diff --git a/Makefile b/Makefile index a6d045b..2f53d61 100644 --- a/Makefile +++ b/Makefile @@ -40,6 +40,9 @@ test: ./$(EXE) -tok ./tests/test_simpleLiterals.alpha ./$(EXE) -tok ./tests/test_real_alpha_file.alpha ./$(EXE) -tok ./tests/test_real_alpha_2.alpha + ./$(EXE) -tok -st ./tests/test_real_alpha_2.alpha + ./$(EXE) -st -tok ./tests/test_operators.alpha + ./$(EXE) -st ./tests/test_keywords.alpha clean: rm -f *.o @@ -49,3 +52,4 @@ clean: rm -f *.tok rm -f grammar.tab.c rm -f grammar.tab.h + rm -f *.st \ No newline at end of file diff --git a/grammar.y b/grammar.y index 4a1c8a4..591d80a 100644 --- a/grammar.y +++ b/grammar.y @@ -6,13 +6,52 @@ #include %} -// Declarations - - -// Grammar Rules +%token ID 101 +%token T_INTEGER 201 +%token T_ADDRESS 202 +%token T_BOOLEAN 203 +%token T_CHARACTER 204 +%token T_STRING 205 +%token C_INTEGER 301 +%token C_NULL 302 +%token C_CHARACTER 303 +%token C_STRING 304 +%token C_TRUE 305 +%token C_FALSE 306 +%token WHILE 401 +%token IF 402 +%token THEN 403 +%token ELSE 404 +%token TYPE 405 +%token FUNCTION 406 +%token RETURN 407 +%token EXTERNAL 408 +%token AS 409 +%token L_PAREN 501 +%token R_PAREN 502 +%token L_BRACKET 503 +%token R_BRACKET 504 +%token L_BRACE 505 +%token R_BRACE 506 +%token SEMI_COLON 507 +%token COLON 508 +%token COMMA 509 +%token ARROW 510 +%token ADD 601 +%token SUB_OR_NEG 602 +%token MUL 603 +%token DIV 604 +%token REM 605 +%token LESS_THAN 606 +%token EQUAL_TO 607 +%token ASSIGN 608 +%token NOT 609 +%token AND 610 +%token OR 611 +%token DOT 612 +%token RESERVE 613 +%token RELEASE 614 +%token COMMENT 700 +%% + %% -semicolon.opt: - %empty -| ";" -; -%% \ No newline at end of file diff --git a/lexicalStructure.lex b/lexicalStructure.lex index 7857730..64b2863 100644 --- a/lexicalStructure.lex +++ b/lexicalStructure.lex @@ -81,6 +81,8 @@ SCHAR \\n|\\t|\\\"|[^\"\n\\] {ID} {if(DEBUG) {printf( "ID: %s (%d)\n", yytext, ID);} else {return ID;}} \n {line_number++; column_number = 1;} -. {column_number++;} +\t {column_number++;} +" " {column_number++;} +. {column_number++; return 1999;} %% diff --git a/print_symbol_table.c b/print_symbol_table.c new file mode 100644 index 0000000..c5e4208 --- /dev/null +++ b/print_symbol_table.c @@ -0,0 +1,98 @@ +#include +#include "symbol_table.h" +#include "symbol_table.c" +#include + +void print_symbol_table(SymbolTable *table, FILE *file_ptr){ + if(table->Parent_Scope == NULL){ + fprintf(file_ptr, "%-17s: %-6s : %-6s : %-21s: %-28s\n", "NAME", "SCOPE", "PARENT", "TYPE", "Extra annotation"); + } + TableNode * entrie = table->entries; + fprintf(file_ptr, "-----------------:--------:--------:----------------------:-----------------------------\n"); + int parant_scope = 0; + int current_scope = 0; + if(table->Parent_Scope != NULL){ + parant_scope = table->Parent_Scope->Line_Number*1000 + table->Parent_Scope->Column_Number; + current_scope = table->Line_Number*1000 + table->Column_Number; + } else { + current_scope = 1001; + } + + for(; entrie != NULL; entrie = entrie->next){ + if (parant_scope == 0){ + fprintf(file_ptr, "%-17s: %06d : : %-21s: %-28s\n", + entrie->theName, current_scope, + entrie->theType, "Extra annotation"); + } else { + fprintf(file_ptr, "%-17s: %06d : %06d : %-21s: %-28s\n", + entrie->theName, current_scope, parant_scope, + entrie->theType, "Extra