regexp_pme.C File Reference

Detailed Description

Class TPMERegexp - API similar to PME - PCRE Made Easy Tries to be as close as possible to PERL syntax and functionality.

Extension of TPRegexp class, see also macro 'regexp.C'.

 
Global matching
----------------------------------------------------------------
Regexp='ba[rz]', Opts='g'
  last string='foobarbaz'
  number of matches = 1
  0 - bar
Regexp='ba[rz]', Opts='g'
  last string='foobarbaz'
  number of matches = 1
  0 - baz
 
Global matching with back-refs
----------------------------------------------------------------
Regexp='(ba[rz])', Opts='g'
  last string='foobarbaz'
  number of matches = 2
  0 - bar
  1 - bar
Regexp='(ba[rz])', Opts='g'
  last string='foobarbaz'
  number of matches = 2
  0 - baz
  1 - baz
 
Matching with nested back-refs
----------------------------------------------------------------
Regexp='([\w\.-]+)@((\d+)\.(\d+)\.(\d+)\.(\d+))', Opts=''
  last string='matevz.tadel@137.138.170.210'
  number of matches = 7
  0 - matevz.tadel@137.138.170.210
  1 - matevz.tadel
  2 - 137.138.170.210
  3 - 137
  4 - 138
  5 - 170
  6 - 210
 
Split
----------------------------------------------------------------
Regexp=':', Opts=''
  last string='root:x:0:0:root:/root:/bin/bash'
  number of matches = 7
  0 - root
  1 - x
  2 - 0
  3 - 0
  4 - root
  5 - /root
  6 - /bin/bash
 
Split with maxfields=5
----------------------------------------------------------------
Regexp=':', Opts=''
  last string='root:x:0:0:root:/root:/bin/bash'
  number of matches = 5
  0 - root
  1 - x
  2 - 0
  3 - 0
  4 - root:/root:/bin/bash
 
Split with empty elements in the middle and at the end
maxfields=0, so trailing empty elements are dropped
----------------------------------------------------------------
Regexp=':', Opts=''
  last string='root::0:0:root:/root::'
  number of matches = 6
  0 - root
  1 - 
  2 - 0
  3 - 0
  4 - root
  5 - /root
 
Split with empty elements at the beginning and end
maxfields=-1, so trailing empty elements are kept
----------------------------------------------------------------
Regexp=':', Opts=''
  last string=':x:0:0:root::'
  number of matches = 7
  0 - 
  1 - x
  2 - 0
  3 - 0
  4 - root
  5 - 
  6 - 
 
Split with no pattern in string
----------------------------------------------------------------
Regexp=':', Opts=''
  last string='A dummy line of text.'
  number of matches = 1
  0 - A dummy line of text.
 
Split with regexp potentially matching a null string 
----------------------------------------------------------------
Regexp=' *', Opts=''
  last string='hi there'
  number of matches = 7
  0 - h
  1 - i
  2 - t
  3 - h
  4 - e
  5 - r
  6 - e
 
Split on patteren with back-refs
----------------------------------------------------------------
Regexp='([,-])', Opts=''
  last string='1-10,20'
  number of matches = 5
  0 - 1
  1 - -
  2 - 10
  3 - ,
  4 - 20
 
Substitute
----------------------------------------------------------------
Regexp='(\d+)\.(\d+)\.(\d+)\.(\d+)', Opts=''
Substitute '137.138.170.210','$4.$3.$2.$1' => '210.170.138.137'
 
Global substitute
----------------------------------------------------------------
Regexp='(\w+)\.(\w+)@[\w\.-]+', Opts='g'
Substitute 'rene.brun@cern.ch, philippe.canal@fnal.gov, fons.rademakers@cern.ch','\u$1 \U$2\E' => 'Rene BRUN, Philippe CANAL, Fons RADEMAKERS'
 

 
void regexp_pme()
{
   static const char *underline =
      "----------------------------------------------------------------\n";
 
 
   // Match tests
 
   {
      printf("Global matching\n%s", underline);
      TPMERegexp re("ba[rz]", "g");
      TString m("foobarbaz");
      while (re.Match(m))
         re.Print("all");
      printf("\n");
 
      printf("Global matching with back-refs\n%s", underline);
      TPMERegexp re1("(ba[rz])", "g");
      TString m1("foobarbaz");
      while (re1.Match(m1))
         re1.Print("all");
      printf("\n");
 
      printf("Matching with nested back-refs\n%s", underline);
      TPMERegexp re2("([\\w\\.-]+)@((\\d+)\\.(\\d+)\\.(\\d+)\\.(\\d+))");
      TString m2("matevz.tadel@137.138.170.210");
      re2.Match(m2);
      re2.Print("all");
      printf("\n");
   }
 
 
   // Split tests
 
   {
      printf("Split\n%s", underline);
      TPMERegexp re(":");
      TString m("root:x:0:0:root:/root:/bin/bash");
      re.Split(m);
      re.Print("all");
      printf("\n");
 
      printf("Split with maxfields=5\n%s", underline);
      re.Split(m, 5);
      re.Print("all");
      printf("\n");
 
      printf("Split with empty elements in the middle and at the end\n"
             "maxfields=0, so trailing empty elements are dropped\n%s", underline);
      m = "root::0:0:root:/root::";
      re.Split(m);
      re.Print("all");
      printf("\n");
 
      printf("Split with empty elements at the beginning and end\n"
             "maxfields=-1, so trailing empty elements are kept\n%s", underline);
      m = ":x:0:0:root::";
      re.Split(m, -1);
      re.Print("all");
      printf("\n");
 
      printf("Split with no pattern in string\n%s", underline);
      m = "A dummy line of text.";
      re.Split(m);
      re.Print("all");
      printf("\n");
   }
 
   {
      printf("Split with regexp potentially matching a null string \n%s", underline);
      TPMERegexp re(" *");
      TString m("hi there");
      re.Split(m);
      re.Print("all");
      printf("\n");
   }
 
   {
      printf("Split on patteren with back-refs\n%s", underline);
      TPMERegexp re("([,-])");
      TString m("1-10,20");
      re.Split(m);
      re.Print("all");
      printf("\n");
   }
 
 
   // Substitute tests
 
   {
      printf("Substitute\n%s", underline);
      TPMERegexp re("(\\d+)\\.(\\d+)\\.(\\d+)\\.(\\d+)");
      TString m("137.138.170.210");
      TString r("$4.$3.$2.$1");
      TString s(m); re.Substitute(s, r);
      re.Print();
      printf("Substitute '%s','%s' => '%s'\n", m.Data(), r.Data(), s.Data());
      printf("\n");
   }
 
   {
      printf("Global substitute\n%s", underline);
      TPMERegexp re("(\\w+)\\.(\\w+)@[\\w\\.-]+", "g");
      TString m("rene.brun@cern.ch, philippe.canal@fnal.gov, fons.rademakers@cern.ch");
      TString r("\\u$1 \\U$2\\E");
      TString s(m); re.Substitute(s, r);
      re.Print();
      printf("Substitute '%s','%s' => '%s'\n", m.Data(), r.Data(), s.Data());
      printf("\n");
   }
}

Author

Eddy Offermann

Definition in file regexp_pme.C.