after creating a language and a grammar , this program will read .compila files and apply the lexical ,syntaxic , symantic analysis
Start_Program
Int_Number : i , j , Aft_5 ;;
Give i : 23 ;;
Real_Number : Aft34_2 ;;
If -- i < j -- Give Aft_5 : 10 ;;
Else
Start
Affect i to j ;;
Give Aft34_2 : 123.54 ;;
Finish
ShowMes : " this is a message " ;;
ShowVal : i ;;
//. this is a comment
End_Program
we need to give every preDefined words a token :
- preDefined words means all reserved array of charachters ( String ) that doesn't change.
- token it's a way to detect the reserved word name.
| Reserved Words | Token | Reserved Words | Token |
|---|---|---|---|
| Start_Program | START_PROGRAM | End_Program | END_PROGRAM |
| Int_Number | INTEGER | Give | GIVE |
| Real_Number | FLOAT | If | IF |
| Affect | AFFECT | //. | COMMENT |
| to | TO | == | EQUAL |
| ShowMes | SHOW_MES | -- | RESERVED_CHARACHTER |
| ShowVal | SHOW_VAL | >= | GTE |
| Start | START | <> | NOT_EQUAL |
| Finish | FINISH | <= | LTE |
| Else | ELSE | < | SMALLER |
| , | RESERVED_CHARACHTER | : | RESERVED_CHARACHTER |
| ;; | END_INSTRUCTION |
- Here is the implementation in c++ :
enum Tokens
{
START_PROGRAM,END_PROGRAM,COMMENT,END_INSTRUCTION,RESERVED_WORD,STRING,
INTEGER,FLOAT,IDENT,GIVE,AFFECT,RESERVED_CHARACHTER,IF,BIGGER,SMALLER,
EQUAL,NOT_EQUAL,ELSE,START,FINISH,SHOW_MES,SHOW_VAL,TO,END_FILE,LTE,
GTE,ERROR,NUMBER
};after that with the given file.compila we need to acces the data in and scan it.
- read the file and scan that data using :
ifstream f;
f.get();- in the function
Tokens Scanner::getToken1()it will scan charachter by charachter and return the given token. strings that aren't reserved words are
- ident the function that detects ident in c++ :
else if (isalpha(ch))
{
s += ch;
f.get(ch);
while (isalpha(ch) || ch == '_' || isdigit(ch))
{
s += ch;
f.get(ch);
}
f.putback(ch);
return checkReserved1(s);
}- integer number the function that detects integer numbers in c++ :
else if (isdigit(ch))
{
s = ch;
f.get(ch);
while (isdigit(ch))
{
s += ch;
f.get(ch);
}
f.putback(ch);
return Tokens::NUMBER;
}- float number
the function that detects float numbers in c++ :
else if (isdigit(ch))
{
s = ch;
f.get(ch);
while (isdigit(ch) || ch == '.')
{
if(ch == '.'){c++;}
if(c==2){ return Token::ERROR}
s += ch;
f.get(ch);
}
f.putback(ch);
return Tokens::NUMBER;
}all other are the same as this two code snippets . after scanning all the file we will have all the tokens and iden's and numbers than we implement a switch to print every word if it's a reserved works and charachters and iden's ...
