M projects/06/assembler1/assembler1.c → projects/06/assembler1/assembler1.c

@@ -6,11 +6,15 @@ #include <string.h>
 
 #include "../bindump.h"
 
-#define DEBUG(...)              printf(__VA_ARGS__)
+//#define DEBUG(...)              printf(__VA_ARGS__)
+#define DEBUG(...)
 #define die(err_msg)            perror(err_msg); exit(-1)
 #define alert(...)              fprintf(stderr, __VA_ARGS__)
 
 #define MAX_LINE_LEN            256
+
+char *g_asm_line;             // copy of currently-read line for easy reference
+size_t g_asm_line_number;     // current line number
 
 
 static uint32_t myatoi(const char *a_field_str)
@@ -76,6 +80,8 @@ */
 static bool parse_c_type_dest(const char *dest_line, uint8_t *dest)
 {
 	size_t len;
+	char *err_dest_fmt = "syntax error: destination format incorrect\n";
+	DEBUG("dest_line: %s\n", dest_line);
 
 	for (len = 0; dest_line[len] != '='; ++len) {}  // read until '='
 
@@ -84,7 +90,8 @@ if (dest_line[0] == 'A' && dest_line[1] == 'D' &&
 		    dest_line[2] == 'M') {
 			*dest = 0x7;
 		} else {
-			alert("syntax error: destination format incorrect\n");
+			alert("%s", err_dest_fmt);
+			alert("%lu | %s\n", g_asm_line_number, g_asm_line);
 			return false;
 		}
 	} else if (len == 2) {
@@ -95,7 +102,8 @@ *dest = 0x5;
 		} else if (dest_line[0] == 'A' && dest_line[1] == 'D') {
 			*dest = 0x6;
 		} else {
-			alert("syntax error: destination format incorrect\n");
+			alert("%s", err_dest_fmt);
+			alert("%lu | %s\n", g_asm_line_number, g_asm_line);
 			return false;
 		}
 	} else if (len == 1) {
@@ -110,7 +118,8 @@ case 'A':
 			*dest = 0x4;
 			break;
 		default:
-			alert("syntax error: destination format incorrect\n");
+			alert("%s", err_dest_fmt);
+			alert("%lu | %s\n", g_asm_line_number, g_asm_line);
 			return false;
 		}
 	} else {
@@ -122,13 +131,115 @@ return true;
 }
 
 /* returns comp bits 0b0acccccc
+ * TODO: refactor (especially the error cases, which are too repetitive)
  */
 static bool parse_c_type_comp(const char *comp_line, uint8_t *comp)
 {
 	size_t len;
 	DEBUG("comp_line: %s\n", comp_line);
 
-	*comp = 0x53;  // STUB: 0b01010011
+	for (len = 0; comp_line[len] == '0' || comp_line[len] == '1' ||
+	              comp_line[len] == '-' || comp_line[len] == 'D' ||
+	              comp_line[len] == 'A' || comp_line[len] == 'M' ||
+	              comp_line[len] == '!' || comp_line[len] == '+' ||
+	              comp_line[len] == '&' || comp_line[len] == '|'; ++len) {}
+	if (len == 1) {  // 0 1 D A M
+		switch (comp_line[0]) {
+		case '0': *comp = 0x2a; break; // 0 101010
+		case '1': *comp = 0x3f; break; // 0 111111
+		case 'D': *comp = 0x0c; break; // 0 001100
+		case 'A': *comp = 0x30; break; // 0 110000
+		case 'M': *comp = 0x70; break; // 1 110000
+		default:
+			alert("syntax error: comp field incorrect value\n");
+			return false;
+		}
+	} else if (len == 2) {  // -1 !D !A !M -D -A -M 
+		if (comp_line[0] == '-') {
+			switch (comp_line[1]) {
+			case '1': *comp = 0x3a; break; // 0 111010
+			case 'D': *comp = 0x0f; break; // 0 001111
+			case 'A': *comp = 0x33; break; // 0 110011
+			case 'M': *comp = 0x73; break; // 1 110011
+			default:
+				alert("syntax error: comp field incorrect value\n");
+				return false;
+			}
+		} else if (comp_line[0] == '!') {
+			switch (comp_line[1]) {
+			case 'D': *comp = 0x0d; break; // 0 001101
+			case 'A': *comp = 0x31; break; // 0 110001
+			case 'M': *comp = 0x71; break; // 1 110001
+			default:
+				alert("syntax error: comp field incorrect value\n");
+				return false;
+			}
+		} else {
+			alert("syntax error: comp field incorrect value\n");
+			return false;
+		}
+	} else if (len == 3) {
+		if (comp_line[0] == 'D') {
+			if (comp_line[2] == '1') {  // D+1 D-1
+				if (comp_line[1] == '+') {
+					*comp = 0x1f; // 0 011111
+				} else if (comp_line[1] == '-') {
+					*comp = 0x0e; // 0 001110
+				} else {
+					alert("syntax error: comp field "
+					      "incorrect value\n");
+					return false;
+				}
+			} else {             // D+A D+M D-A D-M D&A D&M D|A D|M
+				if (comp_line[2] == 'M') {
+					*comp = 0x40;              // _1_ 000000
+				} else if (comp_line[2] == 'A') {
+					*comp = 0x00;              // _0_ 000000
+				} else {
+					alert("syntax error: comp field "
+					      "incorrect value\n");
+					return false;
+				}
+
+				switch (comp_line[1]) {
+				case '+': *comp |= 0x02; break; // 000010
+				case '-': *comp |= 0x13; break; // 010011
+				case '&': *comp |= 0x00; break; // 000000
+				case '|': *comp |= 0x15; break; // 010101
+				default:
+					alert("syntax error: comp field "
+					      "incorrect value\n");
+					return false;
+				}
+			}
+		} else {
+			// A+1 M+1 A-1 M-1 A-D M-D
+			if (comp_line[0] == 'M') {
+				*comp = 0x40;              // _1_ 000000
+			} else if (comp_line[0] == 'A') {
+				*comp = 0x00;              // _0_ 000000
+			} else {
+				alert("syntax error: comp field incorrect value\n");
+				return false;
+			}
+
+			if (comp_line[1] == '+' && comp_line[2] == '1') {
+				*comp |= 0x37;  // 1 110111
+			} else if (comp_line[1] == '-' && comp_line[2] == '1') {
+				*comp |= 0x32;  // 1 110010
+			} else if (comp_line[1] == '-' && comp_line[2] == 'D') {
+				*comp |= 0x07;  // 1 000111
+			} else {
+				alert("syntax error: comp field incorrect value\n");
+				return false;
+			}
+		}
+	} else {
+		alert("syntax error: comp field incorrect length %lu\n", len);
+		return false;
+	}
+
+	//*comp = 0x53;  // STUB: 0b01010011
 	return true;
 }
 
