tokenize.cpp

cppcheck is a static C/C++ code analyzer that checks for memory leaks, mismatching allocation-deallo

/*
 * Cppcheck - A tool for static C/C++ code analysis
 * Copyright (C) 2007-2009 Daniel Marjamäki, Reijo Tomperi, Nicolas Le Cam,
 * Leandro Penz, Kimmo Varis, Vesa Pikki
 *
 * This program is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program.  If not, see <http://www.gnu.org/licenses/
 */


//---------------------------------------------------------------------------
#include "tokenize.h"
#include "filelister.h"

#include <locale>
#include <fstream>
#include <string>
#include <cstring>
#include <iostream>
#include <sstream>
#include <list>
#include <algorithm>
#include <cstdlib>
#include <cctype>

//---------------------------------------------------------------------------

Tokenizer::Tokenizer()
{
    _tokens = 0;
    _tokensBack = 0;
}

Tokenizer::~Tokenizer()
{
    DeallocateTokens();
}

//---------------------------------------------------------------------------

// Helper functions..


//---------------------------------------------------------------------------

const Token *Tokenizer::tokens() const
{
    return _tokens;
}


const std::vector<std::string> *Tokenizer::getFiles() const
{
    return &_files;
}

//---------------------------------------------------------------------------

//---------------------------------------------------------------------------
// addtoken
// add a token. Used by 'Tokenizer'
//---------------------------------------------------------------------------

void Tokenizer::addtoken(const char str[], const unsigned int lineno, const unsigned int fileno)
{
    if (str[0] == 0)
        return;

    // Replace hexadecimal value with decimal
    std::ostringstream str2;
    if (strncmp(str, "0x", 2) == 0)
    {
        str2 << std::strtoul(str + 2, NULL, 16);
    }
    else
    {
        str2 << str;
    }

    if (_tokensBack)
    {
        _tokensBack->insertToken(str2.str().c_str());
        _tokensBack = _tokensBack->next();
    }
    else
    {
        _tokens = new Token;
        _tokensBack = _tokens;
        _tokensBack->str(str2.str().c_str());
    }

    _tokensBack->linenr(lineno);
    _tokensBack->fileIndex(fileno);
}
//---------------------------------------------------------------------------

//---------------------------------------------------------------------------
// SizeOfType - gives the size of a type
//---------------------------------------------------------------------------



int Tokenizer::SizeOfType(const char type[]) const
{
    if (!type)
        return 0;

    std::map<std::string, unsigned int>::const_iterator it = _typeSize.find(type);
    if (it == _typeSize.end())
        return 0;

    return it->second;
}
//---------------------------------------------------------------------------

//---------------------------------------------------------------------------
// InsertTokens - Copy and insert tokens
//---------------------------------------------------------------------------

void Tokenizer::InsertTokens(Token *dest, Token *src, unsigned int n)
{
    while (n > 0)
    {
        dest->insertToken(src->aaaa());
        dest = dest->next();
        dest->fileIndex(src->fileIndex());
        dest->linenr(src->linenr());
        dest->varId(src->varId());
        src  = src->next();
        --n;
    }
}
//---------------------------------------------------------------------------

//---------------------------------------------------------------------------
// Tokenize - tokenizes a given file.
//---------------------------------------------------------------------------

void Tokenizer::tokenize(std::istream &code, const char FileName[])
{
    // The "_files" vector remembers what files have been tokenized..
    _files.push_back(FileLister::simplifyPath(FileName));

    // line number in parsed code
    unsigned int lineno = 1;

    // The current token being parsed
    std::string CurrentToken;

    // lineNumbers holds line numbers for files in fileIndexes
    // every time an include file is complitely parsed, last item in the vector
    // is removed and lineno is set to point to that value.
    std::vector<unsigned int> lineNumbers;

    // fileIndexes holds index for _files vector about currently parsed files
    // every time an include file is complitely parsed, last item in the vector
    // is removed and FileIndex is set to point to that value.
    std::vector<unsigned int> fileIndexes;

    // FileIndex. What file in the _files vector is read now?
    unsigned int FileIndex = 0;

