偵錯的技巧南台科大黎靖偵錯的鄰近原理 v 語句的鄰近

偵錯的技巧南台科大黎靖

偵錯的鄰近原理 v 語句的鄰近 v 執行順序的鄰近 v 空間的鄰近

語句的鄰近 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. #include <stdio. h> #include <stdlib. h> int main() { int i ; for (i = 0; i < 5; i++) printf("n. Hellow World") printf("n. Goodbye Worldn"); system("PAUSE"); return 1; } v v 錯誤訊息指出第 8行 expected `; ‘ before “printf” 。因此實際錯誤是第 7行末尾忘了加 ‘; ’

執行順序的鄰近 int main() 2. { … 40. while(k < 7) 41. { 42. dotest(); …. 240. proc 1(); 241. } 242. } 1. v proc 1()執行完後，若 k<7則執行 dotest() ，所以proc 1() 及 dotest() 為執行順序的鄰近，因此當錯誤訊息指出第 42行有錯時，實際錯誤可能在 240行。

空間的鄰近 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. #include <stdio. h> #include <stdlib. h> int main() { int i, j, k ; i = 0; for(j = 0; j < 19; j++) { for(k = 0; k < j; k++) { putchar(' '); } printf("*n"); } printf("%dn", i); system("PAUSE"); return 1; } v 第 6行及第 15行空間的鄰近，因此找錯誤時，第 6行及第 15行應同時檢查。

偵錯的方法 v 縮小範圍法 v 程式結構追蹤法 v 資料流程追蹤法 v 快照 (Snapshots)

縮小範圍法 v 當編譯器無法正確指出錯誤所在時，可以利用去除部份程式碼的方式，找出錯誤的可能範圍。

找出下列程式的錯誤 1. 2. 3. #include <stdio. h> #include <stdlib. h> #include <string. h> 24. 25. 26. 27. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. #define HEAD 4 #define LENGTH 24 #define P 0 5 #define WIDTH 95 #define ERROR -1 #define MAXLINES 10 char *lines[MAXLINES]; FILE *fd, *printer; int main() { char buff[80]; int max, j, offset, len, count; if ((fd = fopen("address", "r")) == 0) { / * give the user some lelp. . . * / printf("address file does not exist. n"); exit(ERROR); } printer = fopen("LPT 1", "w"); 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. for (count = 0, max = 0; count < MAXLINES; count++) { if (fgets(buff, 80, fd) == 0) break; else { len = strlen(buff); lines[count] = (char *)malloc(80); if (len > max) max = len; strcpy(lines[count], buff); } } offset = ((WIDTH - max)/2)+P 0; for (j = 0; j < offset; j++) buff[j] = ' '; buff[j] = ''; offset = (LENGTH - count)/2 - HEAD; for (j = 0; j < offset; j++) fputs("n", printer); for( j = 0; j < count; j++) fprintf(printer, "%s%s", buff, lines[j]); fputs("14", printer); fclose(fd); fclose(printer); system("PAUSE"); return 1; }

錯誤訊息 v 19: error C 2143: syntax error : missing '; ' before '/‘ v 經檢查發現並沒有少 ‘; ’ 的情況，錯誤在別的地方

先略去紅色部分程式 1. 2. 3. #include <stdio. h> #include <stdlib. h> #include <string. h> 24. 25. 26. 27. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. #define HEAD 4 #define LENGTH 24 #define P 0 5 #define WIDTH 95 #define ERROR -1 #define MAXLINES 10 char *lines[MAXLINES]; FILE *fd, *printer; int main() { char buff[80]; int max, j, offset, len, count; if ((fd = fopen("address", "r")) == 0) { / * give the user some lelp. . . * / printf("address file does not exist. n"); exit(ERROR); } printer = fopen("LPT 1", "w"); 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. for (count = 0, max = 0; count < MAXLINES; count++) { if (fgets(buff, 80, fd) == 0) break; else { len = strlen(buff); lines[count] = (char *)malloc(80); if (len > max) max = len; strcpy(lines[count], buff); } } offset = ((WIDTH - max)/2)+P 0; for (j = 0; j < offset; j++) buff[j] = ' '; buff[j] = ''; offset = (LENGTH - count)/2 - HEAD; for (j = 0; j < offset; j++) fputs("n", printer); for( j = 0; j < count; j++) fprintf(printer, "%s%s", buff, lines[j]); fputs("14", printer); fclose(fd); fclose(printer); system("PAUSE"); return 1; }

