vendredi 18 juin 2021

Would you please help me find the cause for my Lexical Analyzer Error: Invalid Symbol?

Okay, so the issue is that I don't know why I'm getting this error.

For a class, we're writing a compiler piece by piece. This bit of code is supposed to tokenize input symbols. I wrote a series of if/else statements that act like a very simple trie, thinking that it would be able to find all the symbols. It works fine for most of them, but is getting stuck on "<>".

Here's the whole function:

// Process the symbols
void symbol_processor(char *input)
{
    // Initialize symbol_type operator
    int symbol_type = -1;

    printf("Location1: %d\n", input_index);

    // A series of if/else that ape a trie
    if (input[input_index] == '=')
        if (input[input_index + 1] == '=')
        {
            printf("Location2: %d\n", input_index);
            // Set the symbol for "=="
            symbol_type = eqlsym;

            // Move forward two input_index spaces
            input_index += 2;
            printf("Location4: %d\n", input_index);
        }
    else if (input[input_index] == '<')
        if (input[input_index + 1] == '>')
        {
            printf("Location4: %d\n", input_index);
            // Set the symbol for "<>"
            symbol_type = neqsym;

            // Move forward two input_index spaces
            input_index += 2;
        }
        else if (input[input_index + 1] == '=')
        {
            // Set the symbol for "<="
            symbol_type = leqsym;

            // Move forward two input_index spaces
            input_index += 2;
        }
        else
        {
            printf("Location: %d\n", input_index);
            // Set the symbol for "<"
            symbol_type = lessym;

            // Move forward one input_index space
            input_index++;
        }
    else if (input[input_index] == '>')
        if (input[input_index + 1] == '=')
        {
            // Set the symbol for ">="
            symbol_type = geqsym;

            // Move forward two input_index spaces
            input_index += 2;
        }
        else
        {
            // Set the symbol for ">"
            symbol_type = modsym;

            // Move forward one index space
            input_index++;
        }
    else if (input[input_index] == ':')
        if (input[input_index + 1] == '=')
        {
            // Set the symbol for ":="
            symbol_type = becomessym;

            // Move forward two input_index spaces
            input_index += 2;
        }
    // This could cause an issue
    else if (input[input_index] == '/')
        if (input[input_index + 1] == '*')
            comment_error = comment_processor(input);
        else
        {
            
            // Set the symbol for ">"
            symbol_type = slashsym;

            // Move forward one index space
            input_index++;
        }
    else if (input[input_index] == '%')
    {
        // Set the symbol for "%"
        symbol_type = modsym;

        // Move forward one index space
        input_index++;
    }
    else if (input[input_index] == '*')
    {
        // Set the symbol for "*"
        symbol_type = multsym;

        // Move forward one index space
        input_index++;
    }
    else if (input[input_index] == '+')
    {
        // Set the symbol for "+"
        symbol_type = plussym;

        // Move forward one index space
        input_index++;
    }
    else if (input[input_index] == '-')
    {
        // Set the symbol for "-"
        symbol_type = minussym;

        // Move forward one index space
        input_index++;
    }
    else if (input[input_index] == '(')
    {
        // Set the symbol for "("
        symbol_type = lparentsym;

        // Move forward one index space
        input_index++;
    }
    else if (input[input_index] == ')')
    {
        // Set the symbol for ")"
        symbol_type = rparentsym;

        // Move forward one input_index space
        input_index++;
    }
    else if (input[input_index] == ',')
    {
        // Set the symbol for ","
        symbol_type = commasym;

        // Move forward one index space
        input_index++;
    }
    else if (input[input_index] == '.')
    {
        // Set the symbol for "."
        symbol_type = periodsym;

        // Move forward one input_index space
        input_index++;
    }
    else if (input[input_index] == ';')
    {
        // Set the symbol for ";"
        symbol_type = semicolonsym;

        // Move forward one index space
        input_index++;
    }

    // Check to see if an error should be thrown
    if (symbol_type == -1)
        error_processor(1); // Invalid Symbol

    // Append symbol to the list
    list[lex_index].type = symbol_type;
    lex_index++;
}

But I'm pretty sure the problem is in here:

else if (input[input_index] == '<')
        if (input[input_index + 1] == '>')
        {
            printf("Location4: %d\n", input_index);
            // Set the symbol for "<>"
            symbol_type = neqsym;

            // Move forward two input_index spaces
            input_index += 2;
        }
        else if (input[input_index + 1] == '=')
        {
            // Set the symbol for "<="
            symbol_type = leqsym;

            // Move forward two input_index spaces
            input_index += 2;
        }
        else
        {
            printf("Location: %d\n", input_index);
            // Set the symbol for "<"
            symbol_type = lessym;

            // Move forward one input_index space
            input_index++;
        }

I just can't see what the issue is, and was hoping that programmers more wise and experienced than myself could point it out. Also, just ignore the printf statements. I was using those to try and help debug.

Here's the entire input text I'm feeding in, if it helps. The error gets thrown at the '<' that comes right after "var".

const==var<>procedureend<=if>=then.else;while(do)call:=read,write+124-jalapeno*/comment//

Aucun commentaire:

Enregistrer un commentaire