246 lines
8.9 KiB
Python
246 lines
8.9 KiB
Python
"""
|
|
香港赛马会新闻采集模块
|
|
- 列表页: https://racingnews.hkjc.com/chinese/?ny=2026&nm=6
|
|
- 详情页: 列表内详情链接
|
|
|
|
写入分类:
|
|
- la_article_cate.id = 21 (彩票资讯)
|
|
"""
|
|
import html
|
|
import json
|
|
import re
|
|
import time
|
|
from datetime import datetime
|
|
from urllib.parse import urljoin
|
|
|
|
import pymysql
|
|
import requests
|
|
from article_publish_helper import has_publishable_article_content, resolve_crawled_article_is_show
|
|
|
|
ARTICLE_CID = 21
|
|
LIST_URL = "https://racingnews.hkjc.com/chinese/?ny=2026&nm=6"
|
|
SOURCE_NAME = "香港赛马会"
|
|
MAX_ITEMS = 20
|
|
|
|
|
|
def clean_text(text: str) -> str:
|
|
return re.sub(r"\s+", " ", html.unescape(text or "").replace("\u3000", " ")).strip()
|
|
|
|
|
|
def parse_hk_date(date_str: str) -> tuple[int, str]:
|
|
if not date_str:
|
|
now = int(time.time())
|
|
return now, datetime.fromtimestamp(now).strftime("%Y-%m-%d %H:%M:%S")
|
|
for fmt in ("%d/%m/%Y %H:%M", "%d/%m/%Y"):
|
|
try:
|
|
dt = datetime.strptime(date_str.strip(), fmt)
|
|
ts = int(dt.timestamp())
|
|
return ts, dt.strftime("%Y-%m-%d %H:%M:%S")
|
|
except Exception:
|
|
continue
|
|
now = int(time.time())
|
|
return now, datetime.fromtimestamp(now).strftime("%Y-%m-%d %H:%M:%S")
|
|
|
|
|
|
def fetch_list(session: requests.Session) -> list[dict]:
|
|
resp = session.get(LIST_URL, timeout=30)
|
|
resp.raise_for_status()
|
|
html_text = resp.text
|
|
matches = re.findall(
|
|
r'<div class="news-listing-container.*?<div class="news-listing-date">(.*?)</div>.*?<div class="news-listing-content-title">\s*<a href="([^"]+)">(.*?)</a>',
|
|
html_text,
|
|
re.I | re.S
|
|
)
|
|
|
|
items = []
|
|
for date_text, href, title in matches[:MAX_ITEMS]:
|
|
items.append({
|
|
"date_text": clean_text(re.sub(r"<[^>]+>", " ", date_text)),
|
|
"source_url": urljoin(LIST_URL, href),
|
|
"title": clean_text(re.sub(r"<[^>]+>", " ", title)),
|
|
})
|
|
return items
|
|
|
|
|
|
def parse_detail(session: requests.Session, url: str) -> dict:
|
|
resp = session.get(url, timeout=30)
|
|
resp.raise_for_status()
|
|
html_text = resp.text
|
|
|
|
heading_matches = re.findall(r"<h2[^>]*>(.*?)</h2>", html_text, re.I | re.S)
|
|
title = ""
|
|
for raw in heading_matches:
|
|
text = clean_text(re.sub(r"<[^>]+>", " ", raw))
|
|
if text and text != "其他新聞":
|
|
title = text
|
|
break
|
|
|
|
paragraphs = []
|
|
for raw in re.findall(r"<p[^>]*>([\s\S]*?)</p>", html_text, re.I):
|
|
text = clean_text(re.sub(r"<[^>]+>", " ", raw))
|
|
if len(text) > 8:
|
|
paragraphs.append(text)
|
|
|
|
images = []
|
|
for src in re.findall(r'<img[^>]+src="([^"]+res\.hkjc\.com[^"]+)"', html_text, re.I):
|
|
full = urljoin(url, src)
|
|
if full not in images:
|
|
images.append(full)
|
|
|
|
published_text = ""
|
|
if paragraphs and re.match(r"^\d{2}/\d{2}/\d{4}", paragraphs[0]):
|
|
published_text = paragraphs.pop(0)
|
|
|
|
content_parts = [f"<p>{text}</p>" for text in paragraphs]
|
|
for image in images[:3]:
|
|
content_parts.append(f'<img src="{image}" />')
|
|
|
|
desc = paragraphs[0][:255] if paragraphs else title[:255]
|
|
return {
|
|
"title": title[:255],
|
|
"desc": desc,
|
|
"content": "".join(content_parts),
|
|
"image": images[0] if images else "",
