2008-12-18 22:28:57 +01:00
/*
2009-01-21 21:04:20 +01:00
* Cppcheck - A tool for static C / C + + code analysis
2020-05-10 11:16:32 +02:00
* Copyright ( C ) 2007 - 2020 Cppcheck team .
2008-12-18 22:28:57 +01:00
*
* 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
2009-09-27 17:08:31 +02:00
* along with this program . If not , see < http : //www.gnu.org/licenses/>.
2008-12-18 22:28:57 +01:00
*/
2013-09-04 20:59:49 +02:00
//---------------------------------------------------------------------------
# ifndef tokenH
# define tokenH
//---------------------------------------------------------------------------
2008-12-18 22:28:57 +01:00
2017-05-27 04:33:47 +02:00
# include "config.h"
# include "mathlib.h"
# include "valueflow.h"
2018-12-21 13:51:45 +01:00
# include "templatesimplifier.h"
2019-07-14 12:22:33 +02:00
# include "utils.h"
2017-05-27 04:33:47 +02:00
# include <cstddef>
2018-11-14 06:59:25 +01:00
# include <functional>
2014-01-04 20:57:02 +01:00
# include <list>
2019-07-31 09:19:27 +02:00
# include <memory>
2017-05-27 04:33:47 +02:00
# include <ostream>
2008-12-18 22:28:57 +01:00
# include <string>
2009-11-27 23:04:04 +01:00
# include <vector>
2008-12-18 22:28:57 +01:00
2017-05-27 04:33:47 +02:00
class Enumerator ;
class Function ;
2012-08-11 20:47:11 +02:00
class Scope ;
2017-05-27 04:33:47 +02:00
class Settings ;
2015-08-14 20:46:13 +02:00
class Type ;
2015-10-04 19:42:58 +02:00
class ValueType ;
2017-05-27 04:33:47 +02:00
class Variable ;
2020-05-17 17:25:33 +02:00
class TokenList ;
2012-08-11 20:47:11 +02:00
2018-05-25 07:15:05 +02:00
/**
* @ brief This struct stores pointers to the front and back tokens of the list this token is in .
*/
struct TokensFrontBack {
Token * front ;
Token * back ;
2020-05-17 17:25:33 +02:00
const TokenList * list ;
2018-05-25 07:15:05 +02:00
} ;
2019-07-31 09:19:27 +02:00
struct ScopeInfo2 {
ScopeInfo2 ( const std : : string & name_ , const Token * bodyEnd_ , const std : : set < std : : string > & usingNamespaces_ = std : : set < std : : string > ( ) ) : name ( name_ ) , bodyEnd ( bodyEnd_ ) , usingNamespaces ( usingNamespaces_ ) { }
std : : string name ;
const Token * const bodyEnd ;
std : : set < std : : string > usingNamespaces ;
} ;
2018-12-21 13:51:45 +01:00
struct TokenImpl {
2019-07-14 12:22:33 +02:00
nonneg int mVarId ;
nonneg int mFileIndex ;
nonneg int mLineNumber ;
nonneg int mColumn ;
2020-08-20 18:21:29 +02:00
MathLib : : bigint mExprId ;
2018-12-21 13:51:45 +01:00
// AST..
Token * mAstOperand1 ;
Token * mAstOperand2 ;
Token * mAstParent ;
// symbol database information
const Scope * mScope ;
union {
const Function * mFunction ;
const Variable * mVariable ;
const : : Type * mType ;
const Enumerator * mEnumerator ;
} ;
/**
* A value from 0 - 100 that provides a rough idea about where in the token
* list this token is located .
