🏀 https://oralb.com/en-us/products/shop-all
import random, csv, requests, logging, time, json, re
from selenium import webdriver
from copy import deepcopy
from time import sleep
from lxml import etree
from DrissionPage import Chromium, ChromiumOptions
from DrissionPage.common import Keys
class Oralb:
def __init__(self):
self.url = 'https://oralb.com/en-us/products/shop-all'
self.headers = {}
self.id = 0 # 保存到csv文件不用id字段
self.init_data = {
'Handle': '',
'Title': '',
'Body (HTML)': '',
'Vendor': '',
'Type': '',
'Tags': '',
'Published': 'TRUE',
'Option1 Name': '',
'Option1 Value': '',
'Option2 Name': '',
'Option2 Value': '',
'Option3 Name': '',
'Option3 Value': '',
'Variant SKU': '',
'Variant Grams': '',
'Variant Inventory Tracker': 'Shopify',
'Variant Inventory Qty': '99999',
'Variant Inventory Policy': '',
'Variant Fulfillment Service': '',
'Variant Price': '',
'Variant Compare At Price': '',
'Variant Requires Shipping': '',
'Variant Taxable': '',
'Variant Barcode': '',
'Image Src': '',
'Image Position': '',
'Image Alt Text': '',
'Gift Card': '',
'SEO Title': '',
'SEO Description': '',
'Variant Image': '',
'Status': '',
'Collection': '',
}
self.empty_data = {
'Handle': '',
'Title': '',
'Body (HTML)': '',
'Vendor': '',
'Type': '',
'Tags': '',
'Published': '',
'Option1 Name': '',
'Option1 Value': '',
'Option2 Name': '',
'Option2 Value': '',
'Option3 Name': '',
'Option3 Value': '',
'Variant SKU': '',
'Variant Grams': '',
'Variant Inventory Tracker': '',
'Variant Inventory Qty': '',
'Variant Inventory Policy': '',
'Variant Fulfillment Service': '',
'Variant Price': '',
'Variant Compare At Price': '',
'Variant Requires Shipping': '',
'Variant Taxable': '',
'Variant Barcode': '',
'Image Src': '',
'Image Position': '',
'Image Alt Text': '',
'Gift Card': '',
'SEO Title': '',
'SEO Description': '',
'Variant Image': '',
'Status': '',
'Collection': '',
}
self.field_names = ['Handle', 'Title', 'Body (HTML)', 'Vendor', 'Type', 'Tags', 'Published', 'Option1 Name',
'Option1 Value', 'Option2 Name', 'Option2 Value', 'Option3 Name', 'Option3 Value',
'Variant SKU', 'Variant Grams', 'Variant Inventory Tracker', 'Variant Inventory Qty',
'Variant Inventory Policy', 'Variant Fulfillment Service', 'Variant Price',
'Variant Compare At Price', 'Variant Requires Shipping', 'Variant Taxable',
'Variant Barcode', 'Image Src', 'Image Position', 'Image Alt Text', 'Gift Card',
'SEO Title', 'SEO Description', 'Variant Image', 'Status', 'Collection']
self.file = None
self.writer = None
self.browser = None
self.tab = None
self.cnt = 0
def simulated_smooth_scroll(self, driver, step=1000, interval=0.5, timeout=30):
# 平滑移动到底部
start_time = time.time()
last_height = driver.execute_script("return document.documentElement.scrollHeight")
current_position = 0
while time.time() - start_time < timeout:
# 计算剩余滚动距离
remaining = last_height - current_position
# 动态调整步长
current_step = min(step, remaining) if remaining > 0 else 0
if current_step <= 0:
break
# 执行分步滚动
driver.execute_script(f"window.scrollBy(0, {current_step})")
current_position += current_step
# 等待滚动和内容加载
time.sleep(interval * (current_step / step)) # 动态间隔
# 检查新高度
new_height = driver.execute_script(
"return document.documentElement.scrollHeight"
)
# 更新高度（处理动态加载）
if new_height > last_height:
last_height = new_height
def get_driver(self, url, xpath_txt=None, is_turn=False):
options = webdriver.ChromeOptions()
options.add_argument('--headless')
options.add_argument('--disable-gpu')
options.page_load_strategy = "none"
driver = webdriver.Chrome(options=options)
driver.implicitly_wait(10)
driver.maximize_window()
while True:
try:
print('正在获取', url, '的页面数据')
driver.get(url)
if is_turn:
sleep(1)
self.simulated_smooth_scroll(driver)
if xpath_txt:
driver.find_element('xpath', xpath_txt)
else:
self.random_sleep(5)
break
except:
print(url, '没定位到，重新请求...')
