426 lines
13 KiB
C++
426 lines
13 KiB
C++
/*
|
|
* Cppcheck - A tool for static C/C++ code analysis
|
|
* Copyright (C) 2007-2013 Daniel Marjamäki and Cppcheck team.
|
|
*
|
|
* 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 "tokenlist.h"
|
|
#include "token.h"
|
|
#include "mathlib.h"
|
|
#include "path.h"
|
|
#include "preprocessor.h"
|
|
#include "settings.h"
|
|
#include "errorlogger.h"
|
|
|
|
#include <cstring>
|
|
#include <sstream>
|
|
#include <cctype>
|
|
#include <stack>
|
|
|
|
|
|
TokenList::TokenList(const Settings* settings) :
|
|
_front(0),
|
|
_back(0),
|
|
_settings(settings)
|
|
{
|
|
}
|
|
|
|
TokenList::~TokenList()
|
|
{
|
|
deallocateTokens();
|
|
}
|
|
|
|
//---------------------------------------------------------------------------
|
|
|
|
// Deallocate lists..
|
|
void TokenList::deallocateTokens()
|
|
{
|
|
deleteTokens(_front);
|
|
_front = 0;
|
|
_back = 0;
|
|
_files.clear();
|
|
}
|
|
|
|
void TokenList::deleteTokens(Token *tok)
|
|
{
|
|
while (tok) {
|
|
Token *next = tok->next();
|
|
delete tok;
|
|
tok = next;
|
|
}
|
|
}
|
|
|
|
//---------------------------------------------------------------------------
|
|
// add a token.
|
|
//---------------------------------------------------------------------------
|
|
|
|
void TokenList::addtoken(const char str[], const unsigned int lineno, const unsigned int fileno, bool split)
|
|
{
|
|
if (str[0] == 0)
|
|
return;
|
|
|
|
// If token contains # characters, split it up
|
|
if (split && std::strstr(str, "##")) {
|
|
std::string temp;
|
|
for (unsigned int i = 0; str[i]; ++i) {
|
|
if (std::strncmp(&str[i], "##", 2) == 0) {
|
|
addtoken(temp.c_str(), lineno, fileno, false);
|
|
temp.clear();
|
|
addtoken("##", lineno, fileno, false);
|
|
++i;
|
|
} else
|
|
temp += str[i];
|
|
}
|
|
addtoken(temp.c_str(), lineno, fileno, false);
|
|
return;
|
|
}
|
|
|
|
// Replace hexadecimal value with decimal
|
|
std::ostringstream str2;
|
|
if (MathLib::isHex(str) || MathLib::isOct(str) || MathLib::isBin(str)) {
|
|
str2 << MathLib::toLongNumber(str);
|
|
} else if (std::strncmp(str, "_Bool", 5) == 0) {
|
|
str2 << "bool";
|
|
} else {
|
|
str2 << str;
|
|
}
|
|
|
|
if (_back) {
|
|
_back->insertToken(str2.str());
|
|
} else {
|
|
_front = new Token(&_back);
|
|
_back = _front;
|
|
_back->str(str2.str());
|
|
}
|
|
|
|
_back->linenr(lineno);
|
|
_back->fileIndex(fileno);
|
|
}
|
|
|
|
void TokenList::addtoken(const Token * tok, const unsigned int lineno, const unsigned int fileno)
|
|
{
|
|
if (tok == 0)
|
|
return;
|
|
|
|
if (_back) {
|
|
_back->insertToken(tok->str());
|
|
} else {
|
|
_front = new Token(&_back);
|
|
_back = _front;
|
|
_back->str(tok->str());
|
|
}
|
|
|
|
_back->linenr(lineno);
|
|
_back->fileIndex(fileno);
|
|
_back->isUnsigned(tok->isUnsigned());
|
|
_back->isSigned(tok->isSigned());
|
|
_back->isLong(tok->isLong());
|
|
_back->isUnused(tok->isUnused());
|
|
}
|
|
//---------------------------------------------------------------------------
|
|
// InsertTokens - Copy and insert tokens
|
|
//---------------------------------------------------------------------------
|
|
|
|
void TokenList::insertTokens(Token *dest, const Token *src, unsigned int n)
|
|
{
|
|
std::stack<Token *> link;
|
|
|
|
while (n > 0) {
|
|
dest->insertToken(src->str());
|
|
dest = dest->next();
|
|
|
|
// Set links
|
|
if (Token::Match(dest, "(|[|{"))
|
|
link.push(dest);
|
|
else if (!link.empty() && Token::Match(dest, ")|]|}")) {
|
|
Token::createMutualLinks(dest, link.top());
|
|
link.pop();
|
|
}
|
|
|
|
dest->fileIndex(src->fileIndex());
|
|
dest->linenr(src->linenr());
|
|
dest->varId(src->varId());
|
|
dest->type(src->type());
|
|
dest->isUnsigned(src->isUnsigned());
|
|
dest->isSigned(src->isSigned());
|
|
dest->isPointerCompare(src->isPointerCompare());
|
|
dest->isLong(src->isLong());
|
|
dest->isUnused(src->isUnused());
|
|
src = src->next();
|
|
--n;
|
|
}
|
|
}
|
|
|
|
//---------------------------------------------------------------------------
|
|
// Tokenize - tokenizes a given file.