錯誤訊息 v 19: error C 2143: syntax error : missing '; ' before '/‘ v 由於錯誤訊息沒有改變，顯然錯誤不在刪除的地方。

再略去藍色部分程式 1. 2. 3. #include <stdio. h> #include <stdlib. h> #include <string. h> 24. 25. 26. 27. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. #define HEAD 4 #define LENGTH 24 #define P 0 5 #define WIDTH 95 #define ERROR -1 #define MAXLINES 10 char *lines[MAXLINES]; FILE *fd, *printer; int main() { char buff[80]; int max, j, offset, len, count; if ((fd = fopen("address", "r")) == 0) { / * give the user some lelp. . . * / printf("address file does not exist. n"); exit(ERROR); } printer = fopen("LPT 1", "w"); 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. for (count = 0, max = 0; count < MAXLINES; count++) { if (fgets(buff, 80, fd) == 0) break; else { len = strlen(buff); lines[count] = (char *)malloc(80); if (len > max) max = len; strcpy(lines[count], buff); } } offset = ((WIDTH - max)/2)+P 0; for (j = 0; j < offset; j++) buff[j] = ' '; buff[j] = ''; offset = (LENGTH - count)/2 - HEAD; for (j = 0; j < offset; j++) fputs("n", printer); for( j = 0; j < count; j++) fprintf(printer, "%s%s", buff, lines[j]); fputs("14", printer); fclose(fd); fclose(printer); system("PAUSE"); return 1; }

錯誤訊息 v 錯誤訊息消失，顯然錯誤在刪除的地方，即藍色程式碼的地方。

只留下藍色部分程式 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. #include <stdio. h> #include <stdlib. h> #include <string. h> #define ERROR -1 FILE *fd, *printer; int main() { char buff[80]; int max, j, offset, len, count; if ((fd = fopen("address", "r")) == 0) { / * give the user some lelp. . . * / printf("address file does not exist. n"); exit(ERROR); } printer = fopen("LPT 1", "w"); fclose(fd); fclose(printer); system("PAUSE"); return 1; }

錯誤訊息 v 19: error C 2143: syntax error : missing '; ' before '/‘ v 由於錯誤訊息沒有改變，顯然錯誤只在 10 至 15行之間。 v 由於錯誤訊息被限制在很短的程式碼中，可以直接小心的檢查程式碼找出錯誤。

追蹤程式的結構 v 追蹤程式的結構可以觀察程式執行的順序是否符合設計。

追蹤 if-else的結構 v 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 原始程式 if (person. age < 5) { addtopreschool(person); } else if (person. age < 14) { addtogrades(person); } else { addtohigh(person); } v 追蹤程式的結構 1. printf("n Approaching age branch, person. age = %d", person. age); if (person. age < 5) { printf("n. Calling addtopreschool"); addtopreschool(person); } else if (person. age < 14) { printf("n. Calling addtogrades"); addtogrades(person); } else { printf("n. Calling addtohigh"); addtohigh(person); } printf("nreturned from call in age branch"); 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17.