# self.writer_to_file(driver.page_source, 'w', 'utf-8')
return driver
def driver_continue(self, driver, url, xpath_txt=None, is_turn=False):
flag = True
while flag:
flag = False
try:
print('正在获取', url, '的页面数据')
driver.get(url)
if is_turn:
self.random_sleep()
self.simulated_smooth_scroll(driver)
driver.find_element('xpath', xpath_txt)
except:
flag = True
print(url, '没定位到，重新请求...')
# self.writer_to_file(driver.page_source, 'w', 'utf-8')
def get_page_html(self, url, xpath_txt=None, is_turn=False):
driver = self.get_driver(url, xpath_txt, is_turn=is_turn)
page_source = driver.page_source
driver.close()
return etree.HTML(page_source)
def writer_to_file(self, data, mode, encoding=None):
if 'b' in encoding:
open('./text.html', mode).write(data)
else:
open('./text.html', mode, encoding=encoding).write(data)
print('写入文件成功！')
def driver_click(self, driver, timeout=2):
driver.click()
self.random_sleep(timeout)
def driver_back(self, driver, timeout=2):
driver.back()
self.random_sleep(timeout)
def driver_refresh(self, driver, timeout=2):
driver.refresh()
self.random_sleep(timeout)
def tab_wait(self, tab, timeout=3):
tab.wait(timeout)
return tab
def tab_get(self, tab, url, xpath_txt='html', is_turn=False):
print('正在获取', url, '的数据')
while True:
tab.get(url)
self.tab_wait(tab)
if is_turn:
tab.scroll.to_bottom()
self.tab_wait(tab)
ele = tab.ele(f'x:{xpath_txt}')
if ele:
break
print('没有请求到元素，重新请求中')
tab.wait(2)
return tab
def get_dp_html(self, tab, url, xpath_txt='html', is_turn=False):
tab = self.tab_get(tab, url, xpath_txt=xpath_txt, is_turn=is_turn)
res = etree.HTML(tab.html)
return res
def random_sleep(self, timeout=2):
sleep(random.random() + timeout)
def save_csv(self, data):
self.writer.writerow({
'Handle': data['Handle'],
'Title': data['Title'],
'Body (HTML)': data['Body (HTML)'],
'Vendor': data['Vendor'],
'Type': data['Type'],
'Tags': data['Tags'],
'Published': data['Published'],
'Option1 Name': data['Option1 Name'],
'Option1 Value': data['Option1 Value'],
'Option2 Name': data['Option2 Name'],
'Option2 Value': data['Option2 Value'],
'Option3 Name': data['Option3 Name'],
'Option3 Value': data['Option3 Value'],
'Variant SKU': data['Variant SKU'],
'Variant Grams': data['Variant Grams'],
'Variant Inventory Tracker': data['Variant Inventory Tracker'],
'Variant Inventory Qty': data['Variant Inventory Qty'],
'Variant Inventory Policy': data['Variant Inventory Policy'],
'Variant Fulfillment Service': data['Variant Fulfillment Service'],
'Variant Price': data['Variant Price'],
'Variant Compare At Price': data['Variant Compare At Price'],
'Variant Requires Shipping': data['Variant Requires Shipping'],
'Variant Taxable': data['Variant Taxable'],
'Variant Barcode': data['Variant Barcode'],
'Image Src': data['Image Src'],
'Image Position': data['Image Position'],
'Image Alt Text': data['Image Alt Text'],
'Gift Card': data['Gift Card'],
'SEO Title': data['SEO Title'],
'SEO Description': data['SEO Description'],
'Variant Image': data['Variant Image'],
'Status': data['Status'],
'Collection': data['Collection']
})
def get_response(self, url):
print('正在获取', url, '的数据')
while True:
try:
response = requests.get(url, headers=self.headers)
break
except:
print('没有请求到，重新请求')
self.random_sleep()
self.random_sleep()
return response
def get_html(self, url):
response = self.get_response(url)
return etree.HTML(response.text)
def tab_run_js(self, tab, js_code):
while True:
try:
tab.run_js(js_code)
break
except Exception as e:
print('捕获tab_run_js方法的run_js:', e)
tab.wait(1)
def ele_click(self, tab, ele):
while True:
try:
tab.actions.click(ele)
break
except Exception as e:
print('捕获ele_click方法的actions.click:', e)
tab.wait(1)
tab.wait(2)
def dp_click_ad(self, tab, xpath_txt):
ad_ele = tab.ele(f'x:{xpath_txt}', timeout=2)