annotation"); + } + } + if (table->Children_Scope != NULL){ + ListOfTable* node = table->Children_Scope; + for(; node != NULL; node = node->next){ + print_symbol_table(node->table, file_ptr); + } + } + if (table->Parent_Scope == NULL) { + fprintf(file_ptr, "-----------------:--------:--------:----------------------:-----------------------------\n"); + } +} +/* + +int main(void){ + char *prim = strdup("primitive"); + char *inte = strdup("integer"); + SymbolTable * parant = CreateScope(NULL, 1,1); + char *boole = strdup("Boolean"); + char *chare = strdup("character"); + char *str = strdup("string"); + char *arg = strdup("arg"); + // Value* v = calloc(1, sizeof(Value)); + CreateEntry(parant, prim, boole); + CreateEntry(parant, prim, chare); + CreateEntry(parant, prim, inte); + CreateEntry(parant, &"1 -> character", str); + CreateEntry(parant, &"integer -> integer", &"int2int"); + CreateEntry(parant, &"string -> integer", &"string2int"); + CreateEntry(parant, &"int2int", &"square"); + CreateEntry(parant, &"string2int", &"entry"); + SymbolTable * child = CreateScope(parant, 14,14); + CreateEntry(child, inte, &"x"); + SymbolTable * second = CreateScope(parant, 21,15); + CreateEntry(second, str, arg); + CreateEntry(second, inte, &"input"); + CreateEntry(second, inte, &"expected"); + CreateEntry(second, inte, &"actual"); + CreateEntry(second, &"$_undefined_type", &"result"); + SymbolTable * third = CreateScope(second, 33,44); + CreateEntry(third, &"BOO", arg); + CreateEntry(third, &"YAZOO", &"input"); + + TableNode *ret = table_lookup(third, "arg"); + printf("%s == %s\n", ret->theName, "arg"); + + ret = table_lookup(third, "hello"); + printf("This should be nil %p != %s\n", ret, "BOO"); + + ret = look_up(second, "input"); + printf("%s == %s\n", ret->theName, "input"); + + ret = look_up(second, "square"); + printf("%s == %s\n", ret->theName, "square"); + + ret = look_up(second, "spuare"); + printf("This should be nil %p == %s\n", ret, "square"); + + print_symbol_table(parant, stderr); + free(inte); + free(boole); + free(prim); + free(str); + free(chare); + free(arg); + return 0; +} + +*/ diff --git a/runner.c b/runner.c index c927d71..ee872f4 100644 --- a/runner.c +++ b/runner.c @@ -1,99 +1,172 @@ +//#include "symbol_table.h" +#include "symbol_table.c" #include "runner.h" - int main(int argc, char *argv[]) { - char *check_input; - int token; - FILE *output; - - if (argc == 1 || argc > 3) { - fprintf(stderr, "invalid input with 1, >3, or non .alpha arg \n"); - return -1; //no alpha file or too many args + if (argc == 1) { + fprintf(stderr, "INVALID INPUT: Include a .alpha file or use -help for more inputs \n"); + return -1; } else if (argc == 2) { + //can be either -help or .alpha file if (is_help(argv[1])) { return 0; - } else if (is_alpha_file(argv[1], strlen(argv[1])) != 0) { + } else if (is_alpha_file(argv[1], strlen(argv[1])) == 0) { + //run flex for now + no_flag = SET_FLAG; //no argument but valid input + alpha_file = fopen(argv[1], "r"); + + } else { + fprintf(stderr, "INVALID INPUT: Include a .alpha file or use -help for more inputs\n"); return -1; } - arg = NO_ARG; //no argument but valid input - yyin = fopen(argv[1], "r"); } else { - check_input = is_tok(argc, argv); - if (strcmp(CHECK_OTHER, check_input) == 0 || strcmp(INVALID_ARG, check_input) == 