C语言程序参数数量不符时触发core dump问题求助

First, let's tackle the core dump issue you're facing—it's caused by accessing out-of-bounds elements in the argv array before validating the number of arguments. Here's the breakdown:

In your original code, you immediately try to read argv[2] and argv[3] to set Min and Max:

int Min=atoi(argv[2]);
int Max=atoi(argv[3]);
if(Min>Max || argc!=4){
 DisplayHelp();
}

When you pass 2 or 3 arguments, argv[2] (or argv[3]) doesn't exist—argv only contains elements up to argv[argc-1]. Accessing these invalid memory locations triggers a segmentation fault (core dump).

Step 1: Fix the Core Dump #

Move the argument count check before accessing argv[2] and argv[3]. Also, add validation for valid integer values for Min and Max (since atoi returns 0 if the input isn't a number, which could break your length checks).

Here's the corrected start to your main function:

int main(int argc, char * argv[] ){
 char * type1="$1$";
 char * type6="$6$";
 char * salt="$";
 char * mdresult;
 char * sharesult; // Added: missing declaration for sharesult
 char encyption_scheme[20];
 FILE * fp;
 FILE * fp1;
 char input1 [50];
 int length;
 int line=0;
 int line2=0;
 int Min, Max;

 // First validate argument count
 if(argc!=4){
   DisplayHelp();
 }

 // Now safely access argv[2] and argv[3]
 Min=atoi(argv[2]);
 Max=atoi(argv[3]);

 // Validate min/max values meet requirements
 if(Min <=1 || Max < Min){
   printf("Invalid min/max values!\n");
   DisplayHelp();
 }
 // ... rest of your code
}

Step 2: Fix Other Critical Bugs #

Your code has a few other issues that will cause failures even with correct arguments:

• Undefined variable: sharesult is used but never declared—add char *sharesult; alongside mdresult.
• Inefficient file I/O: Opening and closing Filename.txt every iteration is slow and risky. Open it once before processing and close it after:

  // Open output file once before the loop
  fp1=fopen("Filename.txt","a");
  if (fp1==NULL) {
    perror("Failed to open output file");
    exit(-1);
  }
  
  while(fgets(input1,50,fp)){
    // ... your processing logic ...
    fprintf(fp1, "%s:%s\n", input1, mdresult);
    fprintf(fp1, "%s:%s\n", input1, sharesult);
  }
  
  // Close output file after processing all lines
  fclose(fp1);
  

• Check for crypt failures: The crypt function can return NULL if the encryption scheme is invalid or there's an error. Always validate the result:

  mdresult = crypt(input1, encyption_scheme);
  if(mdresult == NULL){
    perror("MD5 encryption failed");
    continue; // Skip this entry or exit, depending on your needs
  }
  // Same check for SHA-512 result
  sharesult = crypt(input1, encyption_scheme);
  if(sharesult == NULL){
    perror("SHA-512 encryption failed");
    continue;
  }
  

Step 3: Optional Robustness Improvements #

• Buffer safety: Use strncat instead of strcat to avoid potential buffer overflow in encyption_scheme:

  strcpy(encyption_scheme, type1);
  strncat(encyption_scheme, salt, sizeof(encyption_scheme) - strlen(encyption_scheme) - 1);
  

• Safe newline trimming: Your current code trims the newline without checking if the line is empty, which could lead to writing to invalid memory. Add a check:

  length=strlen(input1);
  if(length > 0 && input1[length-1] == '\n'){
    input1[length-1] = '\0';
    length--;
  }
  

After making these changes, your program will no longer core dump with incorrect argument counts, and will be more reliable overall.

内容的提问来源于stack exchange，提问作者Ｓｙｕｕバカ