mah lexerHEADmaster

3 files changed, 349 insertions, 0 deletions

diff --git a/driver.c b/driver.c
new file mode 100644
index 0000000..ebee4db
--- /dev/null
+++ b/driver.c
@@ -0,0 +1,180 @@
+#include <stdlib.h>
+#include "lex.yy.c"
+#include "token.h"
+
+/* yyline = 1 for numbering lines 1 to n */
+int yyline = 1;
+int yycolumn = 0;
+int yylength = 0;
+int yylval = 0;
+
+/* the string table is one block separated by null terminators
+ * an int array stores the index of each string
+ * access the strings with pointer arithmetic */
+char *stringTable;
+int stringTableIndices[LIMIT2];
+int nStringsStored = 0;
+int nCharsStored = 0;
+
+int main(int argc, char *argv[])
+{
+    int lexReturn;
+    int i;
+    
+    /* initialize string table */
+    stringTable = (char *) malloc(LIMIT1 + 1);
+    strcpy(stringTable,"");
+    
+    printf("Line      Column      Token         Index in String Table\n");
+    
+    /* get tokens and print table */
+    do {
+        
+        lexReturn = yylex();
+        
+        /* printing line and column here so the switch cases are less wordy */
+        if (lexReturn != EOFnum) {
+            printf("%-8d  %-12d", yyline, yycolumn);
+        }
+        
+        switch (lexReturn) {
+            case ANDnum:
+                printf("%-8s\n","ANDnum");
+                break;
+            case ASSGNnum:
+                printf("%-8s\n","ASSGNnum");
+                break;
+            case DECLARATIONSnum:
+                printf("%-8s\n","DECLARATIONSnum");
+                break;
+            case DOTnum:
+                printf("%-8s\n","DOTnum");
+                break;
+            case ENDDECLARATIONSnum:
+                printf("%-8s\n","ENDDECLARATIONSnum");
+                break;
+            case EQUALnum:
+                printf("%-8s\n","EQUALnum");
+                break;
+            case GTnum:
+                printf("%-8s\n","GTnum");
+                break;
+            case IDnum:
+                printf("%-8s      %d\n","IDnum", yylval);
+                break;
+            case INTnum:
+                printf("%-8s\n","INTnum");
+                break;
+            case LBRACnum:
+                printf("%-8s\n","LBRACnum");
+                break;
+            case LPARENnum:
+                printf("%-8s\n","LPARENnum");
+                break;
+            case METHODnum:
+                printf("%-8s\n","METHODnum");
+                break;
+            case NEnum:
+                printf("%-8s\n","NEnum");
+                break;
+            case ORnum:
+                printf("%-8s\n","ORnum");
+                break;
+            case PROGRAMnum:
+                printf("%-8s\n","PROGRAMnum");
+                break;
+            case RBRACnum:
+                printf("%-8s\n","RBRACnum");
+                break;
+            case RPARENnum:
+                printf("%-8s\n","RPARENnum");
+                break;
+            case SEMInum:
+                printf("%-8s\n","SEMInum");
+                break;
+            case VALnum:
+                printf("%-8s\n","VALnum");
+                break;
+            case WHILEnum:
+                printf("%-8s\n","WHILEnum");
+                break;
+            case CLASSnum:
+                printf("%-8s\n","CLASSnum");
+                break;
+            case COMMAnum:
+                printf("%-8s\n","COMMAnum");
+                break;
+            case DIVIDEnum:
+                printf("%-8s\n","DIVIDEnum");
+                break;
+            case ELSEnum:
+                printf("%-8s\n","ELSEnum");
+                break;
+            case EQnum:
+                printf("%-8s\n","EQnum");
+                break;
+            case GEnum:
+                printf("%-8s\n","GEnum");
+                break;
+            case ICONSTnum:
+                printf("%-8s\n","ICONSTnum");
+                break;
+            case IFnum:
+                printf("%-8s\n","IFnum");
+                break;
+            case LBRACEnum:
+                printf("%-8s\n","LBRACEnum");
+                break;
+            case LEnum:
+                printf("%-8s\n","LEnum");
+                break;
+            case LTnum:
+                printf("%-8s\n","LTnum");
