test

.gitlab-ci.yml

@@ -1,3 +1,4 @@
+---
 lab_3_tests:
   stage: test
   image: gitlab.cs.wallawalla.edu:5050/cs_department/docker-images/cpp
@@ -12,3 +13,15 @@ lab_3_tests:
     - gcc main_test.c scanner.c -o scanner_test.out
     - echo "Lab 3 - Run Tests"
     - ./scanner_test.out
+
+lab_4_tests:
+  stage: test
+  image: gitlab.cs.wallawalla.edu:5050/cs_department/docker-images/cpp
+  script:
+    - apt update
+    - apt install -y flex bison
+    - bison --version
+    - cd lab-4
+    - echo "Lab 4 - Bison Parser"
+    - make test
+    - ./parser_test.out

lab-3/main.c

@@ -8,7 +8,7 @@ extern char *yytext;
 
 int main(int argc, char *argv[]) {
   FILE *file;
-  const char *filename = "samples/quadratic-snippet.cbl"; // Default filename
+  const char *filename = "samples/hello-world.cbl"; // Default filename
 
   // Check if a filename is provided as a command-line argument
   if (argc > 1) {

lab-3/main_test.c

@@ -18,30 +18,6 @@ struct token_st {
   char *p;
 };
 
-// UTEST(scanner, identifier) {
-//   token_t t;
-//   // Must include the null character to terminate input
-//   char string[] = "test\0"; 
-//   YY_BUFFER_STATE buffer = yy_scan_buffer(string, sizeof(string));
-
-//   ASSERT_EQ(TOKEN_EOF, (t = yylex()));
-//   ASSERT_STREQ("", yytext);
-
-//   yy_delete_buffer(buffer);
-// }
-
-// UTEST(scanner, assignment) {
-//   token_t t;
-//   // Must include the null character to terminate input
-//   char string[] = "=\0"; 
-//   YY_BUFFER_STATE buffer = yy_scan_buffer(string, sizeof(string));
-
-//   ASSERT_EQ(TOKEN_EOF, (t = yylex()));
-//   ASSERT_STREQ("", yytext);
-
-//   yy_delete_buffer(buffer);
-// }
-
 UTEST(scanner, hello) {
   struct token_st tokens[] = {
     {TOKEN_IDENTIFICATION, "IDENTIFICATION"},
@@ -64,12 +40,11 @@ UTEST(scanner, hello) {
 
 
   yyin = fopen("samples/hello-world.cbl", "r");
-  yyrestart(yyin);
   ASSERT_TRUE(yyin);
-
   int index = 0;
   token_t t;
   do {
+    printf("index: %d token: %d text: %s\n", index, t, yytext);
     ASSERT_EQ(tokens[index].t, (t = yylex()));
     ASSERT_STREQ(tokens[index].p, yytext);
     ++index;

lab-3/scanner.flex

@@ -3,7 +3,6 @@
 %}
 NAME [a-zA-Z]([a-zA-Z0-9_-]*[a-zA-Z0-9])?
 DIGIT [0-9]+
-
 %%
 