|
|
//---------------------------------------------------------------------------
|
|
|
|
bool TokenList::createTokens(std::istream &code, const std::string& file0)
|
|
{
|
|
_files.push_back(file0);
|
|
|
|
// 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 completely parsed, last item in the vector
|
|
// is removed and lineno is set to point to that value.
|
|
std::stack<unsigned int> lineNumbers;
|
|
|
|
// fileIndexes holds index for _files vector about currently parsed files
|
|
// every time an include file is completely parsed, last item in the vector
|
|
// is removed and FileIndex is set to point to that value.
|
|
std::stack<unsigned int> fileIndexes;
|
|
|
|
// FileIndex. What file in the _files vector is read now?
|
|
unsigned int FileIndex = 0;
|
|
|
|
bool expandedMacro = false;
|
|
|
|
// Read one byte at a time from code and create tokens
|
|
for (char ch = (char)code.get(); code.good(); ch = (char)code.get()) {
|
|
if (ch == Preprocessor::macroChar) {
|
|
while (code.peek() == Preprocessor::macroChar)
|
|
code.get();
|
|
ch = ' ';
|
|
expandedMacro = true;
|
|
} else if (ch == '\n') {
|
|
expandedMacro = false;
|
|
}
|
|
|
|
// char/string..
|
|
// multiline strings are not handled. The preprocessor should handle that for us.
|
|
else if (ch == '\'' || ch == '\"') {
|
|
std::string line;
|
|
|
|
// read char
|
|
bool special = false;
|
|
char c = ch;
|
|
do {
|
|
// Append token..
|
|
line += c;
|
|
|
|
// Special sequence '\.'
|
|
if (special)
|
|
special = false;
|
|
else
|
|
special = (c == '\\');
|
|
|
|
// Get next character
|
|
c = (char)code.get();
|
|
} while (code.good() && (special || c != ch));
|
|
line += ch;
|
|
|
|
// Handle #file "file.h"
|
|
if (CurrentToken == "#file") {
|
|
// Extract the filename
|
|
line = line.substr(1, line.length() - 2);
|
|
|
|
// Has this file been tokenized already?
|
|
++lineno;
|
|
bool foundOurfile = false;
|
|
fileIndexes.push(FileIndex);
|
|
for (unsigned int i = 0; i < _files.size(); ++i) {
|
|
if (Path::sameFileName(_files[i], line)) {
|
|
// Use this index
|
|
foundOurfile = true;
|
|
FileIndex = i;
|
|
}
|
|
}
|
|
|
|
if (!foundOurfile) {
|
|
// The "_files" vector remembers what files have been tokenized..
|
|
_files.push_back(Path::simplifyPath(line.c_str()));
|
|
FileIndex = static_cast<unsigned int>(_files.size() - 1);
|
|
}
|
|
|
|
lineNumbers.push(lineno);
|
|
lineno = 0;
|
|
} else {
|
|
// Add previous token
|
|
addtoken(CurrentToken.c_str(), lineno, FileIndex);
|
|
if (!CurrentToken.empty())
|
|
_back->setExpandedMacro(expandedMacro);
|
|
|
|
// Add content of the string
|
|
addtoken(line.c_str(), lineno, FileIndex);
|
|
if (!line.empty())
|
|
_back->setExpandedMacro(expandedMacro);
|
|
}
|
|
|
|
CurrentToken.clear();
|
|
|
|
continue;
|
|
}
|
|
|
|
if (ch == '.' &&
|
|
CurrentToken.length() > 0 &&
|
|
std::isdigit(CurrentToken[0])) {
|
|
// Don't separate doubles "5.4"
|
|
} else if (std::strchr("+-", ch) &&
|
|
CurrentToken.length() > 0 &&
|
|
std::isdigit(CurrentToken[0]) &&
|
|
(CurrentToken[CurrentToken.length()-1] == 'e' ||
|
|
CurrentToken[CurrentToken.length()-1] == 'E') &&
|
|
!MathLib::isHex(CurrentToken)) {
|
|
// Don't separate doubles "4.2e+10"
|
|
} else if (CurrentToken.empty() && ch == '.' && std::isdigit(code.peek())) {
|
|
// tokenize .125 into 0.125
|
|
CurrentToken = "0";
|
|
} else if (std::strchr("+-*/%&|^?!=<>[](){};:,.~\n ", ch)) {
|
|
if (CurrentToken == "#file") {
|
|
// Handle this where strings are handled
|
|
continue;
|
|
} else if (CurrentToken == "#line") {
|
|
// Read to end of line
|
|
std::string line;
|
|
|
|
