webparse.py

来自「python web programming 部分」· Python 代码 · 共 27 行

import htmllib, urllib, formatter, sys

def Usage():
    print """
Usage: python webparse.py [-s] URL
"""

def parse(url, formatter):
    f = urllib.urlopen(url)
    data = f.read()
    f.close()
    p = htmllib.HTMLParser(formatter)
    p.feed(data)
    p.close()

if len(sys.argv) < 2:
    Usage()
else:
    if sys.argv[1] == "-s":
        fmt  = formatter.NullFormatter()
        del sys.argv[1]
    else:
        fmt = formatter.AbstractFormatter(
                formatter.DumbWriter(sys.stdout))
    parse(sys.argv[1], fmt)