A Cautionary Tale

• I have written two programs that appear quite different.
  • code1
  • code2
• But they are really the same code base.
• I am using a software tool from Stanford which detects similarity in programs.
  • It reports similarity between two programs.
  • 
• Note the main for both were fairly close
  • Code1:

  • int main() {
        string phrase;
        string stripped;
        string reversed;
        string reverseStripped;
    
        phrase = "Hello World ? ? !!...";
        stripped = StripClass(phrase, ispunct);
        reversed = Reverse(phrase);
        reverseStripped = Reverse(stripped);
    
        PrintLabeled(phrase,"Phrase");
        PrintLabeled(stripped,"Stripped");
        PrintLabeled(reversed,"Reversed");
        PrintLabeled(reverseStripped,"Reverse Stripped");
    
        return 0;
    }
        

  • Code2

  • int main() {
        string phrase;
        string stripped;
        string reversed;
        string reverseStripped;
    
        phrase = "Hello World ? ? !!...";
        stripped = bah(phrase, ispunct);
        reversed = hab(phrase);
        reverseStripped = hab(stripped);
    
        PL(phrase,"Phrase");
        PL(stripped,"Stripped");
        PL(reversed,"Reversed");
        PL(reverseStripped,"Reverse Stripped");
    
        return 0;
    }
    	

  • moss nailed it:
  • 
• How about a bunch of messing with variable names
  • Code1: Reverse

  • string Reverse(string s) {
        string rv;
        size_t i;
    
        for(i=0;i<s.size();i++) {
            rv = s[i] + rv;
        }
    
        return rv;
    }
    	

  • Code2: hab

  • // this function will reverse a string.  Isn't the name clever
    // I will just change variable names in this one
    string hab(string x) {
        string r;
        size_t j;
    
        for(j=0;j<x.size();j++) {
            r = x[j] + r;
        }
    
        return r;
    }
    	

  • 
• Ya, but what if I really mess with it, spaces, names, ...
  • Code1:

  • string StripClass(string phrase, int(*TestFunc)(int)) {
        size_t i;
        string rv;
    
        for(i=0;i<phrase.size();i++) {
           if (! TestFunc(phrase[i])) {
              rv += phrase[i];
           }
        }
        return rv;
    }
           

  • Code2:

  • // changing variable names, spacing, indentation
    // adding comments.
    // this should confuse it.
    string bah(string s, int(*x)(int)) {
        // a loop control variable, this will be used to look at the string
        size_t LCV;
        // I want to send this back.
        string back;
    
        // just loop over th string
        for(LCV = 0;  LCV < s.size() ;   LCV++)
        {
    
        if (! x ( s [ LCV ] ) )  // does the passed in function recognise the value
        {
    
    
    back +=  s[LCV];  // if not, add the letter back into the string
    
        }
    
        }
    
    
        return back; // send it back
    }
    
           

  •