 (" "|\t|\n) /* skip whitespace */

lab-3/token.h

@@ -1,6 +1,5 @@
 typedef enum {
   TOKEN_EOF = 0,
-  // Identification Keywords
   TOKEN_IDENTIFICATION,
   TOKEN_KEYWORD_DIVISION,
   TOKEN_KEYWORD_DATA,
@@ -8,8 +7,6 @@ typedef enum {
   TOKEN_PROGRAM_ID,
   TOKEN_WORKING_STORAGE,
   TOKEN_PROCEDURE,
-
-  // Program Keywords
   TOKEN_DISPLAY,
   TOKEN_STOP,
   TOKEN_RUN,
@@ -28,10 +25,7 @@ typedef enum {
   TOKEN_KEYWORD_VALUE,
   TOKEN_KEYWORD_COMPUTE,
   TOKEN_KEYWORD_FUNCTION,
-
-  // Identifiers
   TOKEN_IDENT,
-  // Data types
   TOKEN_STRING,
   TOKEN_INTEGER,
   TOKEN_PICTURE,
@@ -39,19 +33,14 @@ typedef enum {
   TOKEN_NUMERIC,
   TOKEN_SIGNED_NUMERIC,
   TOKEN_IMPLIED_DECIMAL,
-  // https://ibmmainframes.com/about393.html
   TOKEN_COMPUTATION_LEVEL_0,
   TOKEN_COMPUTATION_LEVEL_1,
   TOKEN_COMPUTATION_LEVEL_2,
   TOKEN_COMPUTATION_LEVEL_3,
-
-  // Grammar
   TOKEN_LEFT_PARENTHESIS,
   TOKEN_RIGHT_PARENTHESIS,
   TOKEN_DOT,
   TOKEN_COMMENT,
-
-  // Operators
   TOKEN_ADD,
   TOKEN_SUB,
   TOKEN_MULTIPLY,

lab-4/Makefile

@@ -0,0 +1,29 @@
+
+# The top level rule indicates how to link everything together into main
+
+main: main.o scanner.o parser.o
+	gcc main.o scanner.o parser.o -o parser.out -lm
+
+test: main_test.o scanner.o parser.o
+	gcc main_test.o scanner.o parser.o -o parser_test.out -lm
+
+# Only the files generated by flex and bison need explicit rules.
+
+scanner.c: scanner.flex parser.h
+	flex -oscanner.c scanner.flex
+
+parser.c parser.h: parser.bison
+	bison --defines=token.h --output=parser.c -v parser.bison
+
+# This pattern indicates that any .o file depends
+# upon the .c file of the same name, and all of the .h files.
+# So, if a .o file is needed, it is built automatically.
+
+%.o: %.c *.h
+	gcc -Wall -c $< -o $@
+
+# clean causes all intermediate files to be deleted.
+
+clean:
+	rm -f parser.c parser.output token.h scanner.c *.o parser.out parser_test.out
+

lab-4/README.md

@@ -0,0 +1,98 @@
+# Bison Parser With Tests
+
+Create a scanner for your sample programs with tests.
+The tests should include basic statements and the sample files for your language.
+See the [example lab for Python](https://gitlab.cs.wallawalla.edu/cptr354/language-interpreter-lab-python).
+
+## Build Process
+
+To build the `scanner.c` run the following **flex** command.
+
+```sh
+flex -o scanner.c scanner.flex
+```
+
+The parser is built using **bison**.
+Note this command also builds the `token.h`.
+
+```sh
+bison --defines=token.h --output=parser.c -v parser.bison
+```
+
+THe `-v` flag will create a `parser.output` which represents the grammar in text form.
+
+
+### Running the Parser
+
+Then build the `main.c`, `parcer.c` and `scanner.c` using **gcc**.
+Note that you may get a few warnings that can be ignored.
+
+```sh
+gcc main.c scanner.c  parser.c -o parser.out
+```
+
+Then execute the test suite using `./parser.out` and pass in the input from your sample program.
+For example here is a command for `samples/program.c`
+
+```sh
+./parser.out < samples/program.c
+```
+
+
+### Testing the Parser
+
+Then build the `main_test.c`, `parcer.c` and `scanner.c` using **gcc**.
+Note that you may get a few warnings that can be ignored.
+
+```sh
+gcc main_test.c scanner.c  parser.c -o parser_test.out
+```
+
+Then execute the test suite using `./parser_test.out`.
+
+```sh
+./parser_test.out
+```
+
+
+### Using Make
+
+Alternatively you can use the supplied make file.
+
+* Build main program - `make`
+* Build test program - `make test`
+* Clean up build files - `make clean`
+
+
+## Tests
+
+* Write test for various statements in your language 
+  * assignment
+  * print
+  * mathmatical expressions
+  * boolean expressions
+  * branching
+  * looping
+* Write a test for each sample test file.
+  * Hello World!
+  * Quadratic
+  * Sorting
+
+
+## Grading Rubric
+
+The grading will be over the following categories.
+
+Points      Description
+----------- ------------------------------------
+60 points   Ability to parse — Six different language statements
+10 points   Ability to parse — Hello world
+10 points   Ability to parse — Quadratic equation
+10 points   Ability to parse — Integer sorting
+10 points   Report formatting and readability
+
+
+## Turn In
+
+Please submit the URL of your gitlab project to D2L's Brightspace with your report.
+On gitlab update the `lab-4` folder to have your report and code snippets before the due date.

