C _Regex - RicoJia/notes GitHub Wiki

========================================================================

Regex

======================================================================== 0. Introduced in C++11

Test Program

#include <iostream>
#include <regex>
#include <vector>
#include <string>
using namespace std;

int main()
{
  regex r("abc");         //exact match
  vector<string> vec {"abv", "abc", "ABC"}; 
  for(const string& str: vec){
      bool match = regex_match(str, r); 
      cout << (match ? "matched" : "not matched") << endl;  
  }
  return 0;
}

match anything but newline: .*

exact matches

regex r("abc");         //exact match
vector<string> vec {"abv", "abc", "ABC"}; // not match, match, not match

Case Insensitive, regex_constants::icase

regex r("abc", regex_constants::icase);         //icase = insenstive casing
vector<string> vec {"abv", "abc", "ABC", "aBC"};    //not match, match, match, match

Any char but '\n', . (exactly one char)

regex r("abc.");         
vector<string> vec {"abc\n", "abc", "abcd", "abccc"};    //not match, not match (there's gotta be a char), match, not matched

or |

regex r("ab|cd");         // ab or cd
vector<string> vec { "ab", "cd", "abc"};    //match, match, not match

Be careful with spaces, they count!

regex r("ab ");         // space must be counted insenstive
vector<string> vec {"ab"};    // not match, cuz YOU NEED THAT SPACE!

Escape using \\

regex r("ab\\[");         // 3 chars that are not in [cd]
vector<string> vec { "ab", "ab["};    //not match, match

operators dependent on number of preceeding chars

0 or 1 preceeding char, ?

regex r("abc?");         // ? is [0,1] preceeding char of ?. equivalent to "ab", "abc"
vector<string> vec {"ab", "abc", "abd", "abcd"};    // match, match, not match, not match (just one char only)

any number of the preceeding char, *, could be zero

regex r("abc*");         // * is [0, inf] of the preceeding char. equivalent to "ab", "abc"
vector<string> vec {"ab", "abc", "abcc", "abccc", "abcd", "abcdd"};    // match, match, match, match, not match (must be preceeding char), not match

\s* means any number of white spaces
. is wild card for any character

At least 1 preceeding char, +

regex r("abc+");         // * is [1, inf] of the preceeding char. equivalent to "ab", "abc"
vector<string> vec {"ab", "abc", "abcc", "abccc", "abcd", "abcdd"};    // not match, match, match, match, not match (must be preceeding char), not match

scope operators

A single char from []

regex r("ab[cd]");         // a single char from []
vector<string> vec {"ab", "abc", "abcc", "abccc", "abcd", "abcdd"};    // not match, match, not match, not match, not match, not match

A range of char

regex r("ab[a-z]+");         // [a-z] means anything in a-z
vector<string> vec { "abcdfaasdfa"};    // match

number of chars using {}

Exactly some number of chars

regex r("ab[^cd]{3}");         // 3 chars that are not in [cd] 
vector<string> vec {"abcdc", "abc", "abd", "abeee"};    // not match, not match, not match, match

A range of chars

regex r("ab[^cd]{3,5}");         // 3-5 chars that are not in [cd] 
vector<string> vec { "abeeeeee", "abeee"};    //not match, match

more than some number of chars

regex r("ab[^cd]{3,}");         // 3+ chars that are not in [cd] 
vector<string> vec { "abeeeeee", "abeee"};    //match, match

Composite operators

[0, inf] number of chars from []

regex r("ab[cd]*");         // [0, inf] number of chars from []
vector<string> vec {"ab", "abc", "abcc", "abccc", "abcd", "abcdd"};    // match, match, match, match, match, match

[^ ... ] is anything but these chars

regex r("ab[^c]");         // [^] is anything but these char
vector<string> vec {"ab", "abc", "abd"};    // not match, not match, match
regex r("ab[^cd]");         // ^ is anything but these chars
vector<string> vec {"ab", "abc", "abd"};    // not match, not match, not match

[^...]* is [0, inf] number of anything but these chars

regex r("ab[^cd]*");         // [^...]* is [0, inf] number of anything but these chars
vector<string> vec {"ab", "abc", "abd", "abfg"};    // match, not match, not match, match

() is a sub_group, and \\1 means to repeat the first sub_group

regex r("(ab)\\1");         // 3 chars that are not in [cd]. () is called capture group, which is the sequence matching "ab"
vector<string> vec { "ab", "abab"};    //not match, match

with an +

regex r("(abc)de+\\1");         // 3 chars that are not in [cd] 
vector<string> vec { "ab", "abpabd", "abcdeeabc"};    //not match, not match, match

multiple sub_groups, with + in ()

regex r("(ab)c(de+)\\2\\1");         // 3 chars that are not in [cd] 
vector<string> vec { "ab", "abpabd", "abcdeedeeab"};    //not match, not match, match

^means to be at the beginning of the string, when used alone

regex r("^abc.");         // searching for an email, [[:w]] is a word char, @ is @, \\. is to escape and we get a pure . 
vector<string> vec { "abcd", "bbabcd"};    // match, not matched

Don't use this: Cpp doesn't support this. $ means abc should be at the end of line

regex r("abc$");         //Don't use this: Cpp doesn't support this. $ means abc should be at the end of line 
vector<string> vec { "abcd", "bbabc"};    //not match, not matched

Examples

tell if something is an email:

regex r("[[:w:]]+@[[:w:]]+\\.com");         // searching for an email, [[:w]] is a word char, @ is @, \\. is to escape and we get a pure . 
// word char: digit, number, underscore
vector<string> vec { "[email protected]"};    // match

Tell if a string contains a match:

int main()
{
    regex r("[[:w:]]+@[[:w:]]+\\.com");         // searching for an email, [[:w]] is a word char, @ is @, \\. is to escape and we get a pure .
    vector<string> vec { "[email protected]"};    // match
    for(const string& str: vec){
        bool match = regex_search(str, r);
        cout << (match ? "matched" : "not matched") << endl;
    }
    return 0;
}

C _Regex - RicoJia/notes GitHub Wiki

Regex

⚠️ **GitHub.com Fallback** ⚠️

⚠️ GitHub.com Fallback ⚠️