- The c++ code :
void Scanner::displayTokens1()
{
Tokens a;
if (f.eof()) cout << "END OF FILE \n";
while (!f.eof())
{
a = getToken1();
switch (a)
{
case START_PROGRAM: cout << "Start_Program -----> START OF THE PROGRAM\n";
break;
case END_PROGRAM: cout << "End_Program -----> END OF PROGRAM\n";
break;
case COMMENT: cout << "//. -----> COMMENT STARTER\n";
break;
case END_INSTRUCTION: cout << ";; -----> END OF INSTRUCRUCTION \n";
break;
case RESERVED_WORD: cout << "RESERVED WORD\n";
break;
case STRING: cout << "STRING WORD\n";
break;
case INTEGER: cout << "Int_Number -----> word reserved charachter\n";
break;
case FLOAT: cout << "Real_Number -----> word reserved charachter\n";
break;
case IDENT: cout << "IDENTIFIRE\n";
break;
case GIVE: cout << "Give ----> Reserved Give Word \n";
break;
case AFFECT: cout << "Affect ----> Reserved word for affectation\n";
break;
case RESERVED_CHARACHTER: cout << "RESERVED CHARACHTER\n";
break;
case IF: cout << "IF -----> CONDITION\n";
break;
case BIGGER: cout << "> -----> BIGGER\n";
break;
case SMALLER: cout << "< -----> SMALLER\n";
break;
case EQUAL: cout << "== -----> EQUAL\n";
break;
case NOT_EQUAL: cout << "<> -----> NOT EQUAL\n";
break;
case ELSE: cout << "Else -----> ELSE CONDITION\n";
break;
case START: cout << "Start -----> START OF THE BLOCK\n";
break;
case FINISH: cout << "Finish -----> END OF THE BLOCK\n";
break;
case SHOW_MES: cout << "ShowMes -----> RESERVED TO SHOW MESSAGES\n";
break;
case SHOW_VAL: cout << "ShowVal -----> RESERVED TO SHOW VARIABLES VALUES\n";
break;
case TO: cout << "to -----> TO reserved charachter\n";
break;
case LTE: cout << "<= -----> LESS THEN OR EQUAL reserved charachter\n";
break;
case GTE: cout << ">= -----> GREATER THAN OR EQUAL reserved charachter\n";
break;
case NUMBER: cout << "NUMBER\n";
break;
case END_FILE: cout << "\n ============== THE END OF THE FILE ======================\n";
break;
case ERROR: cout << "ERROR\n";
break;
}
}
}- Here is the output reading txt.compila file :
we need to generate a grammar to use this tokens in a coherent way every token in it's positon than define the meaning of every instruction in the symatix phase
- Grammar :
<programe> ---> Start_Programe <seq> End_Programe
<seq> ---> E | <type> : <identList> ;; <seqTail> | <command> <seqTail>
<seqTail> ---> E | <seq>
<type> ---> Int_Number | Real_Number
<identList> ---> <ident> <identListTail>
<identListTail> ---> E | ,<identList>
<command> ---> Give <ident> : <typeNumber> ;; | ShowMes : <string> ;; | ShowVal : <ident> ;; | <comment> | If -- <exp> -- <command> <commandTail>
<commandTail> ---> ;; | Else Start <command> Finish
<exp> ---> <ident> <exp> <ident>
<typeNumber> ---> <integer> | <real>
<op> ---> == | >= | <= | <>
<string> ---> C++ code implementation
<comment> ---> C++ code implementation
<integer> ---> C++ code implementation
<real> ---> C++ code implementation
<ident> ---> C++ code implementation
- using this grammar and tree Data structure we can detect if the position of the token in every instruction matched to it's place .
- now every none Terminal Variable needs a function to scan and construct the instruction
- non Terminal like
<seq>and<exp> - here is the table describe all the functions needed for all none Terminal varibale
Node* programe(); --------> for non Terminal <program>
Node* seq();--------------> for non Terminal <seq>
Node* seqTail();----------> for non Terminal <seqTail>
Node* typeTest();---------> for non Terminal <typeTest>
Node* type();-------------> for non Terminal <type>
Node* identList();--------> for non Terminal <identList>
Node* identListTail();----> for non Terminal <identListTail>
Node* command();----------> for non Terminal <command>
Node* typeNumber();-------> for non Terminal <typeNumber>
Node* exp();--------------> for non Terminal <exp>
Node* commandTail();------> for non Terminal <commandTail>
Node* op();---------------> for non Terminal <op>- Here is the output parsing sample.compila file :