+                break;
+            case MINUSnum:
+                printf("%-8s\n","MINUSnum");
+                break;
+            case NOTnum:
+                printf("%-8s\n","NOTnum");
+                break;
+            case PLUSnum:
+                printf("%-8s\n","PLUSnum");
+                break;
+            case RBRACEnum:
+                printf("%-8s\n","RBRACEnum");
+                break;
+            case RETURNnum:
+                printf("%-8s\n","RETURNnum");
+                break;
+            case SCONSTnum:
+                printf("%-8s     %d\n", "SCONSTnum", yylval);
+                break;
+            case TIMESnum:
+                printf("%-8s\n","TIMESnum");
+                break;
+            case VOIDnum:
+                printf("%-8s\n","VOIDnum");
+                break;
+            case EOFnum:
+                printf("%28s\n","EOFnum");
+                break;
+            default:
+                printf("%-8s\n","bad token");
+        }
+
+    } while (lexReturn != 0);
+    
+    /* retrieving string table */
+    printf("\nString Table: ");
+    if (nCharsStored > 0) {
+        for (i=0;i<nStringsStored;i++) {
+            printf("%s ",stringTable + stringTableIndices[i]);
+        }
+    }
+    
+    printf("\n\nEnd of File\n");
+    
+    free(stringTable);
+
+return 0;
+}
diff --git a/lexer.l b/lexer.l
new file mode 100644
index 0000000..dc499e6
--- /dev/null
+++ b/lexer.l
@@ -0,0 +1,123 @@
+%{
+#include <stdlib.h>
+#include "token.h"
+
+extern int yycolumn, yylength, yyline, yylval;
+
+extern char *stringTable;
+extern int stringTableIndices[];
+extern int nStringsStored;
+extern int nCharsStored;
+
+/* functions */
+void ReportError(char *, int, int);
+int getStrTableIndex(char *, int, int);
+void RemoveQuotes(char *);
+
+%}
+
+/* regular definitions */
+
+id            ([a-zA-Z])+
+delim         [ \t]
+ws            {delim}+
+newline       [\n]
+
+%%
+{newline}                                                           {yyline++; yycolumn=0;}
+{ws}                                                                {yycolumn+=yyleng;}
+"/*"([^*]|[ \t\n]|(\*+([^*/]|[ \t\n])))*                            {ReportError("comment doesn't terminate",yyline,yycolumn);}
+"/*"([^*]|[ \t\n]|(\*+([^*/]|[ \t\n])))*"*"+"/"                     {yycolumn+=yyleng;};
+
+"&&"                                                                {yycolumn+=yyleng; return (ANDnum);}
+":="                                                                {yycolumn+=yyleng; return (ASSGNnum);}
+[dD][eE][cC][lL][aA][rR][aA][tT][iI][oO][nN][sS]                    {yycolumn+=yyleng; return (DECLARATIONSnum);}
+"."                                                                 {yycolumn+=yyleng; return (DOTnum);}
+[eE][nN][dD][dD][eE][cC][lL][aA][rR][aA][tT][iI][oO][nN][sS]        {yycolumn+=yyleng; return (ENDDECLARATIONSnum);}
+"="                                                                 {yycolumn+=yyleng; return (EQUALnum);}
+">"                               {yycolumn+=yyleng; return (GTnum);}
+[iI][nN][tT]                      {yycolumn+=yyleng; return (INTnum);}
+"["                               {yycolumn+=yyleng; return (LBRACnum);}
+"("                               {yycolumn+=yyleng; return (LPARENnum);}
+[mM][eE][tT][hH][oO][dD]          {yycolumn+=yyleng; return (METHODnum);}
+"!="                              {yycolumn+=yyleng; return (NEnum);}
+"||"                              {yycolumn+=yyleng; return (ORnum);}
+[pP][rR][oO][gG][rR][aA][mM]      {yycolumn+=yyleng; return (PROGRAMnum);}
+"]"                               {yycolumn+=yyleng; return (RBRACnum);}
+")"                               {yycolumn+=yyleng; return (RPARENnum);}
+";"                               {yycolumn+=yyleng; return (SEMInum);}
+[vV][aA][lL]                      {yycolumn+=yyleng; return (VALnum);}
+[wW][hH][iI][lL][eE]              {yycolumn+=yyleng; return (WHILEnum);}
+[cC][lL][aA][sS][sS]              {yycolumn+=yyleng; return (CLASSnum);}
+","                               {yycolumn+=yyleng; return (COMMAnum);}
+"/"                               {yycolumn+=yyleng; return (DIVIDEnum);}
+[eE][lL][sS][eE]                  {yycolumn+=yyleng; return (ELSEnum);}
