How to reliably process web-data in Python

问题

I'm using the following code to get data from a website:

time_out = 4

def tryconnect(turl, timer=time_out, retries=10):
    urlopener = None
    sitefound = 1
    tried = 0
    while (sitefound != 0) and tried < retries:
        try:
            urlopener = urllib2.urlopen(turl, None, timer)
            sitefound = 0
        except urllib2.URLError:
            tried += 1
    if urlopener: return urlopener
    else: return None

[...]

urlopener = tryconnect('www.example.com')
if not urlopener:
    return None
try:
    for line in urlopener:
        do stuff
except httplib.IncompleteRead:
    print 'incomplete'
    return None
except socket.timeout:
    print 'socket'
    return None
return stuff

Is there a way I can handle all these exceptions without having so much boilerplate code everytime?

Thanks!

回答1:

You can avoid some boilerplate code in the first function too:

time_out = 4

def tryconnect(turl, timer=time_out, retries=10):
    for tried in xrange(retries):
        try:
            return urllib2.urlopen(turl, None, timer)
        except urllib2.URLError:
            pass
    return None

and in the second:

urlopener = tryconnect('www.example.com')
if urlopener:
    try:
        for line in urlopener:
            do stuff
    except (httplib.IncompleteRead, socket.timeout), e:
        print e
        return None
else:
    return None

来源：https://stackoverflow.com/questions/5390752/how-to-reliably-process-web-data-in-python