lab-4/main.c

@@ -0,0 +1,11 @@
+#include <stdio.h>
+
+extern int yyparse();
+
+int main() {
+  if (yyparse() == 0) {
+    printf("Parse successful!\n");
+  } else {
+    printf("Parse failed.\n");
+  }
+}

lab-4/main_test.c

@@ -0,0 +1,57 @@
+// https://github.com/sheredom/utest.h/blob/master/utest.h
+#include "utest.h"
+#include <stdio.h>
+
+extern int yyparse();
+
+typedef struct yy_buffer_state *YY_BUFFER_STATE;
+extern int yyrestart();
+extern YY_BUFFER_STATE yy_scan_buffer(char *str, int i);
+extern YY_BUFFER_STATE yy_scan_string(char *str);
+extern void yy_delete_buffer(YY_BUFFER_STATE buffer);
+extern FILE *yyin;
+extern int yylineno;
+
+UTEST_MAIN();
+
+
+UTEST(parser, math) {
+  // Must include the null character to terminate input
+  char string[] = "1+8/4-3;\0"; 
+  YY_BUFFER_STATE buffer = yy_scan_buffer(string, sizeof(string));
+
+  yylineno = 1;
+  int result = yyparse();
+
+  yy_delete_buffer(buffer);
+
+  // Assert the result to test correctness
+  ASSERT_EQ(result, 0); 
+}
+
+UTEST(parser, missing_semi_colon) {
+  // Must include the null character to terminate input
+  char string[] = "1+8/4-3\0"; 
+  YY_BUFFER_STATE buffer = yy_scan_buffer(string, sizeof(string));
+
+  yylineno = 1;
+  int result = yyparse();
+
+  yy_delete_buffer(buffer);
+
+  // Assert the result to test correctness
+  ASSERT_EQ(result, 1); 
+}
+
+UTEST(parser, sample) {
+  // Read sample file as input
+  yyin = fopen("samples/program.c", "r");
+  yyrestart(yyin);
+  ASSERT_TRUE(yyin);
+
+  yylineno = 1;
+  int result = yyparse();
+
+  // Assert the result to test correctness
+  ASSERT_EQ(result, 0); 
+}