    // Read one byte at a time from code and create tokens
    for (char ch = (char)code.get(); code.good(); ch = (char)code.get())
    {
        // We are not handling UTF and stuff like that. Code is supposed to plain simple text.
        if (ch < 0)
            continue;

        if (ch == '\n')
        {
            // Add current token..
            addtoken(CurrentToken.c_str(), lineno++, FileIndex);
            CurrentToken.clear();
            continue;
        }

        // char/string..
        if (ch == '\'' || ch == '\"')
        {
            // Add previous token
            addtoken(CurrentToken.c_str(), lineno, FileIndex);
            CurrentToken.clear();

            // read char
            bool special = false;
            char c = ch;
            do
            {
                // Append token..
                CurrentToken += c;

                if (c == '\n')
                    ++lineno;

                // Special sequence '\.'
                if (special)
                    special = false;
                else
                    special = (c == '\\');

                // Get next character
                c = (char)code.get();
            }
            while (code.good() && (special || c != ch));
            CurrentToken += ch;

            // Add token and start on next..
            addtoken(CurrentToken.c_str(), lineno, FileIndex);
            CurrentToken.clear();

            continue;
        }

        if (ch == '#' && CurrentToken.empty())
        {
            // If previous token was "#" then append this to create a "##" token
            if (Token::simpleMatch(_tokensBack, "#"))
            {
                _tokensBack->str("##");
                continue;
            }

            std::string line("#");
            {
                char chPrev = '#';
                bool skip = false;
                while (code.good())
                {
                    ch = (char)code.get();
                    if (chPrev != '\\' && ch == '\n')
                        break;
                    if (chPrev == '\\')
                        line += chPrev;
                    if (chPrev == '#' && ch == '#')
                    {
                        addtoken("##", lineno, FileIndex);
                        skip = true;
                        break;
                    }
                    if (ch != ' ')
                        chPrev = ch;
                    if (ch != '\\' && ch != '\n')
                    {
                        line += ch;
                    }
                    if (ch == '\n')
                        ++lineno;
                }
                if (skip)
                    continue;
            }
            if (strncmp(line.c_str(), "#file", 5) == 0 &&
                line.find("\"") != std::string::npos)
            {
                // Extract the filename
                line.erase(0, line.find("\"") + 1);
                if (line.find("\"") != std::string::npos)
                    line.erase(line.find("\""));

                // Has this file been tokenized already?
                ++lineno;
                bool foundOurfile = false;
                fileIndexes.push_back(FileIndex);
                for (unsigned int i = 0; i < _files.size(); i++)
                {
                    if (FileLister::SameFileName(_files[i].c_str(), line.c_str()))
                    {
                        // Use this index
                        foundOurfile = true;
                        FileIndex = i;
                    }
                }

                if (!foundOurfile)
                {
                    // The "_files" vector remembers what files have been tokenized..
                    _files.push_back(FileLister::simplifyPath(line.c_str()));
                    FileIndex = static_cast<unsigned int>(_files.size() - 1);
                }

                lineNumbers.push_back(lineno);
                lineno = 1;

                continue;
            }

            else if (strncmp(line.c_str(), "#endfile", 8) == 0)
            {
                if (lineNumbers.empty() || fileIndexes.empty())
                {
                    std::cerr << "####### Preprocessor bug! #######\n";
                    std::exit(0);
                }

                lineno = lineNumbers.back();
                lineNumbers.pop_back();
                FileIndex = fileIndexes.back();
                fileIndexes.pop_back();
                continue;
            }

            else
            {
                addtoken(line.c_str(), lineno, FileIndex);
            }
        }

        if (strchr("#+-*/%&|^?!=<>[](){};:,.~", ch))
        {
            if (ch == '.' &&
                CurrentToken.length() > 0 &&
                std::isdigit(CurrentToken[0]))
            {
                // Don't separate doubles "5.4"
            }
            else if (strchr("+-", ch) &&
                     CurrentToken.length() > 0 &&
                     std::isdigit(CurrentToken[0]) &&
                     CurrentToken[CurrentToken.length()-1] == 'e')
            {
                // Don't separate doubles "4.2e+10"
            }
            else
            {
                addtoken(CurrentToken.c_str(), lineno, FileIndex);
                CurrentToken.clear();
                CurrentToken += ch;
                // Add "++", "--" or ">>" token
                if ((ch == '+' || ch == '-' || ch == '>') && (code.peek() == ch))
                    CurrentToken += (char)code.get();
                addtoken(CurrentToken.c_str(), lineno, FileIndex);
                CurrentToken.clear();
                continue;
            }
        }


        if (std::isspace(ch) || std::iscntrl(ch))
        {
            addtoken(CurrentToken.c_str(), lineno, FileIndex);
            CurrentToken.clear();
            continue;
        }