if ad_ele:
print('有广告:', ad_ele)
self.ele_click(tab, ad_ele)
self.tab_wait(tab, 1)
def infinite_scroll(self, tab):
product_url_list = []
w_cnt = 0
turn_cnt = 0
is_bottom = False
while True:
self.dp_click_ad(tab, '//button[@class="css-2vqtum"]')
self.dp_click_ad(tab, '//button[@class="css-71t821"]')
if is_bottom:
break
tab.scroll.to_bottom()
self.tab_wait(tab, 1)
self.tab_run_js(self.tab, 'window.scrollBy(0, -1000)')
turn_cnt += 1
print(f'翻页了{turn_cnt}次')
n_cnt = 0
self.tab_wait(tab, 5)
product_url_ele_list = tab.eles('x://div[@class="product-thumbnail plpv3 css-1gasjii"]//a[@class="css-avqw6d"]')
for product_url_ele in product_url_ele_list:
product_url = product_url_ele.attr('href')
if 'https:' not in product_url:
product_url = 'https://www.coach.com' + product_url
if product_url in product_url_list:
continue
n_cnt += 1
product_url_list.append(product_url)
print('产品个数：', len(product_url_ele_list))
print('获取到的产品个数', len(product_url_list))
if n_cnt == 0:
w_cnt += 1
else:
w_cnt = 0
if w_cnt >= 5:
print('到底了')
is_bottom = True
return product_url_list
def get_detail_tab(self, url, xpath_txt='html', backup_xpath_txt='html'):
print('正在获取', url, '的数据')
tab = self.browser.latest_tab
while True:
tab.get(url)
self.tab_wait(tab)
ele = tab.ele(f'x:{xpath_txt}')
if ele:
print('第1个xpath找到的')
break
else:
t_ele = tab.ele(f'x:{backup_xpath_txt}')
if t_ele:
print('第2个xpath找到的')
break
print('没有请求到元素，重新请求中')
tab.wait(2)
return tab
def get_product_img_url_list(self, tab, data):
tab.wait(5)
self.dp_click_ad(tab, '//button[@class="css-2vqtum"]')
data['Option1 Value'] = tab.ele('x://p[contains(@class, "color-name")]').text
product_color_img_url_list = []
img_temp_ele = tab.ele('x://li[contains(@class, "is-prev")]')
if img_temp_ele:
product_color_img_url_cnt = int(img_temp_ele.attr('data-slide-index'))
else:
product_color_img_url_cnt = len(tab.eles('x://div[@class="css-1161qt5"]/div')) - 1
print(data['Option1 Value'], '色的图片个数应为：', product_color_img_url_cnt)
tab.wait(6)
raw_product_color_img_ele_list = tab.eles('x://li[contains(@class, "splide__slide")]/div | //li[contains(@class, "splide__slide")]//video/source | //div[@class="css-1161qt5"]/div[@class="css-113el1s"]/div')
for raw_product_color_img_ele in raw_product_color_img_ele_list:
raw_product_color_img_url = raw_product_color_img_ele.attr('src')
if not raw_product_color_img_url:
continue
raw_product_color_img_url = raw_product_color_img_url.split('?')[0]
if raw_product_color_img_url in product_color_img_url_list:
continue
product_color_img_url_list.append(raw_product_color_img_url)
print(data['Option1 Value'], '色获取到的图片个数为：', len(product_color_img_url_list))
return product_color_img_url_list
def init_tab(self):
co = ChromiumOptions()
co.set_browser_path(r'C:\Program Files\Google\Chrome\Application\chrome.exe')
co.auto_port()
# co.headless()
# co.set_argument('--no-sandbox')
self.browser = Chromium(co)
self.tab = self.browser.latest_tab
self.tab.set.window.max()
def handle_img_url(self, raw_img_url):
return raw_img_url
def save_all_data(self, product_data_list):
for product_data in product_data_list:
self.cnt += 1
self.save_csv(product_data)
print('第', self.cnt, '行保存完成！')
# 取消翻译
def cancel_translate(self, url):
tab = self.tab_get(self.tab, url, '//div[@class="collection_item overflow-hidden"]')
tab.actions.move_to(tab.ele('x://div[@class="gt-translate-btn-bg"]'))
tab.wait(2)
tab.actions.click(tab.ele('x://div[@class="gt-translate-switch"]'))
tab.wait(2)
def test_product(self, url, data_type):
self.file = open('./test.csv', 'w', newline='', encoding='utf-8-sig')
self.writer = csv.DictWriter(self.file, fieldnames=self.field_names)
self.writer.writeheader()
logging.captureWarnings(True)
self.init_tab()
data = deepcopy(self.init_data)
