Python：需要从log-fi获取唯一错误

Question

到目前为止我所拥有的

def unique_ips():
f = open('logfile','r')
ips = set()
for line in f:
    ip = line.split()[0]
    print ip
    for date in ip:
       logdate = line.split()[3]
       print "\t", logdate
       for entry in logdate:
           info = line.split()[5:11] 
           print "\t\t", info
    ips.add(ip)
unique_ips()

我遇到的问题是：

^{pr2}$

我有一个日志文件，我必须先按ip排序，然后按时间排序，然后按错误排序

应该看起来像：

199.21.99.83
        [30/Jun/2013:07:18:30
                ['"GET', '/searchme/index.php?f=man_soweth', 'HTTP/1.1"', '200', '8676', '"-"']

但我得到的却是：

199.21.99.83
        [30/Jun/2013:07:18:30
                ['"GET', '/searchme/index.php?f=man_soweth', 'HTTP/1.1"', '200', '8676', '"-"']
                ['"GET', '/searchme/index.php?f=man_soweth', 'HTTP/1.1"', '200', '8676', '"-"']
                ['"GET', '/searchme/index.php?f=man_soweth', 'HTTP/1.1"', '200', '8676', '"-"']
                ['"GET', '/searchme/index.php?f=man_soweth', 'HTTP/1.1"', '200', '8676', '"-"']
                 ...

我确信我遇到了某种语法问题，但希望能得到帮助！在

日志文件如下：

99.21.99.83 - - [30/Jun/2013:07:15:50 -0500] "GET /lenny/index.php?f=13 HTTP/1.1" 200 11244 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
199.21.99.83 - - [30/Jun/2013:07:16:13 -0500] "GET /searchme/index.php?f=being_fruitful HTTP/1.1" 200 7526 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
199.21.99.83 - - [30/Jun/2013:07:16:45 -0500] "GET /searchme/index.php?f=comparing_themselves HTTP/1.1" 200 7369 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
66.249.73.40 - - [30/Jun/2013:07:16:56 -0500] "GET /espanol/displayAncient.cgi?ref=isa%2054:3 HTTP/1.1" 500 167 "-" "Mozilla/5.0 (compatible; Googlebot/2.1; +http://www.google.com/bot.html)"
199.21.99.83 - - [30/Jun/2013:07:17:00 -0500] "GET /searchme/index.php?f=tribulation HTTP/1.1" 200 7060 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
199.21.99.83 - - [30/Jun/2013:07:17:15 -0500] "GET /searchme/index.php?f=proud HTTP/1.1" 200 7080 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
199.21.99.83 - - [30/Jun/2013:07:17:34 -0500] "GET /searchme/index.php?f=soul HTTP/1.1" 200 7063 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.com/bots)"
199.21.99.83 - - [30/Jun/2013:07:17:38 -0500] "GET /searchme/index.php?f=the_flesh_lusteth HTTP/1.1" 200 6951 "-" "Mozilla/5.0 (compatible; YandexBot/3.0; +http://yandex.c

Answer 1

由于示例输出，这个问题有点令人困惑，但我很确定您想要这样的结果：

def unique_ips():
    f = open('logfile','r')
    ips = {}
    # This for loop collects all of the ips with their associated errors
    for line in f:
        ip = line.split()[0]
        try:
            ips[ip].append(line)
        except KeyError:
            ips[ip] = [line]

    # This for loop goes through all the ips that were collected
    # and prints out all errors for those ips
    for ip, errors in ips.iteritems():
        print ip
        errors.sort()
        for e in errors:
           logdate = e.split()[3]
           print "\t", logdate

           info = e.split()[5:11] 
           print "\t\t", info

    f.close()

从示例文件中生成以下输出：

^{pr2}$

Answer 2

你的循环太多了。在logdate循环中不需要for条目。你已经在每一行上循环了。在

删除logdate中的for条目，升级info赋值和print语句。在

（评论已经提到了这一点。）