|
|
"published_text": published_text[:64],
|
|
"source_url": url,
|
|
}
|
|
|
|
|
|
def sync_to_db(conn, articles: list[dict]) -> tuple[int, int, int]:
|
|
if not articles:
|
|
return 0, 0, 0
|
|
|
|
urls = [a["source_url"] for a in articles if a.get("source_url")]
|
|
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()}
|
|
|
|
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()}
|
|
|
|
now = int(time.time())
|
|
new_count = 0
|
|
update_count = 0
|
|
|
|
for article in articles:
|
|
title = article.get("title", "")
|
|
source_url = article.get("source_url", "")
|
|
if not title or not source_url:
|
|
continue
|
|
|
|
ext_json = json.dumps({
|
|
"source": "hkjc_racingnews",
|
|
"list_url": LIST_URL,
|
|
}, ensure_ascii=False)
|
|
|
|
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, abstract=%s, image=%s, author=%s,
|
|
published_at=%s, category=%s, content=%s, ext=%s, is_show=%s, update_time=%s
|
|
WHERE id=%s""",
|
|
(
|
|
title[:255],
|
|
(article.get("desc") or "")[:255],
|
|
(article.get("desc") or "")[:500],
|
|
(article.get("image") or "")[:128],
|
|
SOURCE_NAME,
|
|
(article.get("published_at") or "")[:64],
|
|
"lottery,hk",
|
|
article.get("content") or "",
|
|
ext_json,
|
|
next_is_show,
|
|
now,
|
|
existing_row["id"],
|
|
)
|
|
)
|
|
update_count += 1
|
|
else:
|
|
if 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,
|
|
title[:255],
|
|
(article.get("desc") or "")[:255],
|
|
(article.get("desc") or "")[:500],
|
|
(article.get("image") or "")[:128],
|
|
SOURCE_NAME,
|
|
(article.get("published_at") or "")[:64],
|
|
"lottery,hk",
|
|
source_url[:500],
|
|
article.get("content") or "",
|
|
ext_json,
|
|
resolve_crawled_article_is_show(article.get("content") or ""),
|
|
0,
|
|
0,
|
|
0,
|
|
article.get("published_ts") or now,
|
|
now,
|
|
)
|
|
)
|
|
existing[source_url] = {"id": cur.lastrowid, "content": article.get("content") or "", "is_show": resolve_crawled_article_is_show(article.get("content") or "")}
|
|
new_count += 1
|
|
existing_titles.add(title[:255])
|
|
|
|
return len(articles), new_count, update_count
|
|
|
|
|
|
def run(db_config: dict):
|
|
session = requests.Session()
|
|
session.headers.update({
|
|
"User-Agent": "Mozilla/5.0 (compatible; SportEra/1.0)",
|
|
"Accept": "text/html,*/*",
|
|
})
|
|
|
|
list_items = fetch_list(session)
|
|
print(f"HKJC 资讯列表返回 {len(list_items)} 条")
|
|
articles = []
|
|
for item in list_items:
|
|
detail = parse_detail(session, item["source_url"])
|
|
published_ts, published_at = parse_hk_date(detail.get("published_text") or item.get("date_text", ""))
|
|
articles.append({
|
|
"title": detail.get("title") or item["title"],
|
|
"desc": detail.get("desc", ""),
|
|
"content": detail.get("content", ""),
|
|
"image": detail.get("image", ""),
|
|
"published_ts": published_ts,
|
|
"published_at": published_at,
|
|
"source_url": item["source_url"],
|
|
})
|
|
|
|
if not articles:
|
|
print("无可写入数据，跳过")
|
|
return {"success": True, "candidate_count": 0, "saved_count": 0, "summary_text": "无可写入数据，跳过"}
|
|
|
|
conn = pymysql.connect(**db_config, cursorclass=pymysql.cursors.DictCursor)
|
|
try:
|
|
total, new_count, update_count = sync_to_db(conn, 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()
|