lab-4/parser.bison

@@ -0,0 +1,82 @@
+%{
+#define YYDEBUG 1
+#include <stdio.h>
+extern int yylineno;
+void yyerror(const char*);
+int yylex();
+%}
+
+%debug
+%define parse.error detailed
+
+%token TOKEN_EOF
+%token TOKEN_IDENTIFICATION
+%token TOKEN_KEYWORD_DIVISION
+%token TOKEN_KEYWORD_DATA
+%token TOKEN_KEYWORD_SECTION
+%token TOKEN_PROGRAM_ID
+%token TOKEN_WORKING_STORAGE
+%token TOKEN_PROCEDURE
+%token TOKEN_STOP
+%token TOKEN_RUN
+%token TOKEN_MOVE
+%token TOKEN_KEYWORD_TO
+%token TOKEN_PERFORM
+%token TOKEN_VARYING
+%token TOKEN_KEYWORD_FROM
+%token TOKEN_KEYWORD_BY
+%token TOKEN_UNTIL
+%token TOKEN_END_PERFORM
+%token TOKEN_IF
+%token TOKEN_END_IF
+%token TOKEN_SPACE
+%token TOKEN_KEYWORD_OCCURS
+%token TOKEN_KEYWORD_VALUE
+%token TOKEN_KEYWORD_COMPUTE
+%token TOKEN_KEYWORD_FUNCTION
+%token TOKEN_IDENT
+%token TOKEN_STRING
+%token TOKEN_INTEGER
+%token TOKEN_PICTURE
+%token TOKEN_ALPHANUMERIC
+%token TOKEN_NUMERIC
+%token TOKEN_SIGNED_NUMERIC
+%token TOKEN_IMPLIED_DECIMAL
+%token TOKEN_COMPUTATION_LEVEL_0
+%token TOKEN_COMPUTATION_LEVEL_1
+%token TOKEN_COMPUTATION_LEVEL_2
+%token TOKEN_COMPUTATION_LEVEL_3
+%token TOKEN_LEFT_PARENTHESIS
+%token TOKEN_RIGHT_PARENTHESIS
+%token TOKEN_DOT
+%token TOKEN_COMMENT
+%token TOKEN_ADD
+%token TOKEN_SUB
+%token TOKEN_MULTIPLY
+%token TOKEN_DIVIDE
+%token TOKEN_EQUAL
+%token TOKEN_GREATER_THAN
+%token TOKEN_LESS_THAN
+%token TOKEN_EXPONENTIAL
+
+%%
+program : expr TOKEN_SEMI;
+
+expr : expr TOKEN_PLUS term
+    | expr TOKEN_MINUS term
+    | term
+    ;
+
+term : term TOKEN_MUL factor
+    | term TOKEN_DIV factor
+    | factor
+    ;
+
+factor: TOKEN_MINUS factor
+    | TOKEN_LPAREN expr TOKEN_RPAREN
+    | TOKEN_INT
+    ;
+%%
+void yyerror(const char* msg) {
+    fprintf(stderr, "Error | Line: %d\n%s\n",yylineno,msg);
+}

lab-4/samples/hello-world.cbl

@@ -0,0 +1,5 @@
+IDENTIFICATION DIVISION.
+PROGRAM-ID. HELLO-WORLD.
+PROCEDURE DIVISION.
+    DISPLAY 'Hello World!'
+STOP RUN.

lab-4/samples/quadratic-snippet.cbl

@@ -0,0 +1,35 @@
+       *> Code altered from https://www.quora.com/What-is-a-COBOL-program-that-will-solve-a-quadratic-equation
+       *> Program finds the roots to a simple quadratic equation
+
+       IDENTIFICATION DIVISION.
+       PROGRAM-ID. QuadraticSolver.
+       DATA DIVISION.
+           WORKING-STORAGE SECTION.
+           77 a PIC S9(5)V9(5) COMP-3 VALUE 1.
+           77 b PIC S9(5)V9(5) COMP-3 VALUE 5.
+           77 c PIC S9(5)V9(5) COMP-3 VALUE 6.
+           77 discriminant PIC S9(5)V9(5) COMP-3.
+           77 root1 PIC S9(5)V9(5) COMP-3.
+           77 root2 PIC S9(5)V9(5) COMP-3.
+           77 square-root-discriminant PIC S9(5)V9(5) COMP-3.
+
+       PROCEDURE DIVISION. *> program begins here
+           DISPLAY "EQUATION: (1x^2) + 5x + 6 = 0" 
+           COMPUTE discriminant = (b ** 2) - (4 * a * c)
+
+           IF discriminant > 0
+               COMPUTE square-root-discriminant = FUNCTION SQRT(discriminant)
+               COMPUTE root1 = (-b + square-root-discriminant) / (2 * a)
+               COMPUTE root2 = (-b - square-root-discriminant) / (2 * a)
+               DISPLAY "The equation has two distinct real roots: "
+               DISPLAY "Root 1: " root1
+               DISPLAY "Root 2: " root2
+
+           ELSE IF discriminant = 0
+               COMPUTE root1 = -b / (2 * a)
+               DISPLAY "The equation has one real root: "
+               DISPLAY "Root: " root1
+           ELSE
+               DISPLAY "The equation has no real roots."
+
+       STOP RUN.