std::getline(code, line);
|
|
|
|
// Update the current line number
|
|
std::stringstream(line) >> lineno;
|
|
CurrentToken.clear();
|
|
continue;
|
|
} else if (CurrentToken == "#endfile") {
|
|
if (lineNumbers.empty() || fileIndexes.empty()) { // error
|
|
deallocateTokens();
|
|
return false;
|
|
}
|
|
|
|
lineno = lineNumbers.top();
|
|
lineNumbers.pop();
|
|
FileIndex = fileIndexes.top();
|
|
fileIndexes.pop();
|
|
CurrentToken.clear();
|
|
continue;
|
|
}
|
|
|
|
addtoken(CurrentToken.c_str(), lineno, FileIndex, true);
|
|
if (!CurrentToken.empty())
|
|
_back->setExpandedMacro(expandedMacro);
|
|
|
|
CurrentToken.clear();
|
|
|
|
if (ch == '\n') {
|
|
++lineno;
|
|
continue;
|
|
} else if (ch == ' ') {
|
|
continue;
|
|
}
|
|
|
|
CurrentToken += ch;
|
|
// Add "++", "--", ">>" or ... token
|
|
if (std::strchr("+-<>=:&|", ch) && (code.peek() == ch))
|
|
CurrentToken += (char)code.get();
|
|
addtoken(CurrentToken.c_str(), lineno, FileIndex);
|
|
_back->setExpandedMacro(expandedMacro);
|
|
CurrentToken.clear();
|
|
continue;
|
|
}
|
|
|
|
CurrentToken += ch;
|
|
}
|
|
addtoken(CurrentToken.c_str(), lineno, FileIndex, true);
|
|
if (!CurrentToken.empty())
|
|
_back->setExpandedMacro(expandedMacro);
|
|
_front->assignProgressValues();
|
|
|
|
for (unsigned int i = 1; i < _files.size(); i++)
|
|
_files[i] = Path::getRelativePath(_files[i], _settings->_basePaths);
|
|
|
|
return true;
|
|
}
|
|
|
|
//---------------------------------------------------------------------------
|
|
|
|
void TokenList::createAst()
|
|
{
|
|
// operators that must be ordered according to C-precedence
|
|
const char * const operators[] = {
|
|
" :: ",
|
|
" ++ -- . ",
|
|
"> ++ -- + - ! ~ * & sizeof ", // prefix unary operators, from right to left
|
|
" * / % ",
|
|
" + - ",
|
|
" << >> ",
|
|
" < <= > >= ",
|
|
" == != ",
|
|
" & ",
|
|
" ^ ",
|
|
" | ",
|
|
" && ",
|
|
" || ",
|
|
" = ? : ",
|
|
" throw ",
|
|
" , "
|
|
" [ "
|
|
};
|
|
|
|
for (unsigned int i = 0; i < sizeof(operators) / sizeof(*operators); ++i) {
|
|
// TODO: extract operators to std::set - that should be faster
|
|
if (*operators[i] == '>') { // Unary operators, parse from right to left
|
|
const std::string op(1+operators[i]);
|
|
Token *tok = _front;
|
|
while (tok->next())
|
|
tok = tok->next();
|
|
for (; tok; tok = tok->previous()) {
|
|
if ((!tok->previous() || tok->previous()->isOp()) &&
|
|
op.find(" "+tok->str()+" ")!=std::string::npos) {
|
|
tok->astOperand1(tok->next());
|
|
}
|
|
}
|
|
} else { // parse from left to right
|
|
const std::string op(operators[i]);
|
|
for (Token *tok = _front; tok; tok = tok->next()) {
|
|
if (tok->astOperand1()==NULL && op.find(" "+tok->str()+" ")!=std::string::npos) {
|
|
if (tok->str() != "++" && tok->str() != "--") {
|
|
tok->astOperand1(tok->previous());
|
|
tok->astOperand2(tok->next());
|
|
} else if (tok->previous() && !tok->previous()->isOp()) {
|
|
tok->astOperand1(tok->previous());
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// function calls..
|
|
for (Token *tok = _front; tok; tok = tok->next()) {
|
|
if (Token::Match(tok, "%var% ("))
|
|
tok->astFunctionCall();
|
|
}
|
|
|
|
// parentheses..
|
|
for (Token *tok = _front; tok; tok = tok->next()) {
|
|
if (tok->str() == "(" || tok->str() == ")" || tok->str() == "]") {
|
|
tok->astHandleParentheses();
|
|
}
|
|
}
|
|
}
|
|
|
|
const std::string& TokenList::file(const Token *tok) const
|
|
{
|
|
return _files.at(tok->fileIndex());
|
|
}
|
|
|
|
std::string TokenList::fileLine(const Token *tok) const
|
|
{
|
|
return ErrorLogger::ErrorMessage::FileLocation(tok, this).stringify();
|
|
}
|