使用正则表达式重新排序字符串

Question

我想将第一次出现的日期或一般的正则表达式带到我的文本开头：

示例： "I went out on 1 sep 2012 and it was better than 15 jan 2012" 我想得到 "1 sep 2012, I went out on and it was better than 15 jan 2012"

我正在考虑用",1 sep 2012," 替换"1 sep 2012"，然后从"," 中删除字符串，但我不知道要写什么来代替replace_with：

line = re.sub(r'\d+\s(?:jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\s\d{4}', 'replace_with', line, 1)

有什么帮助吗？

Answer 1

使用capture groups:

>>> import re
>>> s = "I went out on 1 sep 2012 and it was better than 15 jan 2012"
>>> r = re.compile('(^.*)(1 sep 2012 )(.*$)')
>>> r.sub(r'\2\1\3',s)
'1 sep 2012 I went out on and it was better than 15 jan 2012'

括号捕获部分字符串：

(^.*)          # Capture everything from the start of the string
(1 sep 2012 )  # Upto the part we are interested in (captured)
(.*$)          # Capture everything else

然后只需在替换 `\2\1\3' 中重新排序捕获组 注意： 以引用捕获组需要原始字符串 r'\2\1\3'。我的示例中的第二组只是文字字符串(1 sep 2012 )，但当然可以是任何正则表达式，例如您创建的正则表达式（末尾有一个额外的\s）：

(\d+\s(?:jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\s\d{4}\s)

>>> r = re.compile(r'(^.*)(\d+\s(?:aug|sep|oct|nov)\s\d{4}\s)(.*$)')
>>> r.sub(r'\2\1\3',s)
'1 sep 2012 I went out on and it was better than 15 jan 2012'

来自docs.python.org：

当存在 'r' 或 'R' 前缀时，反斜杠后面的字符将直接包含在字符串中。