lab-4/samples/sorting-snippet.cbl

@@ -0,0 +1,55 @@
+IDENTIFICATION DIVISION.
+PROGRAM-ID. sorting.
+DATA DIVISION.
+WORKING-STORAGE SECTION.
+    01  WS-SORT-AREA.
+    05  WS-SORT-TABLE.
+        10  WS-SORT-ROW PIC X(10) OCCURS 100.
+    05  WS-TEMP-ROW     PIC X(10).
+    05  WS-ROW-MAX      PIC S9(4) COMP VALUE 100.
+    05  WS-SORT-MAX     PIC S9(4) COMP.
+    05  WS-I            PIC S9(4) COMP.
+    05  WS-J            PIC S9(4) COMP.
+    05  WS-INDEX        PIC S9(4) COMP.
+
+PROCEDURE DIVISION.
+*> Initialize test data
+    MOVE "30" TO WS-SORT-ROW(1)
+    MOVE "10" TO WS-SORT-ROW(2)
+    MOVE "50" TO WS-SORT-ROW(3)
+    MOVE "20" TO WS-SORT-ROW(4)
+    MOVE "40" TO WS-SORT-ROW(5)
+    MOVE 5 TO WS-SORT-MAX
+
+*> * Display original array
+    DISPLAY "Original Array Contents:"
+    DISPLAY "---------------------"
+    PERFORM VARYING WS-INDEX FROM 1 BY 1
+        UNTIL WS-INDEX > WS-SORT-MAX
+        DISPLAY "Element " WS-INDEX ": " WS-SORT-ROW(WS-INDEX)
+    END-PERFORM
+    DISPLAY SPACE
+
+*> * Simplified bubble sort
+    PERFORM VARYING WS-I FROM 1 BY 1 
+        UNTIL WS-I > WS-SORT-MAX - 1
+        PERFORM VARYING WS-J FROM 1 BY 1 
+            UNTIL WS-J > WS-SORT-MAX - WS-I
+            IF WS-SORT-ROW(WS-J) > WS-SORT-ROW(WS-J + 1)
+                MOVE WS-SORT-ROW(WS-J) TO WS-TEMP-ROW
+                MOVE WS-SORT-ROW(WS-J + 1) TO WS-SORT-ROW(WS-J)
+                MOVE WS-TEMP-ROW TO WS-SORT-ROW(WS-J + 1)
+            END-IF
+        END-PERFORM
+    END-PERFORM
+
+*> * Display sorted array
+    DISPLAY "Sorted Array Contents:"
+    DISPLAY "--------------------"
+    PERFORM VARYING WS-INDEX FROM 1 BY 1
+        UNTIL WS-INDEX > WS-SORT-MAX
+        DISPLAY "Element " WS-INDEX ": " WS-SORT-ROW(WS-INDEX)
+    END-PERFORM.
+
+    STOP RUN.
+    

lab-4/scanner.flex

@@ -0,0 +1,25 @@
+%{
+#include "token.h"
+%}
+
+%option nounput
+%option noinput
+%option yylineno
+
+DIGIT [0-9]
+LETTER [a-zA-Z]
+
+%%
+(" "|\t|\n) /* skip whitespace */
+\+ { return TOKEN_PLUS; }
+\- { return TOKEN_MINUS; }
+\* { return TOKEN_MUL; }
+\/ { return TOKEN_DIV; }
+\( { return TOKEN_LPAREN; }
+\) { return TOKEN_RPAREN; }
+\; { return TOKEN_SEMI; }
+{DIGIT}+ { return TOKEN_INT; }
+. { return TOKEN_ERROR; }
+%%
+
+int yywrap() { return 1; }