Detailed Description

A regular expression, often called a pattern, is an expression that describes a set of strings.

They are usually used to give a concise description of a set, without having to list all elements. The Unix utilities like sed and grep make extensive use of regular expressions. Scripting languages like Perl have regular expression engines built directly into their syntax .

Extensive documentation about Regular expressions in Perl can be found at: http://perldoc.perl.org/perlre.html

ROOT has this capability through the use of the P(erl) C(ompatible) R(egular) E(xpression)

library, PCRE, see http://www.pcre.org

Its functionality can be accessed through the TPRegexp and TString class . Note that in patterns taken from Perl all backslash character have to be replaced in the C/C++ strings by two backslashes .

This macro shows several ways how to use the Match/Substitute capabilities of the the TPRegexp class . It can be run as follows :

.x regexp.C

Processing /mnt/build/workspace/root-makedoc-v614/rootspi/rdoc/src/v6-14-00-patches/tutorials/regexp.C...
pepernotenkoek
lekkere walnotenboom
two one three
on 24-09-1959 the world stood still
Check if the email address "fons.rademakers@cern.ch" is valid: TRUE
neutrino proton electron neutrino

#include "Riostream.h"
#include "TString.h"
#include "TPRegexp.h"
#include "TClonesArray.h"
#include "TObjString.h"
void regexp()
{
   // Substitute example :
   // Find a word that starts with "peper" and ends with "koek" .
   TString s1("lekkere pepernotenkoek");
   TPRegexp r1("\\bpeper(\\w+)koek\\b");
   // Note that the TString class gives access to some of the simpler TPRegexp functionality .
   // The following command returns the fully matched string .
   cout << s1(r1) << endl;
   // In the "Substitute" command, keep the middle part (indicated in the regexp by "(\\w+)"
   // and the substitute string by "$1") and sandwich it between "wal" and "boom" .
   r1.Substitute(s1,"wal$1boom");
   cout << s1 << endl;
   // Substitute example :
   // Swap first two words in a string
   TString s2("one two three");
   TPRegexp("^([^ ]+) +([^ ]+)").Substitute(s2,"$2 $1");
   cout << s2 << endl;
   // Substitute example :
   // $1, $2, and so on, in the substitute string are equivalent to whatever the corresponding set
   // of parentheses match in the regexp string, counting opening parentheses from left to right .
   // In the following example, we are trying to catch a date MMDDYYYY in a string and rearrange
   // it to DDMMYYY . "(\\d{1,2}) matches only 1 or 2 digits etc .
   TString s3("on 09/24/1959 the world stood still");
   TPRegexp("\\b(\\d{1,2})/(\\d{1,2})/(\\d{4})\\b").Substitute(s3,"$2-$1-$3");
   cout << s3 << endl;
   // Match Example :
   // The following example shows how to extract a protocol and port number from an URL string .
   // Note again the parentheses in the regexp string : "(\\w+)" requires a non-empty
   // alphanumeric string while "(\\d+)" wants a pure digital string .
   // The matched substrings together with the full matched string are returned in a
   // TObjArray . The first entry is the full string while next entries are the substrings
   // in the order as listed in the regexp string .
   //
   // Note that there is also a Match(..) command that returns the positions of the
   // substrings in the input string .
   TString s4("http://fink.sourceforge.net:8080/index/readme.html");
   TObjArray *subStrL = TPRegexp("^(\\w+)://[^/]+:(\\d+)/$").MatchS(s4);
   const Int_t nrSubStr = subStrL->GetLast()+1;
   if (nrSubStr > 2) {
     const TString proto = ((TObjString *)subStrL->At(1))->GetString();
     const TString port  = ((TObjString *)subStrL->At(2))->GetString();
     cout << "protocol: " << proto << "  port: " << port << endl;
   }
   // Match Example :
   // This example returns kTRUE if the email address is valid . For that it has to fulfil the following
   // criteria:
   // 1) It should be of the form string1@string2 . The "^" and "$" ensure that we compare the complete
   //    email string
   // 2) ([\\w-\\.]+)  :
   //    string1 is only allowed to be composed out of the alphanumeric characters, "-" and "." .
   //    The "+" ensures that string1 can not be empty .
   // 3) string2 is matched against three different parts :
   //    a. ((\\[[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}\\.)|(([\\w-]+\\.)+))  :
   //       This regular expression ensures that EITHER the string starts with "[" followed by three groups
   //       of numbers, separated by "." , where each group has 1 to 3 numbers, OR alphanumeric strings,
   //       possibly containing "-" characters, separated by "." .
   //    b. ([a-zA-Z]{2,4}|[0-9]{1,3})  :
   //       This part contains EITHER 2 to 4 alpha characters OR 1 to 3 numbers
   //    c. (\\]?)  :
   //       At most one "]" character .
   TString s5("fons.rademakers@cern.ch");
   TPRegexp r5("^([\\w-\\.]+)@((\\[[0-9]{1,3}\\.[0-9]{1,3}\\.[0-9]{1,3}\\.)|(([\\w-]+\\.)+))([a-zA-Z]{2,4}|[0-9]{1,3})(\\]?)$");
   cout << "Check if the email address \"" << s5 << "\" is valid: " << (r5.MatchB(s5) ? "TRUE" : "FALSE") << endl;
   // Substitute Example with pattern modifier :
   // Like in Perl, Substitute/Match commands accept modifier arguments . For instance a "g" modifier causes to
   // match the regexp globally . In the example below, all words starting and ending with the character "n"
   // are replaced by the word neutrino .
   TString s6("neutron proton electron neutron");
   TPRegexp("(n\\w+n)").Substitute(s6,"neutrino","g");
   cout << s6 << endl;
}

Author: Eddy Offermann

Definition in file regexp.C.