使用Python进行网页抓取[关闭]

I’d like to grab daily sunrise/sunset times from a web site. Is it possible to scrape web content with Python? what are the modules used? Is there any tutorial available?

Use urllib2 in combination with the brilliant BeautifulSoup library:

import urllib2
from BeautifulSoup import BeautifulSoup
# or if you're using BeautifulSoup4:
# from bs4 import BeautifulSoup

soup = BeautifulSoup(urllib2.urlopen('http://example.com').read())

for row in soup('table', {'class': 'spad'})[0].tbody('tr'):
    tds = row('td')
    print tds[0].string, tds[1].string
    # will print date and sunrise

I’d really recommend Scrapy.

Quote from a deleted answer:

• Scrapy crawling is fastest than mechanize because uses asynchronous operations (on top of Twisted).
• Scrapy has better and fastest support for parsing (x)html on top of libxml2.
• Scrapy is a mature framework with full unicode, handles redirections, gzipped responses, odd encodings, integrated http cache, etc.
• Once you are into Scrapy, you can write a spider in less than 5 minutes that download images, creates thumbnails and export the extracted data directly to csv or json.

I collected together scripts from my web scraping work into this bit-bucket library.

Example script for your case:

from webscraping import download, xpath
D = download.Download()

html = D.get('http://example.com')
for row in xpath.search(html, '//table[@class="spad"]/tbody/tr'):
    cols = xpath.search(row, '/td')
    print 'Sunrise: %s, Sunset: %s' % (cols[1], cols[2])

Output:

Sunrise: 08:39, Sunset: 16:08
Sunrise: 08:39, Sunset: 16:09
Sunrise: 08:39, Sunset: 16:10
Sunrise: 08:40, Sunset: 16:10
Sunrise: 08:40, Sunset: 16:11
Sunrise: 08:40, Sunset: 16:12
Sunrise: 08:40, Sunset: 16:13

I would strongly suggest checking out pyquery. It uses jquery-like (aka css-like) syntax which makes things really easy for those coming from that background.

For your case, it would be something like:

from pyquery import *

html = PyQuery(url='http://www.example.com/')
trs = html('table.spad tbody tr')

for tr in trs:
  tds = tr.getchildren()
  print tds[1].text, tds[2].text

Output:

5:16 AM 9:28 PM
5:15 AM 9:30 PM
5:13 AM 9:31 PM
5:12 AM 9:33 PM
5:11 AM 9:34 PM
5:10 AM 9:35 PM
5:09 AM 9:37 PM

You can use urllib2 to make the HTTP requests, and then you’ll have web content.

You can get it like this:

import urllib2
response = urllib2.urlopen('http://example.com')
html = response.read()

Beautiful Soup is a python HTML parser that is supposed to be good for screen scraping.

In particular, here is their tutorial on parsing an HTML document.

Good luck!

I use a combination of Scrapemark (finding urls – py2) and httlib2 (downloading images – py2+3). The scrapemark.py has 500 lines of code, but uses regular expressions, so it may be not so fast, did not test.

Example for scraping your website:

import sys
from pprint import pprint
from scrapemark import scrape

pprint(scrape("""
    <table class="spad">
        <tbody>
            {*
                <tr>
                    <td>{{[].day}}</td>
                    <td>{{[].sunrise}}</td>
                    <td>{{[].sunset}}</td>
                    {# ... #}
                </tr>
            *}
        </tbody>
    </table>
""", url=sys.argv[1] ))

Usage:

python2 sunscraper.py http://www.example.com/

Result:

[{'day': u'1. Dez 2012', 'sunrise': u'08:18', 'sunset': u'16:10'},
 {'day': u'2. Dez 2012', 'sunrise': u'08:19', 'sunset': u'16:10'},
 {'day': u'3. Dez 2012', 'sunrise': u'08:21', 'sunset': u'16:09'},
 {'day': u'4. Dez 2012', 'sunrise': u'08:22', 'sunset': u'16:09'},
 {'day': u'5. Dez 2012', 'sunrise': u'08:23', 'sunset': u'16:08'},
 {'day': u'6. Dez 2012', 'sunrise': u'08:25', 'sunset': u'16:08'},
 {'day': u'7. Dez 2012', 'sunrise': u'08:26', 'sunset': u'16:07'}]

Make your life easier by using CSS Selectors

I know I have come late to party but I have a nice suggestion for you.

Using BeautifulSoup is already been suggested I would rather prefer using CSS Selectors to scrape data inside HTML

import urllib2
from bs4 import BeautifulSoup

main_url = "http://www.example.com"

main_page_html  = tryAgain(main_url)
main_page_soup = BeautifulSoup(main_page_html)

# Scrape all TDs from TRs inside Table
for tr in main_page_soup.select("table.class_of_table"):
   for td in tr.select("td#id"):
       print(td.text)
       # For acnhors inside TD
       print(td.select("a")[0].text)
       # Value of Href attribute
       print(td.select("a")[0]["href"])

# This is method that scrape URL and if it doesnt get scraped, waits for 20 seconds and then tries again. (I use it because my internet connection sometimes get disconnects)
def tryAgain(passed_url):
    try:
        page  = requests.get(passed_url,headers = random.choice(header), timeout = timeout_time).text
        return page
    except Exception:
        while 1:
            print("Trying again the URL:")
            print(passed_url)
            try:
                page  = requests.get(passed_url,headers = random.choice(header), timeout = timeout_time).text
                print("-------------------------------------")
                print("---- URL was successfully scraped ---")
                print("-------------------------------------")
                return page
            except Exception:
                time.sleep(20)
                continue 

If we think of getting name of items from any specific category then we can do that by specifying the class name of that category using css selector:

import requests ; from bs4 import BeautifulSoup

soup = BeautifulSoup(requests.get('https://www.flipkart.com/').text, "lxml")
for link in soup.select('div._2kSfQ4'):
    print(link.text)

This is the partial search results:

Puma, USPA, Adidas & moreUp to 70% OffMen's Shoes
Shirts, T-Shirts...Under ₹599For Men
Nike, UCB, Adidas & moreUnder ₹999Men's Sandals, Slippers
Philips & moreStarting ₹99LED Bulbs & Emergency Lights

Here is a simple web crawler, i used BeautifulSoup and we will search for all the links(anchors) who’s class name is _3NFO0d. I used Flipkar.com, it is an online retailing store.

import requests
from bs4 import BeautifulSoup
def crawl_flipkart():
    url = 'https://www.flipkart.com/'
    source_code = requests.get(url)
    plain_text = source_code.text
    soup = BeautifulSoup(plain_text, "lxml")
    for link in soup.findAll('a', {'class': '_3NFO0d'}):
        href = link.get('href')
        print(href)

crawl_flipkart()

Python has good options to scrape the web. The best one with a framework is scrapy. It can be a little tricky for beginners, so here is a little help.
1. Install python above 3.5 (lower ones till 2.7 will work).
2. Create a environment in conda ( I did this).
3. Install scrapy at a location and run in from there.
4. Scrapy shell will give you an interactive interface to test you code.
5. Scrapy startproject projectname will create a framework.
6. Scrapy genspider spidername will create a spider. You can create as many spiders as you want. While doing this make sure you are inside the project directory.

The easier one is to use requests and beautiful soup. Before starting give one hour of time to go through the documentation, it will solve most of your doubts. BS4 offer wide range of parsers that you can opt for. Use user-agent and sleep to make scraping easier. BS4 returns a bs.tag so use variable[0]. If there is js running, you wont be able to scrape using requests and bs4 directly. You could get the api link then parse the JSON to get the information you need or try selenium.