lex-yacc

Simple calculator design (lex and yacc)

BIL-395 HW-1 Nursize Turan 201104091

PS: I work on Windows so I use flex and bison since lex and yacc not supported directly in Windows.

I first design the lex code by implementing the instructions step by step.

• Lex Code Design:

  • The lexer identifies tokens based on the given arithmetic expressions.

  • It supports both integer (NUMBER) and floating-point (FLOAT) numbers, assigning their values to yylval.val to pass them to the parser.

  • Spaces and tabs are ignored by the lexer.

  • The lexer is designed to recognize valid numbers, arithmetic operators, and parentheses. So, if an unexpected character appears, it triggers an error message.

  Additional Details:

  atoi is used for integer conversion, atof is used for floating-point conversion.

  FLOAT is explicitly defined for floating-point number support.

  NEWLINE is treated as an end-of-expression signal, ensuring that the expression result is printed after pressing enter.

Then I started to design yacc code by following insturctions step by step.

• Yacc Code Design:

  • The Yacc file defines the grammar rules for arithmetic expressions.

  • The parser provides error messages without terminating execution.

  • Tokens are used to represent arithmetic expressions, and precedence rules are applied to follow mathematical conventions.

  • Operations follow a left-to-right evaluation order, except exponentiation, which has right-associativity.

  • FLOAT is treated similarly to NUMBER in expressions.

  • %union is used to store token values, with double val handling both integers and floating-point numbers.

  • The program is designed to handle multiple expressions in a single run.

  • NEWLINE is used as an end-of-expression signal.

  Expression Rules:

  $1 → Represents the first part of the rule.

  $2 → Represents the middle part (e.g., value inside parentheses).

  $3 → Represents the third part of the rule.

To run code in Windows;

1. install flex, bison, dev-cpp
2. go to the path of calculator.l and calculator.y and run the command givn below using cmd;
3. flex calculator.l
4. bison -d calculator.y (Different from Linux yacc (y.tab.c and y.tab.h), bison creates calculator.tab.c and calculator.tab.h)
5. gcc lex.yy.c calculator.tab.c -o calculator
6. calculator

To run code in Linux;

1. sudo apt install flex bison
2. go to the path of calculator.l and calculator.y
3. lex calculator.l
4. yacc -d calculator.y
5. gcc lex.yy.c y.tab.c -o calculator -lm
6. ./calculator

PS: In Linux, it is necessary to add -lm to the end of gcc compiler line like given below because of the "pow": gcc lex.yy.c y.tab.c -o calculator -lm

Test Cases PDF Drive Link : https://drive.google.com/file/d/1J_rKJpMc0ByZa3r9aj-jss61SoD9ivgc/view?usp=sharing