Regex

This is a regular expression engine implemented in Python that uses NFA and DFA and optionally minimizes DFA sets.

Begin to use

from regex import Regex


st = 'THISISREGEXTEST'
pattern = '([A-Z]*|[0-9]+)'
regex = Regex(st, pattern)
result = regex.match()
log(result)

Regex build parameter

(input_string, pattern_string, mode=1, minimize=True)

• mode

  Finite state automata used

  mode = 1 - NFA

  mode = 2 - DFA

• minimize

  Whether to minimize if using DFA

see sample.py for details.

BNF

group ::= ("(" expr ")")*
expr ::= factor_conn ("|" factor_conn)*
factor_conn ::= factor | factor factor*
factor ::= (term | term ("*" | "+" | "?"))*
term ::= char | "[" char "-" char "]" | .

Implementation of the

Implemented all the basic syntax

Tokens = {
    '.': Token.ANY,
    '^': Token.AT_BOL,
    '$': Token.AT_EOL,
    ']': Token.CCL_END,
    '[': Token.CCL_START,
    '}': Token.CLOSE_CURLY,
    ')': Token.CLOSE_PAREN,
    '*': Token.CLOSURE,
    '-': Token.DASH,
    '{': Token.OPEN_CURLY,
    '(': Token.OPEN_PAREN,
    '?': Token.OPTIONAL,
    '|': Token.OR,
    '+': Token.PLUS_CLOSE,
}

in lex.token

Code structure

For the sake of simplicity, the code style of this implementation is not very good.

• lex

  Lexical analysis of regular expressions

• nfa

  Definition of an NFA node the construction of an NFA

• dfa

  Definition of an DFA node the construction of an DFA

• parse

  About parsing DFA or NFA based on input