0) { - fprintf(stderr, "check_other or invalid_arg \n"); - return -1; //invalid argument (need to update as we add more valid arguments) + //last input must be .alpha + if (is_alpha_file(argv[argc - 1], strlen(argv[argc - 1])) != 0) { + fprintf(stderr, "INVALID INPUT: Include a .alpha file at end of input or use -help for more inputs \n"); + return -1; + } else { + //now check that other args are valid (flags will not be null if flag is present) + for (int i = 1; i < argc - 1; i++) { + if (check_flag(argv[i], argv[argc - 1]) != 0) { + fprintf(stderr, "INVALID FLAG(S): Use -help to view valid inputs \n"); + return -1; + } + } + alpha_file = fopen(argv[argc - 1], "r"); } - output = fopen(check_input, "w"); - arg = TOK_ARG; //it is a -tok arg with a valid alpha file at argv[2] - yyin = fopen(argv[2], "r"); } - while (0 != (token = yylex())) { - if (arg == TOK_ARG) { - fprintf(output, "%d %d %3d \"%s\"\n", line_number, column_number, token, yytext); + return run(alpha_file); +} + +int check_flag(char *arg, char* alpha) { + if (strcmp("-tok", arg) == 0) { + if (tok_flag == NULL) { + return new_file(arg, alpha); } - if(token == COMMENT){ - for (int i = 0; i < yyleng; i++) { - if(yytext[i] == '\n'){ - line_number++; - column_number = 0; - } - column_number++; - } - continue; + fprintf(stderr, "FLAGS REPEAT\n"); + return -1; + } else if (strcmp("-st", arg) == 0) { + if (st_flag == NULL) { + return new_file(arg, alpha); + } + fprintf(stderr, "FLAGS REPEAT\n"); + return -1; + } else { + fprintf(stderr, "INVALID FLAG: Use -help for valid inputs\n"); + return -1; + } +} + + +int run(FILE *alpha) { + int token; + //check that file exists + if (alpha == NULL) { + fprintf(stderr, "INPUT FILE NOT FOUND\n"); + return -1; + } + yyin = alpha; + while (0 != (token = yylex())) { + if (tok_flag != NULL) { + fprintf(tok_flag, "%d %d %3d \"%s\"\n", line_number, column_number, token, yytext); + } + if(token == COMMENT){ + for (int i = 0; i < yyleng; i++) { + if(yytext[i] == '\n'){ + line_number++; + column_number = 0; + } + column_number++; + } + continue; } + if(token == 1999){ + printf("On line number %d and column number %d we have an invalid character:%s\n",line_number,column_number,yytext); + } column_number += yyleng; } + if (st_flag != NULL) { + //output symbol table, file pointer is + print_symbol_table(curr,st_flag); + } + if (yyin != NULL) { fclose(yyin); } - if (output != NULL && arg == TOK_ARG) { - fclose(output); + + if (tok_flag != NULL) { + fclose(tok_flag); } + return 0; } + + bool is_help(char * input) { if (strcmp(input, "-help") == 0) { - printf("HELP:\nHow to run the alpha compiler:\n./alpha [options] program\nValid options:\n-tok output the token number, token, line number, and column number for each of the tokens to the .tok file\n"); + printf("%s", HELP); return true; } return false; } -char *is_tok(int argc, char *argv[]) { - if (argc == 3 && strcmp("-tok", argv[1]) == 0) { - char *input_prog = argv[2]; - int file_len = strlen(input_prog); - //check that input program is a .alpha file - if (is_alpha_file(input_prog, file_len) != 0) { - return INVALID_ARG; - } - - const char *basename = input_prog; - const char *slash = strrchr(input_prog, '/'); - if (slash != NULL) { - basename = slash + 1; - } - - // Calculate lengths - int basename_len = strlen(basename); - char* FILE_tok = calloc(basename_len - ALPHA_OFFSET + TOK_LEN + 2, sizeof(char)); - - // Copy filename and add .tok extension - strncpy(FILE_tok, basename, basename_len - ALPHA_OFFSET); - //fprintf(stderr, "hello"); - strcat(FILE_tok, ".tok"); - - return FILE_tok; +int new_file(char *arg, char *alpha) { + int type_len; + const char *basename = alpha; + const char *slash = strchr(alpha, '/'); + while (slash != NULL) { + basename = slash + 1; + slash = strchr(basename, '/'); } - return CHECK_OTHER; + if (strcmp(arg, "-tok") == 0) { + type_len = TOK_LEN; + } else if (strcmp(arg, "-st") == 0) { + type_len = ST_LEN; + } else { + fprintf(stderr, "INVALID FLAG: Use -help to view valid inputs\n"); + return -1; + } + + + // calculate lengths + int basename_len = strlen(basename); + char *file_name = calloc(basename_len - ALPHA_OFFSET + type_len + 2, sizeof(char)); + + //coy filename and add extension + strncpy(file_name, basename, basename_len - ALPHA_OFFSET); + strcat(file_name, "."); + strcat(file_name, arg + 1); + + + if (strcmp(arg, "-tok") == 0) { + tok_flag = fopen(file_name, "w"); + } else if (strcmp(arg, "-st") == 0) { + st_flag = fopen(file_name, "w"); + } + return 0; } -int is_alpha_file(char *file, int file_len) { - if (strcmp(".alpha", file + sizeof(char) * (file_len - ALPHA_OFFSET)) != 0) { + +int is_alpha_file(char *alpha, int file_len) { + if (strcmp(".alpha", alpha + sizeof(char) * (file_len - ALPHA_OFFSET)) != 0) { return -1; //not alpha file } return 0; //is alpha file } +void enter_scope(int line, int column){ + curr = CreateScope(curr, line, column); +} +void exit_scope() { + if(curr->Parent_Scope == NULL){ + printf("Can't close top"); + return; + } + curr = curr->Parent_Scope; +} + + + + diff --git a/runner.h b/runner.h index c96c3df..73c94f2 100644 --- a/runner.h +++ b/runner.h @@ -1,14 +1,9 @@ #define ALPHA_OFFSET 6 #define TOK_LEN 3 - -//returns for is_tok -#define INVALID_ARG "invalid" -#define CHECK_OTHER "diff" - -//argument type in main -#define NO_ARG 0 -#define DIFF_ARG 1 -#define TOK_ARG 2 +#define ST_LEN 2 +#define HELP "HELP:\nHow to run the alpha compiler:\n./alpha [options] program\nValid options:\n-tok output the token number, token, line number, and column number for each of the tokens to the .tok file\n-st output the symbol table for the program to the .st file\n-help print this message and exit the alpha compiler\n" +//use to set flags for arg types +#define SET_FLAG 1 #include #include @@ -22,7 +17,23 @@ extern char *yytext; extern FILE *yyin; int arg; -int main(int argc, char* argv[]); +SymbolTable * top; +SymbolTable * curr; + +// int main(int argc, char* argv[]); char *is_tok(int argc, char* argv[]); -int is_alpha_file(char *file, int file_len); +// int is_alpha_file(char *file, int file_len); +void enter_scope(int, int); +void exit_scope(void); + +FILE *alpha_file; +FILE *tok_flag = NULL; +FILE *st_flag = NULL; +int no_flag = 0; + +int main(int argc, char* argv[]); +int new_file(char *arg, char *alpha); +int is_alpha_file(char *alpha, int file_len); bool is_help(char * input); +int run(FILE*alpha); +int check_flag(char * arg, char* alpha); diff --git a/symbol_table.c b/symbol_table.c new file mode 100644 index 0000000..a8592eb --- /dev/null +++ b/symbol_table.c @@ -0,0 +1,83 @@ +//Defining a symbol table +//Using a Linked List