zhaojie/weidian
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: 通过点击商品图片获取真实itemID,解决商品链接为空问题

Browse Source 
cart_service: 对没有itemID的预售商品,点击图片触发跳转
从跳转URL中提取真实itemID(支持多种URL格式)
点击后自动关闭新tab并回到购物车继续处理下一个商品
This commit is contained in:
Jeason
2026-04-01 16:23:19 +08:00
parent 28fa04c407
commit 400569ad03
1 changed files with 121 additions and 93 deletions
Download Patch File Download Diff File Expand all files Collapse all files

214
server/services/cart_service.py
View File

@@ -1,16 +1,17 @@
"""
购物车预售商品抓取服务
通过 Playwright 打开购物车页面，拦截 API + DOM 提取商品信息
通过 Playwright 打开购物车页面，从 DOM 提取商品信息，
并通过点击商品图片获取跳转 URL 来提取真实 itemID。
"""
import asyncio
import json
import re
from playwright.async_api import async_playwright
from utils.stealth import stealth_async
from server.services.auth_service import get_browser_context, has_auth
CART_URL = "https://weidian.com/new-cart/index.php"
# 从 DOM 提取购物车商品（含尝试从 Vue 组件获取 itemID）
# 从 DOM 提取购物车商品基本信息
EXTRACT_JS = """() => {
const R = [];
const sws = document.querySelectorAll(
@@ -29,21 +30,35 @@ EXTRACT_JS = """() => {
sale_time: '', presale_type: ''
};
// 尝试从 Vue 组件数据中提取 itemID
// 尝试多种方式提取 itemID
// 1. Vue 组件数据
try {
const vue = iw.__vue__ || (iw.__vue_app__ && iw.__vue_app__._instance);
const vue = iw.__vue__;
if (vue) {
const d = vue.$data || vue.data || vue;
o.item_id = String(d.itemID || d.itemId || d.item_id || '');
const d = vue.$data || vue._data || vue;
o.item_id = String(d.itemID || d.itemId || d.item_id
|| d.goodsId || d.goods_id || '');
// 也检查 props
if (!o.item_id && vue.$props) {
const pp = vue.$props;
o.item_id = String(pp.itemID || pp.itemId
|| pp.item_id || pp.goodsId || '');
}
// 检查 item 对象
if (!o.item_id && d.item) {
o.item_id = String(d.item.itemID || d.item.itemId
|| d.item.item_id || '');
}
}
} catch(e) {}
// 尝试从 data-* 属性提取
// 2. data-* 属性
if (!o.item_id) {
o.item_id = iw.dataset.itemId || iw.dataset.itemid || '';
o.item_id = iw.dataset.itemId || iw.dataset.itemid
|| iw.dataset.goodsId || iw.dataset.id || '';
}
// 尝试从内部链接提取
// 3. 内部链接
if (!o.item_id) {
const a = iw.querySelector('a[href*="itemID"]');
if (a) {
@@ -52,6 +67,10 @@ EXTRACT_JS = """() => {
}
}
// 4. 图片 URL 中可能有商品信息（备用）
const img = iw.querySelector('.item_img img');
o.img_src = img ? (img.src || img.dataset.src || '') : '';
const te = iw.querySelector('.item_title');
if (te) o.title = te.textContent.trim();
const sk = iw.querySelector('.item_sku');
@@ -89,15 +108,12 @@ EXTRACT_JS = """() => {
async def fetch_cart_presale_items(account_id):
"""
获取指定账号购物车中的预售商品列表。
双重提取：拦截购物车 API 获取 itemID 映射 + DOM 提取预售信息。
对于没有 itemID 的商品，通过点击图片获取跳转 URL 来提取。
返回: (success, items_or_msg)
"""
if not has_auth(account_id):
return False, "账号未登录"
# 用于存储 API 返回的 cart_item_id -> itemID 映射
api_item_map = {}
async with async_playwright() as p:
browser, context = await get_browser_context(
p, account_id, headless=True
@@ -105,20 +121,6 @@ async def fetch_cart_presale_items(account_id):
page = await context.new_page()
await stealth_async(page)
# 拦截购物车相关 API，提取 itemID
async def on_response(response):
url = response.url
# 购物车 API 通常包含 cart 相关路径
if any(k in url for k in ['cart/list', 'cart/query', 'cartList', 'getCart',
'cart-server', 'newcart']):
try:
data = await response.json()
_extract_item_ids(data, api_item_map)
except Exception:
pass
page.on("response", on_response)
try:
await page.goto(
CART_URL, wait_until="networkidle", timeout=20000
@@ -137,74 +139,100 @@ async def fetch_cart_presale_items(account_id):
await browser.close()
return False, f"打开购物车失败: {e}"
# 也尝试从页面内嵌的 JS 变量/window 对象提取
try:
extra_map = await page.evaluate("""() => {
const m = {};
// 尝试从 window.__INITIAL_STATE__ 或类似全局变量提取
const sources = [
