts, pub_fmt = parse_rfc2822(pub_date)
|
||
articles.append({
|
||
"title": title[:255],
|
||
"desc": desc,
|
||
"content": strip_html(desc_raw),
|
||
"source_url": link,
|
||
"image": image,
|
||
"author": author[:64],
|
||
"published_ts": pub_ts,
|
||
"published_at": pub_fmt,
|
||
"category": "CBA",
|
||
"ext": {
|
||
"source": "google_news",
|
||
"feed_name": name,
|
||
"feed_url": url,
|
||
},
|
||
})
|
||
|
||
return articles
|
||
|
||
|
||
# ─── 数据库同步 ───
|
||
def sync_to_db(conn, articles: list[dict]) -> tuple[int, int, int]:
|
||
"""同步到 la_article 表，返回 (总数, 新增数, 更新数)"""
|
||
if not articles:
|
||
return 0, 0, 0
|
||
|
||
urls = [a["source_url"] for a in articles if a.get("source_url")]
|
||
if not urls:
|
||
return 0, 0, 0
|
||
|
||
with conn.cursor() as cur:
|
||
placeholders = ",".join(["%s"] * len(urls))
|
||
cur.execute(
|
||
f"SELECT id, source_url, content, is_show FROM la_article WHERE source_url IN ({placeholders}) AND cid = %s",
|
||
urls + [ARTICLE_CID]
|
||
)
|
||
existing = {row["source_url"]: row for row in cur.fetchall()}
|
||
|
||
# 按标题去重：同分类下相同标题的文章不再插入
|
||
titles = [a["title"][:255] for a in articles if a.get("title")]
|
||
existing_titles = set()
|
||
if titles:
|
||
with conn.cursor() as cur:
|
||
t_placeholders = ",".join(["%s"] * len(titles))
|
||
cur.execute(
|
||
f"SELECT title FROM la_article WHERE title IN ({t_placeholders}) AND cid = %s",
|
||
titles + [ARTICLE_CID]
|
||
)
|
||
existing_titles = {row["title"] for row in cur.fetchall()}
|
||
|
||
new_count = 0
|
||
update_count = 0
|
||
now = int(time.time())
|
||
|
||
for article in articles:
|
||
source_url = article.get("source_url", "")
|
||
if not source_url or not article.get("title"):
|
||
continue
|
||
|
||
create_ts = article.get("published_ts") or now
|
||
ext_json = json.dumps(article.get("ext", {}), ensure_ascii=False) if article.get("ext") else None
|
||
|
||
if source_url in existing:
|
||
existing_row = existing[source_url]
|
||
next_is_show = existing_row["is_show"] if has_publishable_article_content(existing_row.get("content", "")) else 0
|
||
with conn.cursor() as cur:
|
||
cur.execute(
|
||
"""UPDATE la_article SET
|
||
title = %s, `desc` = %s, image = %s, author = %s,
|
||
category = %s, ext = %s, is_show = %s, update_time = %s
|
||
WHERE id = %s""",
|
||
(
|
||
article["title"][:255],
|
||
(article.get("desc") or "")[:255],
|
||
article.get("image", ""),
|
||
(article.get("author") or "")[:64],
|
||
(article.get("category") or "")[:128],
|
||
ext_json,
|
||
next_is_show,
|
||
now,
|
||
existing_row["id"],
|
||
)
|
||
)
|
||
update_count += 1
|
||
else:
|
||
# 标题去重：相同标题跳过
|
||
if article["title"][:255] in existing_titles:
|
||
continue
|
||
with conn.cursor() as cur:
|
||
cur.execute(
|
||
"""INSERT INTO la_article
|
||
(cid, title, `desc`, abstract, image, author, published_at,
|
||
category, source_url, content, ext, is_show, sort,
|
||
click_virtual, click_actual, create_time, update_time)
|
||
VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s)""",
|
||
(
|
||
ARTICLE_CID,
|
||
article["title"][:255],
|
||
(article.get("desc") or "")[:255],
|
||
(article.get("desc") or "")[:500],
|
||
article.get("image", ""),
|
||
(article.get("author") or "")[:64],
|
||
(article.get("published_at") or "")[:64],
|
||
(article.get("category") or "")[:128],
|
||
source_url[:500],
|
||
article.get("content") or "",
|
||
ext_json,
|
||
resolve_crawled_article_is_show(article.get("content") or ""),
|
||
0,
|
||
0,
|
||
0,
|
||
create_ts,
|
||
now,
|
||
)
|
||
)
|
||
existing[source_url] = {"id": cur.lastrowid, "content": article.get("content") or "", "is_show": resolve_crawled_article_is_show(article.get("content") or "")}
|
||
existing_titles.add(article["title"][:255])
|
||
new_count += 1
|
||
|
||
return len(articles), new_count, update_count
|
||
|
||
|
||
# ─── 主入口 ───
|
||
def run(db_config: dict):
|
||
"""主入口，db_config 从 dongqiudi-crawler 配置中传入"""
|
||
session = requests.Session()
|
||
session.headers.update({
|
||
"User-Agent": "Mozilla/5.0 (compatible; SportEra/1.0)",
|
||
"Accept": "application/xml, text/xml, application/rss+xml, */*",
|
||
})
|
||
|
||
all_articles = []
|
||
for feed in RSS_FEEDS:
|
||
print(f"=== {feed['name']} 采集 ===")
|
||
articles = fetch_rss(session, feed)
|
||
print(f" {feed['name']} 返回 {len(articles)} 条")
|
||
all_articles.extend(articles)
|
||
|
||
# 按 source_url 去重
|
||
seen = set()
|
||
deduped = []
|
||
for a in all_articles:
|
||
url = a.get("source_url", "")
|
||
if url and url not in seen:
|
||
seen.add(url)
|
||
deduped.append(a)
|
||
all_articles = deduped
|
||
print(f"去重后共 {len(all_articles)} 条")
|
||
|
||
if not all_articles:
|
||
print("所有数据源均无数据，跳过")
|
||
return {"success": True, "candidate_count": 0, "saved_count": 0, "summary_text": "所有数据源均无数据，跳过"}
|
||
|
||
conn = pymysql.connect(**db_config, cursorclass=pymysql.cursors.DictCursor)
|
||
try:
|
||
total, new_count, update_count = sync_to_db(conn, all_articles)
|
||
conn.commit()
|
||
summary = f"共 {total} 条, 新增 {new_count}, 更新 {update_count}"
|
||
print(summary)
|
||
return {"success": True, "candidate_count": total, "saved_count": new_count + update_count, "summary_text": summary}
|
||
finally:
|
||
conn.close()
|