data['Type'] = data_type
data['Collection'] = data['Type']
product_data_list = self.product_detail_parse(url, data)
self.save_all_data(product_data_list)
if self.file:
self.file.close()
if self.browser:
self.browser.quit()
def product_detail_parse(self, url, data):
product_data_list = []
tab = self.get_detail_tab(url, '//span[@class="addLineHeight"] | //div[contains(@class, "genius-rechargeable-ET-container")]', '//div[contains(@class, "productView-thumbnails")]//div[@class="slick-track"]/div/a | //div[@class="productView-thumbnails"]/div/a') # 第2个请求
# ----------------------------------------------
# 获取 颜色链接 和 颜色文字
product_color_url_list = []
product_color_txt_list = []
t_ele = tab.ele('x://div[@class="color-swatches"]//label[contains(@class, "form-label")]')
if t_ele:
data['Option1 Name'] = t_ele.text.strip().replace('s:', '')
temp_color_ele_list = tab.eles('x://div[@class="color-swatches"]//label[contains(@class, "form-option")]/a')
for temp_color_ele in temp_color_ele_list:
product_color_txt_list.append(temp_color_ele.ele('x:./img').attr('alt'))
product_color_url = temp_color_ele.attr('href')
if not product_color_url:
product_color_url_list.append(url)
continue
if 'https:' not in product_color_url:
product_color_url = 'https://oralb.com' + product_color_url
product_color_url_list.append(product_color_url)
else:
product_color_url_list.append(url)
product_color_txt_list.append('')
# ----------------------------------------------
data['Handle'] = tab.ele('x://h1/a').text.strip().split(',')[0].replace(' ', '-').lower()
temp_id = 0
for product_color_url in product_color_url_list:
tab = self.get_detail_tab(product_color_url, '//span[@class="addLineHeight"] | //div[contains(@class, "genius-rechargeable-ET-container")]', '//div[contains(@class, "productView-thumbnails")]//div[@class="slick-track"]/div/a | //div[@class="productView-thumbnails"]/div/a') # 第3个请求
_data = deepcopy(data)
body_ele_list = tab.eles('x://span[@class="addLineHeight"] | //div[contains(@class, "genius-rechargeable-ET-container")]')
for body_ele in body_ele_list:
_data['Body (HTML)'] += body_ele.html # 到时候看看效果，div标签应该会自己换行吧？会，但是不会跟随颜色换
try:
_data['Variant Price'] = tab.ele('x://div[@class="productView-price"]//span[contains(@class, "price--withoutTax")]').text.replace('$', '').strip()
try:
_data['Variant Compare At Price'] = tab.ele('x://div[@class="productView-price"]//span[contains(@class, "price--rrp")]').text.replace('$', '').strip()
except:
_data['Variant Compare At Price'] = ''
except:
_data['Variant Price'] = '9.9'
_data['Title'] = tab.ele('x://h1/a').text.strip() # 颜色标题不一样，到时候看看是什么效果？不会跟随颜色换
if ',' in _data['Title']:
t_color = _data['Title'].split(',')[-1].strip()
try:
int(t_color.strip().split(' ')[0])
_data['Option1 Value'] = product_color_txt_list[temp_id]
except:
_data['Option1 Value'] = t_color
_data['Variant SKU'] = product_color_url.split('/')[-1] if product_color_url.split('/')[-1] else product_color_url.split('/')[-2]
# ----------------------------------------------
# 获取图片
temp_img_url_ele_list = tab.eles('x://div[contains(@class, "productView-thumbnails")]//div[@class="slick-track"]/div/a | //div[@class="productView-thumbnails"]/div/a')
product_img_url_list = []
for temp_img_url_ele in temp_img_url_ele_list:
temp_img_url = temp_img_url_ele.attr('href').strip()
if not temp_img_url or temp_img_url in product_img_url_list:
continue
product_img_url_list.append(temp_img_url)
product_img_url_len = len(product_img_url_list)
# ----------------------------------------------
_data['Image Src'] = self.handle_img_url(product_img_url_list[0])
_data['Image Position'] = 1
_data['Variant Image'] = _data['Image Src']
print('第', self.id, '个产品的第', temp_id + 1, '个sku', '颜色个数为：', len(product_color_url_list), '图片个数为：', product_img_url_len)
