正则表达式匹配的所有引用的字符串

Question

我试图做一个正则表达式，如果可能的话，这将匹配从文本的所有引用的字符串。 一个例子：

ABC released its full midseason schedule today, and it features premiere dates for several new shows, along with one rather surprising timeslot change.</p><p>First of all, ABC's previously reported plans for dramas 'Once Upon A Time,' 'Revenge,' 'Grey's Anatomy,' and 'Scandal' haven't changed. 

我想有结果：

's previously reported plans for dramas ' (not useful but i can manage it) 
'Once Upon A Time,' 
' ' 
'Revenge,' 
' 'Grey' 
'Grey's Anatomy,' 
etc 

，所以我会basicly需要两次，每次报价相匹配。如果我使用标准的正则表达式，我不会有'从前'和'灰色的解剖学'，原因很明显。

感谢您的任何建议！

Answer 1

下面是Perl的的解决方案，对于给定的例子工程。请参阅live demo。

#!/usr/bin/perl -w 

use strict; 
use warnings; 

while (<DATA>) { 

# \1/ Starting at the beginning of a string or non-word character, 
# \2/ MATCH a single-quote character followed by a character that is 
#  *not* a single quote character, 
# \3/ And continue matching one or more times: 
#  - a white space character, 
#  - a word character, 
#  - a comma, 
#  - or a single-quote that is followed by a lower-case 's' or 't'. 
# \4/ And END the match on a single quote. 
# \5/ Continue searching for additional matches. 

    my @matches = /(?:\A|\W)('[^'](?:\w|\s|,|'(?=[st]\b))+')/g; 

#     \___1___/\__2_/\___________3__________/4/\5/ 

    print join("\n", @matches), "\n"; 
} 

__END__ 
'At the Beginning' ABC released its full midseason schedule today, and it features premiere dates for several new shows, along with one rather surprising timeslot change.</p><p>First of all, ABC's previously reported plans for dramas 'Once Upon A Time,' 'Revenge,' 'Grey's Anatomy,' and 'Scandal' haven't changed. 

预期输出：

'At the Beginning' 
'Once Upon A Time,' 
'Revenge,' 
'Grey's Anatomy,' 
'Scandal'