+"=="                              {yycolumn+=yyleng; return (EQnum);}
+">="                              {yycolumn+=yyleng; return (GEnum);}
+[iI][fF]                          {yycolumn+=yyleng; return (IFnum);}
+"{"                               {yycolumn+=yyleng; return (LBRACEnum);}
+"<="                              {yycolumn+=yyleng; return (LEnum);}
+"<"                               {yycolumn+=yyleng; return (LTnum);}
+"-"                               {yycolumn+=yyleng; return (MINUSnum);}
+"!"                               {yycolumn+=yyleng; return (NOTnum);}
+"+"                               {yycolumn+=yyleng; return (PLUSnum);}
+"}"                               {yycolumn+=yyleng; return (RBRACEnum);}
+[rR][eE][tT][uU][rR][nN]          {yycolumn+=yyleng; return (RETURNnum);}
+"*"                               {yycolumn+=yyleng; return (TIMESnum);}
+[vV][oO][iI][dD]                  {yycolumn+=yyleng; return (VOIDnum);}
+
+'(.|[ \t])*'                      {RemoveQuotes(yytext); yylval = getStrTableIndex(yytext,yyline,yycolumn); yycolumn+=yyleng; return (SCONSTnum);}
+'(.|[ \t\n])*'                    {ReportError("string constant crosses line boundary",yyline,yycolumn); yyline++; yycolumn=0;}
+[1-9][0-9]*                       {yylval = atoi(yytext); yycolumn+=yyleng; return (ICONSTnum);}
+[0-9][a-zA-Z0-9]*                 {ReportError("identifier starting with a digit",yyline,yycolumn); yycolumn+=yyleng;}
+[a-zA-Z][a-zA-Z0-9]*              {yylval = getStrTableIndex(yytext,yyline,yycolumn); yycolumn+=yyleng; return (IDnum);}
+
+
+%%
+
+void ReportError(char *message,int line, int column) {
+    fprintf( stderr, "\nError at line %d, column %d (first char): %s\n\n", line, column, message);
+    return;
+}
+
+int getStrTableIndex(char *string, int line, int column) {
+    int i;
+    
+    /* use index array to search table for string*/
+    for (i=0; i < nStringsStored; i++) {
+        /* return index if found */
+        if (strcmp(string, stringTable + stringTableIndices[i]) == 0) {
+            return stringTableIndices[i];
+        }
+    }
+    
+    /* error if string table is full */
+    if (nStringsStored == LIMIT1) {
+        ReportError("can't add string to string table: max character count.",line, column);
+        return -1;
+    }
+    
+    /* error if index table is full */
+    if (nStringsStored == LIMIT2) {
+        ReportError("can't add string to string table: max number of unique strings",line, column);
+        return -1;
+    }
+    
+    /* add string to table if not full and return index */
+    stringTableIndices[i] = nCharsStored;
+    memcpy(stringTable + stringTableIndices[i],string,strlen(string)+1);
+    nCharsStored+=strlen(string)+1;
+    nStringsStored++;
+    
+    return stringTableIndices[i];
+}
+
+/* this lexer detects string constants by their quotes, so yytext includes them
+ * even though they aren't really part of the string 
+ * so this removes them */
+void RemoveQuotes(char *string) {
+    memmove(string, string+1, strlen(string));
+    string[strlen(string) - 1] = '\0';
+    
+    return;
+}
diff --git a/token.h b/token.h
new file mode 100644
index 0000000..20bfa3f
--- /dev/null
+++ b/token.h
@@ -0,0 +1,46 @@
+/* limit on total characters */
+#define LIMIT1 5000
+
+/* limit on number of unique strings */
+#define LIMIT2 1000
+
+#define	ANDnum			257
+#define ASSGNnum		258
+#define DECLARATIONSnum		259
+#define DOTnum			260
+#define ENDDECLARATIONSnum	261
+#define EQUALnum		262
+#define GTnum			263
+#define IDnum			264
+#define INTnum			265
+#define LBRACnum		266
+#define LPARENnum		267
+#define METHODnum		268
+#define NEnum			269
+#define ORnum			270
+#define PROGRAMnum		271
+#define RBRACnum		272
+#define RPARENnum		273
+#define SEMInum			274
+#define VALnum			275
+#define WHILEnum		276
+#define CLASSnum		277
+#define COMMAnum		278
+#define DIVIDEnum		279
+#define ELSEnum			280
+#define EQnum			281
+#define GEnum			282
+#define ICONSTnum		283
+#define IFnum			284
+#define LBRACEnum		285
+#define LEnum			286
+#define LTnum			287
+#define MINUSnum		288
+#define NOTnum			289
+#define PLUSnum			290
+#define RBRACEnum		291
+#define RETURNnum		292
+#define SCONSTnum		293
+#define TIMESnum		294
+#define VOIDnum			295
+#define EOFnum			0