        CurrentToken += ch;
    }
    addtoken(CurrentToken.c_str(), lineno, FileIndex);

    // Combine tokens..
    for (Token *tok = _tokens; tok && tok->next(); tok = tok->next())
    {
        static const char* combineWithNext[][3] =
        {
            { "<", "<", "<<" },

            { "&", "&", "&&" },
            { "|", "|", "||" },

            { "+", "=", "+=" },
            { "-", "=", "-=" },
            { "*", "=", "*=" },
            { "/", "=", "/=" },
            { "&", "=", "&=" },
            { "|", "=", "|=" },

            { "=", "=", "==" },
            { "!", "=", "!=" },
            { "<", "=", "<=" },
            { ">", "=", ">=" },

            { ":", ":", "::" },
            { "-", ">", "." },  // Replace "->" with "."

            { "private", ":", "private:" },
            { "protected", ":", "protected:" },
            { "public", ":", "public:" }
        };

        for (unsigned ui = 0; ui < sizeof(combineWithNext) / sizeof(combineWithNext[0]); ui++)
        {
            if (tok->str() == combineWithNext[ui][0] && tok->next()->str() == combineWithNext[ui][1])
            {
                tok->str(combineWithNext[ui][2]);
                tok->deleteNext();
            }
        }
    }

    // typedef..
    for (Token *tok = _tokens; tok;)
    {
        if (Token::Match(tok, "typedef %type% %type% ;"))
        {
            const char *type1 = tok->strAt(1);
            const char *type2 = tok->strAt(2);
            tok = const_cast<Token*>(tok->tokAt(4));
            for (Token *tok2 = tok; tok2; tok2 = tok2->next())
            {
                if (tok2->str() == type2)
                    tok2->str(type1);
            }
            continue;
        }

        else if (Token::Match(tok, "typedef %type% %type% %type% ;"))
        {
            const char *type1 = tok->strAt(1);
            const char *type2 = tok->strAt(2);
            const char *type3 = tok->strAt(3);
            tok = const_cast<Token*>(tok->tokAt(5));
            for (Token *tok2 = tok; tok2; tok2 = tok2->next())
            {
                if (tok2->str() == type3)
                {
                    tok2->str(type1);
                    tok2->insertToken(type2);
                    tok2 = tok2->next();
                }
            }
            continue;
        }

        tok = tok->next();
    }

    // Remove __asm..
    for (Token *tok = _tokens; tok; tok = tok->next())
    {
        if (Token::simpleMatch(tok->next(), "__asm {"))
        {
            while (tok->next())
            {
                bool last = Token::simpleMatch(tok->next(), "}");

                // Unlink and delete tok->next()
                tok->deleteNext();

                // break if this was the last token to delete..
                if (last)
                    break;
            }
        }
    }

    // Remove "volatile"
    while (Token::simpleMatch(_tokens, "volatile"))
    {
        Token *tok = _tokens;
        _tokens = _tokens->next();
        delete tok;
    }
    for (Token *tok = _tokens; tok; tok = tok->next())
    {
        while (Token::simpleMatch(tok->next(), "volatile"))
        {
            tok->deleteNext();
        }
    }

    // Remove "mutable"
    while (Token::simpleMatch(_tokens, "mutable"))
    {
        Token *tok = _tokens;
        _tokens = _tokens->next();
        delete tok;
    }
    for (Token *tok = _tokens; tok; tok = tok->next())
    {
        while (Token::simpleMatch(tok->next(), "mutable"))
        {
            tok->deleteNext();
        }
    }

}
//---------------------------------------------------------------------------


void Tokenizer::setVarId()
{
    // Clear all variable ids