regex2dfa 0.2.1

Convert regular expressions to minimized DFAs in AT&T FST format

regex2dfa

Convert regular expressions to minimized DFAs in AT&T FST format.

Pure Python · Zero dependencies · Results cached in memory

Installation

pip install regex2dfa

Usage

from regex2dfa import regex2dfa

dfa = regex2dfa("(a|b)+")
print(dfa)

Output:

0	1	97	97
0	1	98	98
1	1	97	97
1	1	98	98
1

Features

• Pure Python — No external dependencies
• Fast — Results cached with LRU (1024 entries)
• Complete pipeline — Thompson's → Subset → Hopcroft minimization

Supported Syntax

Feature	Syntax	Example
Literal	a	abc
Any char	.	a.b
Character class	[abc], [a-z], [^abc]	[0-9]+
Quantifiers	*, +, ?	a*, b+, c?
Alternation	|	cat|dog
Grouping	(...)	(ab)+
Escapes	\x00, \d, \w, \s, \C	\xFF, \d+

Output Format

AT&T FST format with tab-separated fields:

• Transitions: src dst input output
• Final states: state_id

Labels are ASCII byte values (e.g., 97 = 'a').

Links

• GitHub
• Documentation

License

MIT