Thread: Regular Expression using regex_search

I search the board using "Regular Expression" and regex_search as keywords. I get "Sorry - no matches. Please try some different terms."

Anyway, I'm trying to get the output below using regex_search only. I'm playing it by ear because all I read seems to take me out the regex_search function. For instance out of all the great things I learned here, he cut it off right here:

These leaves us with:
Replace any * characters with .*
Replace any ? characters with .
Leave square brackets as they are.
Replace any characters which are metacharacters with a backslashified version.

Code:

#include <iostream>
#include <string>
#include <boost/regex.hpp>
using namespace std;


int main()
{
   boost::regex expression( ""x*\.o"" );

   string string1 = "How to re";

   boost::smatch match;

   while ( boost::regex_search( string1, match, expression,
      boost::match_not_dot_newline ) )
   {
      cout << match << endl;
      string1 = match.suffix();    } }

Code:

//  Output needed:   oH to tore

After over 30 hours of trying I found nothing that show me how to move back to the beginning to grab the {T} or how to store it for farther use. All of this and hundreds more ... after 30 hours, I still stuck. All I ask is how to move backwards or what do I have to do to get the first word in my result list in that order. I got enough below and more to take care of the rest I think.

Thanks in advance.

Code:

//	boost::regex expression( "(\w)\o" );   // back reference DON'T WORK
//   boost::regex expression( "[A-Z][a-zA-Z]*" );   // match first word, no symbols.
//   boost::regex expression( "[A-Z][A-Za-z]*" );   // same as above
//   boost::regex expression( "[a-zA-Z]+" );   // match all words and char only.
//   boost::regex expression( "[^a-zA-Z0-9]+" );   // match symbols only.
//   boost::regex expression( "[*.jpg]" );   // match word
//   boost::regex expression( "[*.jpg]+" );   // with + inc any matching char
//   boost::regex expression( "*.[ch]pp" );   // DON'T WORK FOR BOOST
//	boost::regex expression( "[^B]" );   // don't find this char or sequence
//	boost::regex expression( "[*B]" );   // Find all
//	boost::regex expression( "B+" );   // find one or more
//	boost::regex expression( "[a-z]+://" );   // get the header https://
//	boost::regex expression( "yM?" );   // must find more than one char of one
//	boost::regex expression( "{1}y" );   // match one DON'T WORK
//	boost::regex expression( "( "( ?B)+" );   // will only get 4 char
//	boost::regex expression( "( "( ?B)+" )";  // will only get 4 char
//	boost::regex expression( "[ ?Bith]+" );   // find any of these
//	boost::regex expression( "( ?ho)+" );   // whole word per line sequence
//	boost::regex expression( "cow(ard|age|boy|l)?" );   // connect parts


//	boost::regex expression( "^(\\D{2})" );   // get 5 char - (d) digit
// [^\5]
// [^(\k<1>)]
// [^${1}]
//	boost::regex expression( "(\\D+)(\\s*)(cow|M(M)?)" );   // back up and M to M
	//boost::regex expression( "[^M]+(y+ig)?[big(igbay|age|boy|l)?]?+(ho|M(y)?)" );   // works great
//	boost::regex expression( "^(\\D{2})+(\\D{2}) " );   // don't find this char or sequence
//	boost::regex expression( ".*\.o" );   // copy all o
//	boost::regex expression( "x*\.o" );   // copy only first o

... but I still read nothing about:

These leaves us with:
Replace any * characters with .*
Replace any ? characters with .
Leave square brackets as they are.
Replace any characters which are metacharacters with a backslashified version.

Regular Expressions Tutorial

... boost::regex_search for what I'm trying to do.

flat assembler - View topic - Feature Request: Regular Expressions

Regular Expression Matching Can Be Simple And Fast

Benchmark of Regex Libraries

Qt 4.7: QRegExp Class Reference

Regular expression - Wikipedia, the free encyclopedia

Perl Regular Expression Syntax - Boost 1.43.0 ... now things are getting easier. It's a new way to do your abc's ... That took me until Junior High ... heehee


PS: After posting this thread now I see 4 links to Regular Expression below but 9 out of 10 they are not related. Going to read them now.

The new forum look is very nice and it feels smooth

The "This leaves us with" part is just instructions on how to translation a shell glob to a regular expression. Other than that, I'm afraid I don't understand your question.