Scribd
Upload
9 views11 pages

Lecture 5

The document discusses converting an NFA to a DFA and implementing lexical analysis using DFAs. It covers the epsilon closure calculation, constructing the DFA transition table from the NFA, and using DFA states to distinguish lexemes and handle lexical errors.

Uploaded by

Vedang Chavan
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
9 views11 pages

Lecture 5

The document discusses converting an NFA to a DFA and implementing lexical analysis using DFAs. It covers the epsilon closure calculation, constructing the DFA transition table from the NFA, and using DFA states to distinguish lexemes and handle lexical errors.

Uploaded by

Vedang Chavan
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
Download as pdf or txt
Download as pdf or txt
You are on page 1/ 11

CS327 - Compilers

Implementing Lexical
Specifications

Abhishek Bichhawat 24/01/2024


NFA to DFA
● NFA can be in many states at any time
○ Combine them into a set of states (a new state) for DFA
● Compute ε-closure of different states
○ Set of all states that are reachable from the current state in ε-moves

● Start state of DFA → subset of NFA states reachable through ε-moves


from the NFA start state
● Each state of DFA → non-empty subset of NFA states (including
those reachable through ε-moves)
● Transition in DFA → S0 →c S1 iff S1 is the subset of states
reachable from state S0 after seeing input c (including ε-moves)
NFA to DFA
((ab)*|c)*
ε

ε a ε b ε
B C E G H I
ε ε
A ε J
ε ε ε
c
D F K
ε
ε
NFA to DFA
((ab)*|c)*

b
a EG HIJKABCD

ABCDIJK c
FJKABCDI
NFA to DFA
((ab)*|c)*

a
b
a EG HIJKABCD
a
ABCDIJK c
c
FJKABCDI

c
NFA to DFA
a
B

a a
b a

a b
A C D
a
Implementing DFAs
● Use the state transitions to construct a lookup table
○ Rows can be states and columns can be input alphabets
○ Initialize all cells to empty state (Φ)
○ For each of the alphabets, fill the cell with the state that the “current” state
(represented by the row) can transition to.
○ Based on the “current” state, determine the next state

1
0 1

s0
1 s1 s2 s0 s1 s2
0 s1 s2 s0

0 s2 Φ Φ
Distinguishing lexemes
Suppose we have a regex : (break | [a-z]+)
How do we distinguish break from breaker using the DFA?

b r e a k
K ε
ε

ε [a-z] ε
I

[a-z]+
Distinguishing lexemes
Suppose we have a regex : (break | [a-z]+)
How do we distinguish break from breaker using the DFA?

r e a k
I I I I KI
b
[a-qs-z] [a-df-z] [b-z] [a-jl-z]
[ac-z] [a-z]
I

[a-z]+
Lexical Errors
● Lexical analyzer generally reports lexically invalid errors:
○ Length of number, string, identifier etc.
○ Incomplete number (e.g., 1. etc.)
○ Quotes on strings, characters
○ …
Lexical Analysis
How do you represent the string of balanced parentheses using RE?

You might also like