The Hartley Clinic Clinic Contact: Ailing Jeavons 01256 856289 , , Fleet, RG27 8NZ Monday 8:30 — 17:00 Tuesday 8:30 — 19:00 Wednesday 8:30— 17:00 Thursday 8:30 — 17:00 Friday 8:30 — 15:00 Saturday 9:00 — 17:00 Sunday Closed https://www.saveface.co.uk/clinic/the-hartley-clinic/
The Hartley Clinic Clinic Contact: Ailing Jeavons 01256 856289 , , Fleet, RG27 8NZ Monday 8:30 — 17:00 Tuesday 8:30 — 19:00 Wednesday 8:30— 17:00 Thursday 8:30 — 17:00 Friday 8:30 — 15:00 Saturday 9:00 — 17:00 Sunday Closed https://www.saveface.co.uk/clinic/the-hartley-clinic/
The Hartley Clinic Clinic Contact: Ailing Jeavons 01256 856289 , , Fleet, RG27 8NZ Monday 8:30 — 17:00 Tuesday 8:30 — 19:00 Wednesday 8:30— 17:00 Thursday 8:30 — 17:00 Friday 8:30 — 15:00 Saturday 9:00 — 17:00 Sunday Closed https://www.saveface.co.uk/clinic/the-hartley-clinic/
The Hartley Clinic Clinic Contact: Ailing Jeavons 01256 856289 , , Fleet, RG27 8NZ Monday 8:30 — 17:00 Tuesday 8:30 — 19:00 Wednesday 8:30— 17:00 Thursday 8:30 — 17:00 Friday 8:30 — 15:00 Saturday 9:00 — 17:00 Sunday Closed https://www.saveface.co.uk/clinic/the-hartley-clinic/
The Hartley Clinic Clinic Contact: Ailing Jeavons 01256 856289 , , Fleet, RG27 8NZ Monday 8:30 — 17:00 Tuesday 8:30 — 19:00 Wednesday 8:30— 17:00 Thursday 8:30 — 17:00 Friday 8:30 — 15:00 Saturday 9:00 — 17:00 Sunday Closed https://www.saveface.co.uk/clinic/the-hartley-clinic/
The Hartley Clinic Clinic Contact: Ailing Jeavons 01256 856289 , , Fleet, RG27 8NZ Monday 8:30 — 17:00 Tuesday 8:30 — 19:00 Wednesday 8:30— 17:00 Thursday 8:30 — 17:00 Friday 8:30 — 15:00 Saturday 9:00 — 17:00 Sunday Closed https://www.saveface.co.uk/clinic/the-hartley-clinic/

... = blocks[block].find_element(...)

location = blocks[block].find_element(By.XPATH, "//li[@class='cls_loc']")

import requests
from bs4 import BeautifulSoup
url = 'https://www.saveface.co.uk/search/'
soup = BeautifulSoup(requests.get(url).text, "html.parser")
css_selector = {
    'company name' : ".title",
    'person name'  : ".name_wrapper",
    'phone number' : ".phone",
    'location'     : ".cls_loc",
    'opening hours': ".opening-hours",
    'profile link' : ".visitpage",
}
data = {key:[] for key in list(css_selector)+['description','website link']}
number_of_pages = int(str(soup).split('total_pages":')[1].split('}')[0])
for page in range(2,number_of_pages+2):
    
    blocks = soup.select('.clientresult')
    for idx,block in enumerate(blocks):
        print(f'blocks {idx+1}/{len(blocks)}',end='\r')
        for key in list(css_selector):
            try:
                if 'link' in key:
                    data[key] += [ block.select_one(css_selector[key])['href'] ]
                else:
                    data[key] += [ block.select_one(css_selector[key]).text.strip().replace('\r\n', ', ') ]
            except AttributeError:
                data[key] += ['*missing value']
    if page <= number_of pages:
        print('\nloading page', page)
        url_page = f'{url}?fwp_paged={page}&'
        soup = BeautifulSoup(requests.get(url_page).text, "html.parser")
print('\nno more pages to load, moving to scrape profile links...')
for idx,url in enumerate(data['profile link']):
    
    print(f"profile link {idx+1}/{len(data['profile link'])}  ",end='\r')
    soup_profile = BeautifulSoup(requests.get(url).text, "html.parser")
    try:
        data['description'] += [soup_profile.select_one('.clinicContent > .description').text.strip()]
    except AttributeError:
        data['description'] += ['*missing value']
    try:
        data['website link'] += [soup_profile.select_one('.visitwebsite')['href']]
    except AttributeError:
        data['website link'] += ['*missing value']
Output（执行完需要约8分钟）