window.__INITIAL_STATE__,
window.__NUXT__,
window.__APP_DATA__,
window.cartData,
window.__data__,
];
function walk(obj, depth) {
if (!obj || depth > 5) return;
if (Array.isArray(obj)) {
for (const item of obj) walk(item, depth + 1);
} else if (typeof obj === 'object') {
const cid = String(obj.cartItemId || obj.cart_item_id || obj.cartId || '');
const iid = String(obj.itemID || obj.itemId || obj.item_id || obj.goodsId || '');
if (cid && iid && iid !== cid) m[cid] = iid;
// 也存 itemUrl
if (cid && obj.itemUrl) m[cid + '_url'] = obj.itemUrl;
for (const v of Object.values(obj)) walk(v, depth + 1);
}
}
for (const s of sources) { if (s) walk(s, 0); }
return m;
}""")
if extra_map:
api_item_map.update(extra_map)
except Exception:
pass
# 提取基本信息
raw_items = await page.evaluate(EXTRACT_JS)
# 筛选预售商品
presale = [it for it in raw_items if it.get("is_presale")]
# 对没有 itemID 的预售商品，通过点击图片获取跳转 URL
for item in presale:
if item.get('item_id'):
continue
cid = item.get('cart_item_id', '')
if not cid:
continue
item_id = await _get_item_id_by_click(page, context, cid)
if item_id:
item['item_id'] = item_id
item['url'] = f'https://weidian.com/item.html?itemID={item_id}'
await browser.close()
# 合并 API 数据到 DOM 提取结果
for item in raw_items:
cid = item.get('cart_item_id', '')
if not item.get('item_id') and cid in api_item_map:
item['item_id'] = api_item_map[cid]
# 检查是否有 URL
url_key = cid + '_url'
if url_key in api_item_map:
item['url'] = api_item_map[url_key]
# 只返回预售商品
presale = [it for it in raw_items if it.get("is_presale")]
return True, presale
def _extract_item_ids(data, result_map):
"""递归遍历 API 响应 JSON，提取 cart_item_id -> itemID 映射"""
if isinstance(data, list):
for item in data:
_extract_item_ids(item, result_map)
elif isinstance(data, dict):
# 常见字段名
cid = str(data.get('cartItemId', data.get('cart_item_id',
data.get('cartId', ''))))
iid = str(data.get('itemID', data.get('itemId',
data.get('item_id', data.get('goodsId', '')))))
if cid and iid and cid != iid and iid != 'None':
result_map[cid] = iid
# 也提取 URL
item_url = data.get('itemUrl', data.get('item_url', ''))
if cid and item_url:
result_map[cid + '_url'] = item_url
for v in data.values():
if isinstance(v, (dict, list)):
_extract_item_ids(v, result_map)
async def _get_item_id_by_click(page, context, cart_item_id):
"""
通过点击购物车中商品的图片，拦截跳转 URL 来提取 itemID。
点击后会打开新 tab，从新 tab 的 URL 中提取 itemID，然后关闭。
"""
try:
# 定位商品图片
img_locator = page.locator(
f'.item_warp[id="{cart_item_id}"] .item_img')
if await img_locator.count() == 0:
return None
# 监听新页面打开事件
async with context.expect_page(timeout=5000) as new_page_info:
await img_locator.first.click()
new_page = await new_page_info.value
# 等待 URL 加载
await asyncio.sleep(1)
url = new_page.url
# 从 URL 提取 itemID
item_id = _extract_item_id_from_url(url)
# 关闭新 tab
await new_page.close()
return item_id
except Exception:
# 如果 expect_page 超时，可能是在当前页面跳转了
# 检查当前页面 URL
try:
current_url = page.url
item_id = _extract_item_id_from_url(current_url)
if item_id:
# 跳回购物车
await page.goto(CART_URL, wait_until="networkidle",
timeout=15000)
await asyncio.sleep(2)
return item_id
except Exception:
pass
# 确保回到购物车页面
try:
if 'new-cart' not in page.url:
await page.goto(CART_URL, wait_until="networkidle",
timeout=15000)
await asyncio.sleep(2)
except Exception:
pass
return None
def _extract_item_id_from_url(url):
"""从 URL 中提取 itemID"""
if not url:
return None
# https://weidian.com/item.html?itemID=123456
m = re.search(r'itemID=(\d+)', url, re.IGNORECASE)
if m:
return m.group(1)
# https://shop.weidian.com/item/123456
m = re.search(r'/item/(\d+)', url)
if m:
return m.group(1)
# https://weidian.com/...?id=123456
m = re.search(r'[?&]id=(\d+)', url)
if m:
return m.group(1)
return None