product_data_list.append(_data)
print(_data)
for i in range(1, product_img_url_len):
temp_data = deepcopy(self.empty_data)
temp_data['Handle'] = _data['Handle']
temp_data['Published'] = 'TRUE'
temp_data['Image Src'] = self.handle_img_url(product_img_url_list[i])
temp_data['Image Position'] = i + 1
product_data_list.append(temp_data)
print(temp_data)
temp_id += 1
return product_data_list
def parse(self, url):
data = deepcopy(self.init_data)
data['Type'] = 'Shop All'
data['Collection'] = data['Type']
page = 0
while True:
tab = self.tab_get(self.tab, url, '//li[@class="product"]/article/figure/a') # 第1个请求
for i in range(page):
self.tab.scroll.to_bottom()
self.tab_wait(tab, 1)
self.tab_run_js(tab, 'window.scrollBy(0, -1000)')
self.tab_wait(tab, 1)
self.ele_click(tab, tab.ele('x://ul[@class="pagination-list"]/li[contains(@class, "next")]/a'))
print('翻页到第', page + 1, '页')
index_html = etree.HTML(tab.html)
product_url_list = index_html.xpath('//li[@class="product"]/article/figure/a/@href')
product_title_list = index_html.xpath('//li[@class="product"]/article//div[@class="card-title"]/a/text()')
product_page_i = 0
for product_url in product_url_list:
# ---------------------------------------
# 标题去重
with open('./title_filter.txt', 'r', encoding='utf-8') as f:
title_filter = f.read()
if product_title_list[product_page_i].split(',')[0] in title_filter:
print(product_title_list[product_page_i].split(',')[0], '之前采集过')
product_page_i += 1
continue
with open('./title_filter.txt', 'a', encoding='utf-8') as f:
f.write(product_title_list[product_page_i].split(',')[0])
# ---------------------------------------
self.id += 1
with open('./filter.txt', 'r', encoding='utf-8') as f:
filter_txt = f.read()
if product_url in filter_txt:
print(self.id, '已完成')
continue
print(self.id, '开始')
product_data_list = self.product_detail_parse(product_url, data)
self.save_all_data(product_data_list)
print(self.id, '结束')
with open('./filter.txt', 'a', encoding='utf-8') as f:
f.write(product_url + '\n')
product_page_i += 1
is_next = index_html.xpath('//ul[@class="pagination-list"]/li[contains(@class, "next")]/a/@href')
if not is_next:
break
page += 1
def run(self, is_continue=False):
if is_continue:
self.file = open('./oralb.csv', 'a', newline='', encoding='utf-8-sig')
self.writer = csv.DictWriter(self.file, fieldnames=self.field_names)
else:
self.file = open('./oralb.csv', 'w', newline='', encoding='utf-8-sig')
self.writer = csv.DictWriter(self.file, fieldnames=self.field_names)
self.writer.writeheader()
with open('./filter.txt', 'w', encoding='utf-8') as f:
f.write('')
with open('./title_filter.txt', 'w', encoding='utf-8') as f:
f.write('')
self.init_tab()
logging.captureWarnings(True)
self.parse(self.url)
if self.file:
self.file.close()
if self.browser:
self.browser.quit()
if __name__ == '__main__':
oralb = Oralb()
oralb.run()
url1 = 'https://oralb.com/en-us/products/oral-b-io-series-10-twin-pack-cosmic-black-cosmic-black/' # 正常的，有打折价格
url2 = 'https://oralb.com/en-us/products/electric-toothbrushes/genius-7500-rechargeable-electric-toothbrush-orchid-purple/' # 正常的，没打折价格，没有描述，主颜色不是第一个
url3 = 'https://oralb.com/en-us/products/toothbrushes/battery-toothbrushes/pro-health-battery-replacement-brush-heads/' # 图片只有一个，且没价格，没颜色
url4 = 'https://oralb.com/en-us/products/kids-products/Kids-Extra-Soft-Replacement-Brush-Heads-featuring-Disney-Prinesses-2-count' # 格式错误，颜色不在标题，主颜色不是第一个，没有描述
# oralb.test_product(url1, 'Shop All')
# https://oralb.com/en-us/products/shop-all
# 去重
# 没价格的填9.9
# 没颜色的就不填颜色字段
# title、body等很多字段是随着颜色变化，那几乎所有字段随着颜色变
# 测一下详情是否正确
# 上传代码的时候把True给去掉 √
# 试试全部产品数据 √
# 试试部分产品数据（搜索“到时候”） √
# 重新采集 √
# 试试那几个产品详情 √
# 看笔记有没有漏掉的，并且检查下xpath在那几个网站有没有用 √
# 获取不到所有的产品，得一个一个翻页 √
# 让单个产品也能保存数据 √
# 图片要随着颜色变 √