追蹤 do-while的結構 v 原始程式 v 1. do { 1. 2. part 1(); part 2(); part 3(); k = part 4(); } while(k < 15); 3. 4. 5. 6. 2. 3. 4. 5. 6. 7. 8. 9. 追蹤程式的結構 printf("n. Approaching 4 -part loop"); do { part 1(); part 2(); part 3(); k = part 4(); printf("n. In 4 -part loop, after part 4, k = %d", k); } while(k < 15); printf("n. Past four part loop");

追蹤 break & continue的結構 v 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 原始程式 while(k--) { checka(); if ((stat = checkb()) == 13) { break; } else if (stat > 25) { continue; } checkc(); } v 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 追蹤程式的結構 printf("n. Approaching checkit loop, k = %d", k); while(k--) { checka(); if ((stat = checkb()) == 13) { printf("nbreaking, stat = 13"); break; } else if (stat > 25) { printf("ncontinuing, stat = %d, k = %d", stat, k); continue; } checkc(); printf("nat end of checks, k = %d", k); } printf("n. Past checkit loop");

找出下列程式的錯誤 1. 2. 3. 4. 5. 6. 7. 8. 9. while (c = getchar() != 'X') { switch(c) { case 'R': dorwork(); break; case 'L': dolwork(); break; case 'Z': dozwork(); break; } }

加入追蹤結構 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. printf("n. Approaching while-loop"); while (c = getchar() != 'X') { switch(c) { case 'R': printf("n. In R case"); dorwork(); break; case 'L': printf("n. In L case"); dolwork(); break; case 'Z': printf("n. In Z case"); dozwork(); break; default: printf("n. Enter an error character %c“, c); exit(); } } printf("n. Exiting while-loop");

加入追蹤結構的結果 v 錯誤仍然存在 v 加入的追蹤程式未發生作用 v 結論: 第 2行可能有問題 v 動作: 加入第 2行之追蹤

新的追蹤結構 v v v v v printf("n. Approaching while-loop"); char ch = getchar(); printf("ngetchar returns %c", ch); while (c = ch != 'X') { printf("n. Loop entered, selecting on %c %d", c, c); switch(c) { case 'R': printf("n. In R case"); dorwork(); break; case 'L': printf("n. In L case"); dolwork(); break; case 'Z': printf("n. In Z case"); dozwork(); break; default: printf("n. Enter an error character“); exit(); } } printf("n. Exiting while-loop");

由鍵盤輸入 R，追蹤結果預期輸出 : Approaching while-loop getchar returns R Loop entered, selecting on R 52 Enter an error character 實際輸出 : Approaching while-loop getchar returns R Loop entered, selecting on 1 Enter an error character

找到錯誤 while (c = getchar() != ‘X’) 應改為 while ((c = getchar()) != ‘X’)

函數的追蹤 v v 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 加入追蹤結構的程式原始程式 k = n+j; m = foobar(k); k = m+j; . . . 1. k = n+j; int foobar(int parm) { int result; result = table[parm]; return result; } 7. 2. 3. 4. 5. 6. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. printf("nready to call foobar with %d", k); m = foobar(k); printf("nreturned %d from foobar", m); k = m+j; . . . int foobar(int parm) { int result; printf("nenter foobar with %d", parm); result = table[parm]; if (result >= 0) { printf("npositive exit, result=%d", result); return result; } printf("nneg result, returning %d", parm); return parm; }

資料流程追蹤 (變數追蹤 ) v 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 原始程式 k=process(m); . . // no references to k in missing code . proc 2(k); . . // no references to k in missing code . j = k + m; k = 2 * process(n); v 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 加入追蹤結構的程式 k=process(m); printf("nafter process k = %d", k); . . // no references to k in missing code . printf("nbefore proc 2 k = %d", k); proc 2(k); . . // no references to k in missing code . printf("nin addition, k = %d", k); j = k + m; k = 2 * process(n); printf("nafter 2&process, k = %d", k);

快照 (Snapshots) 原始程式 #include <stdio. h> #include <stdlib. h> void main() { int x[10]; for (int i=1; i <= 10; i++) x[i] += i*i; } v v 加入快照的程式 #include <stdio. h> #include <stdlib. h> void snaptable(int x[], int n) { for (int i = 0; i < n; i++) printf(“%d “, x[i]); } void main() { int x[10]; snaptable(x, 10); for (int i=1; i <= 10; i++) x[i] += i*i; snaptable(x, 10); }

偵錯的技巧 南台科大 黎靖 偵錯的 鄰近原理 v 語句的鄰近

偵錯的技巧南台科大黎靖偵錯的鄰近原理 v 語句的鄰近