Re: [Boost-users] boost::regex_replace problem

Subject: Re: [Boost-users] boost::regex_replace problem
From: Steven Watanabe (watanabesj_at_[hidden])
Date: 2014-04-03 14:28:18

AMDG

On 04/03/2014 09:18 AM, Илья wrote:
> Greetings everyone! My name is Ilya and i have such problem with boost::regex_replace : I want to process some text messages to find and replace with * swears there. I wrote regular expressions, this is an example of one of them: "([[:punct:]]|[[:space:]]|[[:digit:]]|^)(?:a)(?:[[:punct:]]|[[:space:]]|[[:digit:]])*(?:b)(?:[[:punct:]]|[[:space:]]|[[:digit:]])*(?:c)([[:punct:]]|[[:space:]]|[[:digit:]]|$)". The format string i used is: "$1***$2" 
> Then i run boost::regex_replace() and it works almoust correctly, except one case: when i have message with abc's standing next to each other with one separator beetween them, like this: "abc abc", "--abc-abc--" etc. So i need to run boost::regex_replace() twice to process corectly such messages. 
> Are there any other ways to solve this problem? Thank you for answers. 
> 

If I'm understanding correctly, you should be
able to use lookahead (?=) and/or lookbehind (?<=),

I've simplified it to use just \s, as a separator:
"(?:^|(?<=\\s))(?:a)\\s*(?:b)\\s*(?:c)(?:(?=\\s)|$)"
and then just use "***" as the format. (warning: untested)

In Christ,
Steven Watanabe