1. Regular Definition and Transition Diagrams
CSE-415

2. Contents Regular Definition Transition Diagrams - Node + Edges
Identifiers
Keywords
Relational Operators
Constants
Unsigned Number
White Spaces

3. Regular Definition
Giving names to regular expressions is referred to as a Regular definition.
If Σ is an alphabet of basic symbols, then a regular definition is a sequence of definitions of the form:
d1→r1
d2→r2
………
dn→rn
where,
Each di is a new symbol, not in Σ .
Each ri is a regular expression.

4. Example
Strings of letters, digits and underscores are C Identifiers. Regular definition for the language of C identifiers:
letter →A|B|….|Z|a|b|….|z|_
digit→0|1|….|9
id→letter(letter|digit)*

5. Transition Diagrams
We represent patterns in form of flow charts, called transition diagram.
Consists of set of nodes and edges that connect one state to another.

6. White Space delim * other delim start q0 q1 q2
delim -> blank| new line | tab

7. Node Nodes represents states.
State represents what characters we have seen between lexemeBegin pointer and forward pointer.
Transition diagram always begin with start state q0.
One or more states are final state(s). We may attach action to final states to indicate that token is returned to parser.

8. Continue…
Retraction: lexeme does not include the symbol that got us to accepting state, then we retract forward pointer one position.
We place a * near the final state.

9. Identifiers/Keywords
letter or digit *
other
letter
return(getToken(), InstallID())
start q0 q1 q2
Pattern for identifier:
([a-z]|[A-Z]) ([a-z]|[A-Z]|[0-9])*
Regular Definition:
Id -> letter(letter | digit)*
First Approach:

10. More…
Install reserved words in symbol table initially as special entries.
installID(): When an identifier is found , a call to installID() places it in symbol table if it is not already there and returns a pointer to symbol table entry for the lexeme found. Any identifier not in symbol table during LA can not be a reserved word so it’s token is id.
getToken(): The function getToken() examines the symbol table entry for the lexeme found and return whatever token name the symbol table says this lexeme represents- either id or one of the keyword token initially installed in table.

11. Keywords Second Approach: blank or newline * return(1 ,) start E N Dq0 q1 q2 q3 q4
blank or
newline L *
return(2 ,) S E q5 q6 q7 q8
blank or
newline I F *
return(3 ,) q9
q10
q11
Pattern for keywords:
e( nd | lse ) |if
Second Approach:

12. Relational Operators * start < Not = or < return(relop ,LT ) q0
return(relop , LE) = q8 *
return(relop,EQ)
q11

13. Constants digit * not digit digit return(number, INSTALL()) start q0
Pattern for constant:
[0-9] ([0-9])*
Regular Definition:
Id -> digit(digit)*

14. Unsigned Numbers Speed of Light: 3.0 E+8 m/sec
digit
digit
digit
+ or - * .
digit
other
start
digit
digit E q7 q0 q1 q2 q3 q4 q5 q6 E * *
digit
other
other q8 q9
return(number, pointer to table)
Pattern for Unsigned Number:
[0-9] ([0-9])* (.[0-9] ([0-9])*)) ? (E[+|-]? [0-9] ([0-9])*))?

15. End