🎉 https://www.llbean.com/llb/shop/33381?page=boat-and-tote-bag-open-top&bc=1000010392&csp=a&attrValue_0=38875&pos=1&feat=hp-F7t33381
import random, csv, requests, logging, time, json, re, execjs
from selenium import webdriver
from copy import deepcopy
from time import sleep, time
from lxml import etree
from DrissionPage import Chromium, ChromiumOptions
class Llbean:
def __init__(self):
self.url = 'https://www.llbean.com/llb/shop/33381?page=boat-and-tote-bag-open-top&bc=1000010392&csp=a&attrValue_0=38875&pos=1&feat=hp-F7t33381'
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
self.total_product_num = 0
self.total_page = 0
self.sku_id = 0
self.row_path = []
self.col_path = []
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', backup_xpath_txt='html', is_turn=False, timeout=3, off_xpath='//h1', off_txt='Oops, Something Went Wrong.', wait_minute=15):
while True:
print('正在获取', url, '的数据')
tab.get(url)
self.tab_wait(tab, timeout)
if off_xpath and off_txt:
try:
t_ele = tab.ele(f'x:{off_xpath}')
except Exception as e:
print('off_xpath这里错了')
print(e)
tab.wait(timeout)
continue
if t_ele and off_txt in t_ele.text:
print(f'ip被封了，等待{wait_minute}分钟再尝试请求')
sleep(60 * wait_minute)
continue
if is_turn:
tab.scroll.to_bottom()
self.tab_wait(tab, timeout)
ele = tab.ele(f'x:{xpath_txt}')
if ele:
print('第1个xpath找到的')
break
else:
if backup_xpath_txt != 'html':
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(timeout)
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)
sleep(2)
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, timeout=2):
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(timeout)
tab.wait(timeout)
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"]')
self.dp_click_ad(tab, '//button[@class="css-71t821"]')
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()
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 click_accept(self, tab=None, url=None, xpath_txt='html'):
if not tab and url:
tab = self.get_detail_tab(url, xpath_txt=xpath_txt)
temp_ele = tab.ele('x://div[contains(@class, "policy_acceptBtn")]')
if temp_ele:
self.ele_click(tab, temp_ele)
def check_size(self, product_size_list):
size_filter = []
for product_size in product_size_list:
if product_size in size_filter:
return False
size_filter.append(product_size)
return True
def test_product(self, url, file_name='test', _type='One'):
self.file = open(f'./{file_name}.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()
self.id += 1
print(self.id, '开始')
product_data_list = self.product_detail_parse(url, _type)
self.save_all_data(product_data_list)
print(self.id, '结束')
if self.file:
self.file.close()
if self.browser:
self.browser.quit()
def is_have_dot_line(self, tab, row, col):
print('is_have_dot_line', row, col)
return 'hmf-option-unavailable' in tab.ele(f'x://pdp-attributes-components-attributes[@class="hmf-mx-s hmf-mx-m-0"]/section/div[{str(row)}]//div[contains(@class, "hmf-flex-wrap")]/div[{str(col)}]/button[contains(@class, "hmf-selectable-option")]').attr('class')
def get_product_color_url(self, tab, url, row, col):
color_txt = tab.ele(f'x://pdp-attributes-components-attributes[@class="hmf-mx-s hmf-mx-m-0"]/section/div[{str(row)}]//div[contains(@class, "sliding-row-inner")]/div[{str(col)}]//button[contains(@class, "pdp-color-swatch-selected")]/img').attr('alt').strip().replace(' ', '/')
return url.split('?')[0] + '?color=' + color_txt
def dfs_click(self, tab, timeout=2):
for click_id in range(len(self.row_path)):
row = self.row_path[click_id]
col = self.col_path[click_id]
print('dfs_click：', row, col)
t_ele = tab.eles(f'x://ul[contains(@role, "radiogroup")]')[row - 1].ele(f'x:./li[{col}]')
self.ele_click(tab, t_ele, timeout)
self.random_sleep(3)
def get_img_url(self, tab):
product_img_url_list = []
product_img_url_ele_list = tab.eles('x://div[contains(@class, "MultiCarousel_vertical-wrapper")]//div[contains(@class, "Card")]/button/img')
for product_img_url_ele in product_img_url_ele_list:
product_img_url = product_img_url_ele.attr('src')
product_img_url = product_img_url.split('?')[0] + '?hei=764&wid=665&resMode=sharp2&defaultImage=llbprod/' + product_img_url.split('?')[-1].split('/')[-1]
product_img_url_list.append(product_img_url)
return product_img_url_list
def insert_data(self, data_list, img_url_list, data, tab, url, name_list, total_row, col_list):
_data = deepcopy(data)
self.sku_id += 1
img_url_len = len(img_url_list)
_data['Title'] = self.tab.ele('x://h1[contains(@class, "NewProductDetailHeader_product-name")]').text.strip()
_data['Handle'] = (_data['Type'] + '-' + _data['Title']).replace(' ', '-').lower()
