使用regex从特定的文本表单获取信息

Question

我有一个文本，包含如下内容：

(some text)
libncursesw5-dev:amd64 depends on libc6-dev | libc-dev;(some text)
libx32ncursesw5 depends on libc6-x32 (>= 2.16);(some text)
libx32ncurses5-dev depends on libncurses5-dev (= 5.9+20150516-2ubuntu1);(some text)
libx32ncursesw5-dev depends on libc6-dev-x32;(some text)
lib32tinfo-dev depends on lib32c-dev;(some text)

下面是其中一个句子的完整示例：

dpkg: error processing package lib32tinfo5 (--install):
 dependency problems - leaving unconfigured
dpkg: dependency problems prevent configuration of libncurses5-dev:amd64:
 libncurses5-dev:amd64 depends on libc6-dev | libc-dev; however:
    Package libc6-dev is not installed.
    Package libc-dev is not installed.

整篇课文被分成几个段落，例如上面的一段，每个段落包含其中一个句子。你知道吗

我想要一个在python中使用re-library的regex，使用findall选项可以得到如下结果：

('libc6-dev', '', 'libc-dev', '')
('libc6-x32','2.16')
('libncurses5-dev','5.9+20150516-2ubuntu1')
('libc6-dev-x32','')
('lib32c-dev','')

换句话说，我需要你的帮助，以便从这样的文本中获取一个元组，其中包含包及其版本（如果指定的话）。你知道吗

我做了这个正则表达式：

(?<=depends on )([a-zA-Z0-9\-]*)(?: \([=> ]*([a-zA-Z0-9-+.]*)(?:\)))?|(?: \| )([a-zA-Z0-9\-]*)(?: \([=> ]*([a-zA-Z0-9-+.]*)(?:\)))?(?=;)

我得到这个结果：

('libc6-dev', '', '', '')
('', '', 'libc-dev', '')
('libc6-x32', '2.16', '', '')
('libncurses5-dev', '5.9+20150516-2ubuntu1', '', '')
('libc6-dev-x32', '', '', '')
('lib32c-dev', '', '', '')

如你所见，对于这句话：

libncursesw5-dev:amd64 depends on libc6-dev | libc-dev;

我得到了这个答案：

('libc6-dev', '', '', '')
('', '', 'libc-dev', '')

而不是这个：

('libc6-dev', '', 'libc-dev', '')

谢谢你的帮助。你知道吗

Answer 1

#!/usr/bin/python2
# -*- coding: utf-8 -*-

import re

input = """(some text)
libncursesw5-dev:amd64 depends on libc6-dev | libc-dev;(some text)
libx32ncursesw5 depends on libc6-x32 (>= 2.16);(some text)
libx32ncurses5-dev depends on libncurses5-dev (= 5.9+20150516-2ubuntu1);(some text)
libx32ncursesw5-dev depends on libc6-dev-x32;(some text)
lib32tinfo-dev depends on lib32c-dev;(some text)"""

#a = []
#m = re.findall("depends on ([^\s;]+)\ \|\ ([^\s;]+)", input) # 1
#a = a + m
#m = re.findall("depends on ([^\s;]+)\ \([><=]{,2} ([^;]+)\)", input) # 2, 3
#a = a + m
#m = re.findall("depends on ([^\s;]+)", input) # 4, 5
#a = a + m

m = re.findall("depends on ([^\s;]+)\ \|\ ([^\s;]+)|depends on ([^\s;]+)\ \([><=]{,2} ([^;]+)\)|depends on ([^\s;]+)", input)

print m

输出：

[
    ('libc6-dev', 'libc-dev', '', '', ''),
    ('', '', 'libc6-x32', '2.16', ''),
    ('', '', 'libncurses5-dev', '5.9+20150516-2ubuntu1', ''),
    ('', '', '', '', 'libc6-dev-x32'),
    ('', '', '', '', 'lib32c-dev')
]

你可以一个接一个或者全部一起拿到，我不知道这是否能帮到你