@@ -206,6 +317,7 @@ }
 
 /* Instruction format: 0b111accccccdddjjj
  * Assumes line begins with actual instruction (prepended whitespace stripped)
+ * TODO: eventually just replace all wasteful c-instruction parsing w/ strcmp()
  */
 static bool parse_c_type(const char *line, uint16_t *instruction)
 {
@@ -230,8 +342,8 @@ dest_start = &line[0];   // start of line
 				// this 'i+1' might be dangerous!
 				comp_start = &line[i+1]; // after "[dest]="
 			} else {
-				alert("syntax error: destination field"
-				      " incorrect length\n");
+				alert("syntax error: jump field incorrect "
+				      "length\n");
 				return false;
 			}
 		} else if (c == ';') {
@@ -349,44 +461,48 @@ char *usage_msg = "Usage: assembler1 [path/to/file.asm]\n";
 
 int main(int argc, char *argv[])
 {
+	DEBUG("DEBUG 1 2 3...\n");
+
 	bool result = false; 
 	uint16_t instruction;
 	char in_line[MAX_LINE_LEN];
-	size_t in_line_len, i, count;
+	size_t in_line_len, i, file_line;
 	char *in_file_path;
 	FILE *fp;
 
 	if (argc != 2) {                            // requires 1 argument
-		die(usage_msg);
+		alert(usage_msg);
 		exit(-1);
 	}
 
 	in_file_path = argv[1];
 	fp = fopen(in_file_path, "r");
 	if (fp == NULL) {
-		die("failed to open file for reading\n");
+		alert("failed to open file for reading\n");
 		exit(-1);
 	}
 
-	count = 1;
-	while (fgets(in_line, MAX_LINE_LEN, fp) != NULL) {
+	file_line = 1;
+	while (fgets(in_line, MAX_LINE_LEN, fp) != NULL) {  // parse loop
 		in_line_len = strlen(in_line);
 
 		for (i = 0; i < in_line_len; ++i) { // remove newlines
-			if (in_line[i] == '\n') {
+			if (in_line[i] == '\n') {   // TODO handle DOS '\r\n'
 				in_line[i] = '\0';
 				break;
 			}
 		}
 
-		DEBUG("%lu|%s\n", count, in_line);
+		DEBUG("%lu|%s | ", file_line, in_line);
+		g_asm_line = in_line;
+		g_asm_line_number = file_line;
 		result = parse_line(in_line, in_line_len, &instruction);
 		if (result) {
-			DEBUG("instruction: 0x%x  |  ", instruction);
+			DEBUG("instruction: 0x%x | ", instruction);
 			bindump_word16(instruction); // output instruction as binary
 			putchar('\n');
 		}
-	++count;
+	++file_line;
 	}
 
 	return 0;