_data['Body (HTML)'] = self.tab.ele('x://div[contains(@class, "NewProductInformation_copy-wrap")]/div[1]').html.replace('<svg class="icon AccordionButton_icon" xmlns="http://www.w3.org/2000/svg" fill="currentColor" preserveAspectRatio="xMidYMid meet" viewBox="0 0 19 32">', '').replace('</svg>', '').replace('<path transform="" d="M17.6 14.4q.96 0 .96 1.6t-.96 1.6h-6.72v6.72q0 .96-1.6.96t-1.6-.96V17.6H.96Q0 17.6 0 16t.96-1.6h6.72V7.68q0-.96 1.6-.96t1.6.96v6.72h6.72z"></path>', '')
_data['Variant SKU'] = _data['Handle']
for row in range(total_row):
_data[f'Option{str(row + 1)} Value'] = tab.eles('x://form[contains(@name, "pdpBuyboxForm")]//fieldset/legend//span[contains(@class, "name")] | //form[contains(@name, "pdpBuyboxForm")]//fieldset/legend[./span[contains(@aria-hidden, "true")]]/span[last()] | //form[contains(@name, "pdpBuyboxForm")]//fieldset/legend/span[./strong]/span[last()]')[row].text.strip()
_data['Variant SKU'] = data['Variant SKU'] + '-' + data[f'Option{str(row + 1)} Value']
_data['Variant SKU'] = (_data['Variant SKU'] + '-' + str(self.sku_id)).replace(' ', '-').lower()
_data['Variant Price'] = tab.ele('x://article[contains(@class, "ProductPage_product-buybox")]//span[contains(@itemprop, "price")]').text.replace(',', '').strip()
try:
_data['Image Src'] = img_url_list[0]
except:
print('这个产品没有图片')
return
_data['Image Position'] = 1
_data['Variant Image'] = _data['Image Src']
print(f'第{self.id}个产品 {url} 的第{self.sku_id}个sku，', end='')
for row in range(total_row):
print(f'字段{name_list[row]}的个数为：{col_list[row]},', end='')
print(f'图片个数为：{img_url_len}')
data_list.append(_data)
print(_data)
for i in range(1, img_url_len):
temp_data = deepcopy(self.empty_data)
temp_data['Handle'] = _data['Handle']
temp_data['Published'] = 'TRUE'
temp_data['Image Src'] = img_url_list[i]
temp_data['Image Position'] = i + 1
data_list.append(temp_data)
print(temp_data)
def check_field_val_is_all_exist(self, tab):
pass
def dfs_insert_data(self, data_list, img_url_list, data, tab, url, name_list, total_row, col_list, row):
if row < 1:
return
for col in range(1, col_list[row - 1] + 1):
self.row_path.append(row)
self.col_path.append(col)
if 1 == row:
self.dfs_click(tab)
img_url_list = self.get_img_url(tab)
self.insert_data(data_list, img_url_list, data, tab, url, name_list, total_row, col_list)
self.dfs_insert_data(data_list, img_url_list, data, tab, url, name_list, total_row, col_list, row - 1)
self.row_path.pop()
self.col_path.pop()
def check_field_val_num(self, field_num, field_val_num_list):
for field_id in range(3, field_num):
if field_val_num_list[field_id] > 1:
return False
return True
def product_detail_parse(self, url, _type):
_type = _type.replace('-', ' ').title()
product_data_list = []
data = deepcopy(self.init_data)
data['Type'] = _type
data['Collection'] = data['Type']
tab = self.tab_get(self.tab, url, '//div[contains(@class, "NewProductInformation_copy-wrap")]/div[1]') # 第1个请求
self.dp_click_ad(tab, '//button[contains(@class, "Modal_modal-close")]')
# ------------------------------------
# 获取字段的名称 和 字段的数量 和 每个字段拥有的值的数量
field_name_list = []
field_val_num_list = []
field_name_ele_list = tab.eles('x://form[contains(@name, "pdpBuyboxForm")]//fieldset/legend//span[contains(@class, "title")] | //form[contains(@name, "pdpBuyboxForm")]//fieldset/legend//span/strong')
field_num = len(field_name_ele_list)
for field_i in range(field_num):
data[f'Option{str(field_i + 1)} Name'] = field_name_ele_list[field_i].text.replace(':', '').strip().split('$')[0]
field_name_list.append(data[f'Option{str(field_i + 1)} Name'])
field_val_ele_len = len(tab.eles(f'x://ul[contains(@role, "radiogroup")]')[field_i].eles('x:./li'))
field_val_num_list.append(field_val_ele_len)
# ------------------------------------
print(f'有 {field_num} 个字段')
print('字段的名称有：', field_name_list)
for field_i in range(field_num):
print(f'{field_name_list[field_i]}拥有的值的数量为：{field_val_num_list[field_i]}')
if field_num > 3 and (not self.check_field_val_num(field_num, field_val_num_list)):
return product_data_list
self.sku_id = 0
self.dfs_insert_data(product_data_list, [], data, tab, url, field_name_list, field_num, field_val_num_list, field_num)
return product_data_list
def parse(self):
pass
def multi_product_run(self, is_continue=False):
pass
def run(self, is_continue=False):
self.test_product(self.url, 'www_llbean_product')
if __name__ == '__main__':
llbean = Llbean()
llbean.run()
# https://www.llbean.com/llb/shop/33381?page=boat-and-tote-bag-open-top&bc=1000010392&csp=a&attrValue_0=38875&pos=1&feat=hp-F7t33381
# 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 √
# 改url
# 改文件名
# 改类型