Structure. Head of linked List points to parent scope (if one exists) + +//Tail of Linked List points to a List of child scopes + + +#include "symbol_table.h" + +SymbolTable* CreateScope(SymbolTable* ParentScope, int Line, int Column){ + SymbolTable* table = (SymbolTable*)malloc(sizeof(SymbolTable)); + table->Line_Number = Line; + table->Column_Number = Column; + table->Parent_Scope = ParentScope; + table->Children_Scope = NULL; + table->entries = NULL; + if(ParentScope != NULL){ + if(ParentScope->Children_Scope == NULL){ + ListOfTable* newEntry = (ListOfTable*)malloc(sizeof(ListOfTable)); + newEntry->next = NULL; + //newEntry->prev = NULL; + newEntry->table = table; + ParentScope->Children_Scope = newEntry; + } else{ + ListOfTable* newEntry = (ListOfTable*)malloc(sizeof(ListOfTable)); + //newEntry->prev = NULL; + newEntry->table= table; + ListOfTable* oldEntry = ParentScope->Children_Scope; + ParentScope->Children_Scope = newEntry; + newEntry->next = oldEntry; + } + } + return table; + } + + +TableNode* CreateEntry(SymbolTable* table, char* typeOf, char* id){ + TableNode* newEntry = (TableNode*)malloc(sizeof(TableNode)); + newEntry->theType = typeOf; + newEntry->theName = id; + if(table->entries == NULL){ + table->entries = newEntry; + return newEntry; + } else{ + TableNode* oldEntry = table->entries; + table->entries = newEntry; + newEntry->next = oldEntry; + return newEntry; + } +} +TableNode * table_lookup(SymbolTable * table, char * x){ + TableNode * entrie = table->entries; + for(; entrie != NULL; entrie = entrie->next){ + if (!strcmp(entrie->theName, x)){ + return entrie; + } + } + return NULL; +} +TableNode * look_up(SymbolTable * table, char * x){ + if(table == NULL){ + return NULL; + } + TableNode * ret = table_lookup(table, x); + if (ret != NULL){ + return ret; + } + return look_up(table->Parent_Scope, x); +} + +//uncomment the below main function along with the headers above for a simple standalone test of table and entry creation + +/* +int main(){ + char* String = "STRING"; + char* X = "X"; + SymbolTable* Second = CreateScope(NULL, 2,2); + printf("Line number is %d, Column number of scope is %d\n",Second->Line_Number,Second->Column_Number); + TableNode* First_Entry = CreateEntry(Second,String,X); + + printf("The type of the first entry is %s\n",First_Entry->theType); + return 0; + } + */ diff --git a/symbol_table.h b/symbol_table.h new file mode 100644 index 0000000..4830345 --- /dev/null +++ b/symbol_table.h @@ -0,0 +1,29 @@ +#include +#include +#include +#include + +typedef struct ListOfTable{ + struct SymbolTable* table; + //struct ListOfTable* prev; + struct ListOfTable* next; + +}ListOfTable; + +typedef struct TableNode{ + char* theType; + char* theName; + struct TableNode* next; +}TableNode; + +typedef struct SymbolTable{ + TableNode* entries; + struct SymbolTable* Parent_Scope; + struct ListOfTable* Children_Scope; + int Line_Number; + int Column_Number; +}SymbolTable; + + +TableNode* CreateEntry(SymbolTable* table, char* typeOf, char* id); +SymbolTable* CreateScope(SymbolTable* ParentScope, int Line, int Column); diff --git a/tests/test_comment_issues.tok b/tests/test_comment_issues.tok new file mode 100644 index 0000000..e69de29 diff --git a/tests/test_wrong_type.txt b/tests/test_wrong_type.txt new file mode 100644 index 0000000..e69de29