然后，您可以轻松通过运行`pd.DataFrame(data)`来创建数据框。
<details>
<summary>英文:</summary>
To restric the search within a subtree rooted at the context node, your expression should start with `.//` so you have to replace `//` with `.//` in each of the commands
... = blocks[block].find_element(...)
The meaning of `//` is to search the document from the document&#39;s root, ignoring the context node `blocks[block]` altogether.
Moreover, notice that not all the blocks have a location as you can see from this image
[![enter image description here][1]][1]
in this case 
location = blocks[block].find_element(By.XPATH, &quot;//li[@class=&#39;cls_loc&#39;]&quot;)
will raise a `NoSuchElementException`. To avoid this you have to put the command in a `try...except...` block
# UPDATE
Scraping 400 blocks with selenium takes about 1 minute on my computer, I tried with BeautifulSoup and it just takes less than 1 second! The slow part is to scrape the profiles, because for each of them we have to download a new webpage, however is still way faster with BeautifulSoup.
So I write a script without using selenium, just BeautifulSoup (you can install by running `pip install beautifulsoup4` in the terminal)
import requests
from bs4 import BeautifulSoup
url = &#39;https://www.saveface.co.uk/search/&#39;
soup = BeautifulSoup(requests.get(url).text, &quot;html.parser&quot;)
css_selector = {
&#39;company name&#39; : &quot;.title&quot;,
&#39;person name&#39;  : &quot;.name_wrapper&quot;,
&#39;phone number&#39; : &quot;.phone&quot;,
&#39;location&#39;     : &quot;.cls_loc&quot;,
&#39;opening hours&#39;: &quot;.opening-hours&quot;,
&#39;profile link&#39; : &quot;.visitpage&quot;,
}
data = {key:[] for key in list(css_selector)+[&#39;description&#39;,&#39;website link&#39;]}
number_of_pages = int(str(soup).split(&#39;total_pages&quot;:&#39;)[1].split(&#39;}&#39;)[0])
for page in range(2,number_of_pages+2):
blocks = soup.select(&#39;.clientresult&#39;)
for idx,block in enumerate(blocks):
print(f&#39;blocks {idx+1}/{len(blocks)}&#39;,end=&#39;\r&#39;)
for key in list(css_selector):
try:
if &#39;link&#39; in key:
data[key] += [ block.select_one(css_selector[key])[&#39;href&#39;] ]
else:
data[key] += [ block.select_one(css_selector[key]).text.strip().replace(&#39;\r\n&#39;,&#39;, &#39;) ]
except AttributeError:
data[key] += [&#39;*missing value*&#39;]
if page &lt;= number_of_pages:
print(&#39;\nloading page&#39;, page)
url_page = f&#39;{url}?fwp_paged={page}&#39;
soup = BeautifulSoup(requests.get(url_page).text, &quot;html.parser&quot;)
print(&#39;\nno more pages to load, moving to scrape profile links...&#39;)
for idx,url in enumerate(data[&#39;profile link&#39;]):
print(f&quot;profile link {idx+1}/{len(data[&#39;profile link&#39;])}  &quot;,end=&#39;\r&#39;)
soup_profile = BeautifulSoup(requests.get(url).text, &quot;html.parser&quot;)
try:
data[&#39;description&#39;] += [soup_profile.select_one(&#39;.clinicContent &gt; .description&#39;).text.strip()]
except AttributeError:
data[&#39;description&#39;] += [&#39;*missing value*&#39;]
try:
data[&#39;website link&#39;] += [soup_profile.select_one(&#39;.visitwebsite&#39;)[&#39;href&#39;]]
except AttributeError:
data[&#39;website link&#39;] += [&#39;*missing value*&#39;]
Output (it took about 8 minutes to complete the execution)
blocks 400/400
loading page 2
blocks 109/109
no more pages to load, moving to scrape profile links...
profile link 509/509  
Then you can easily create the dataframe by running `pd.DataFrame(data)`
[![enter image description here][2]][2]
[1]: https://i.stack.imgur.com/RLjpLm.png
[2]: https://i.stack.imgur.com/vsGn0.png
</details>
# 答案2
**得分**: 0
这是新的代码
但为什么每一页都返回相同的输出：
<details>
<summary>英文:</summary>
this is the new code
but it returns the same output on every page why:

blocks = driver.find_elements(By.XPATH, ".//div[@class='result clientresult']")
for block in range(10):
try:    
company_name = blocks[block].find_element(By.XPATH, ".//h3[@class='resulttitle']").text.strip()
companies_names.append(company_name)
except:
companies_names.append("Not found on the site")
try:    
person_name = blocks[block].find_element(By.XPATH, ".//p[@class='name_wrapper']").text.strip()
persons_names.append(person_name)
except:
persons_names.append("Not found on the site")
try:    
phone_number = blocks[block].find_element(By.XPATH, ".//div[@class='searchContact phone']").text.strip()
phones_numbers.append(phone_number)
except:
phones_numbers.append("Not found on the site")
try:    
location = blocks[block].find_element(By.XPATH, ".//li[@class='cls_loc']").text.strip()
locations.append(location)
except:
locations.append("Not found on the site")
try:    
opening_hour = blocks[block].find_element(By.XPATH, ".//li[@class='opening-hours']").text.strip()
opening_hours.append(opening_hour)
except:
opening_hours.append("Not found on the site")
try:    
profile = blocks[block].find_element(By.XPATH, ".//a[@class='visitpage']").get_attribute("href")
all_profiles.append(profile)
except:
all_profiles.append("Not found on the site")
two_page = driver.find_element(By.XPATH, ".//a[@class='facetwp-page']")
two_page.click()

try:
driver.get(all_profiles[i])
driver.implicitly_wait(10)
try:    
description = driver.find_element(By.XPATH, ".//div[@class='desc-text-left']").text.strip()
descriptions.append(description)
except:
descriptions.append("Not found on the site")
try:    
website_link = driver.find_element(By.XPATH, ".//a[@class='visitwebsite website']").get_attribute("href")
websites_links.append(website_link)        
except:
websites_links.append("Not found on the site")
except:
descriptions.append("Not found on the site")
websites_links.append("Not found on the site")

</details>