*/
2019-07-14 12:22:33 +02:00
nonneg int mProgressValue ;
2018-12-21 13:51:45 +01:00
2019-06-21 22:16:23 +02:00
/**
* Token index . Position in token list
*/
2019-07-14 12:22:33 +02:00
nonneg int mIndex ;
2019-06-21 22:16:23 +02:00
2018-12-21 13:51:45 +01:00
// original name like size_t
std : : string * mOriginalName ;
// ValueType
ValueType * mValueType ;
// ValueFlow
std : : list < ValueFlow : : Value > * mValues ;
static const std : : list < ValueFlow : : Value > mEmptyValueList ;
// Pointer to a template in the template simplifier
2020-05-15 09:02:59 +02:00
std : : set < TemplateSimplifier : : TokenAndName * > * mTemplateSimplifierPointers ;
2018-12-21 13:51:45 +01:00
2019-07-31 09:19:27 +02:00
// Pointer to the object representing this token's scope
std : : shared_ptr < ScopeInfo2 > mScopeInfo ;
2019-07-31 11:11:01 +02:00
2019-07-12 11:09:24 +02:00
// __cppcheck_in_range__
struct CppcheckAttributes {
enum Type { LOW , HIGH } type ;
MathLib : : bigint value ;
struct CppcheckAttributes * next ;
} ;
struct CppcheckAttributes * mCppcheckAttributes ;
2019-09-02 20:31:01 +02:00
// For memoization, to speed up parsing of huge arrays #8897
enum class Cpp11init { UNKNOWN , CPP11INIT , NOINIT } mCpp11init ;
2020-05-15 09:02:59 +02:00
/** Bitfield bit count. */
unsigned char mBits ;
2019-07-12 11:09:24 +02:00
void setCppcheckAttribute ( CppcheckAttributes : : Type type , MathLib : : bigint value ) ;
bool getCppcheckAttribute ( CppcheckAttributes : : Type type , MathLib : : bigint * value ) const ;
2018-12-21 13:51:45 +01:00
TokenImpl ( )
: mVarId ( 0 )
, mFileIndex ( 0 )
, mLineNumber ( 0 )
, mColumn ( 0 )
2020-08-20 18:21:29 +02:00
, mExprId ( 0 )
2018-12-21 13:51:45 +01:00
, mAstOperand1 ( nullptr )
, mAstOperand2 ( nullptr )
, mAstParent ( nullptr )
, mScope ( nullptr )
, mFunction ( nullptr ) // Initialize whole union
, mProgressValue ( 0 )
2019-06-21 22:16:23 +02:00
, mIndex ( 0 )
2018-12-21 13:51:45 +01:00
, mOriginalName ( nullptr )
, mValueType ( nullptr )
, mValues ( nullptr )
2020-05-15 09:02:59 +02:00
, mTemplateSimplifierPointers ( nullptr )
2019-07-31 09:19:27 +02:00
, mScopeInfo ( nullptr )
2019-07-12 11:09:24 +02:00
, mCppcheckAttributes ( nullptr )
2019-09-02 20:31:01 +02:00
, mCpp11init ( Cpp11init : : UNKNOWN )
2020-05-15 09:02:59 +02:00
, mBits ( 0 )
2018-12-21 13:51:45 +01:00
{ }
~ TokenImpl ( ) ;
} ;
2009-07-17 10:49:01 +02:00
/// @addtogroup Core
/// @{
2009-07-13 13:35:33 +02:00
/**
2012-05-05 18:33:26 +02:00
* @ brief The token list that the TokenList generates is a linked - list of this class .
2009-07-13 13:35:33 +02:00
*
* Tokens are stored as strings . The " if " , " while " , etc are stored in plain text .
* The reason the Token class is needed ( instead of using the string class ) is that some extra functionality is also needed for tokens :
2017-05-17 14:57:54 +02:00
* - location of the token is stored ( fileIndex , linenr , column )
2009-07-13 13:35:33 +02:00
* - functions for classifying the token ( isName , isNumber , isBoolean , isStandardType )
*
* The Token class also has other functions for management of token list , matching tokens , etc .
*/
2012-06-10 14:19:09 +02:00
class CPPCHECKLIB Token {
2010-01-06 20:19:27 +01:00
private :
2018-06-16 16:22:35 +02:00
TokensFrontBack * mTokensFrontBack ;
2010-01-06 20:19:27 +01:00
// Not implemented..
2012-12-01 02:16:03 +01:00
Token ( const Token & ) ;
Token operator = ( const Token & ) ;
2010-01-06 20:19:27 +01:00
public :
2012-04-23 21:05:26 +02:00
enum Type {
2014-06-04 22:33:08 +02:00
eVariable , eType , eFunction , eKeyword , eName , // Names: Variable (varId), Type (typeId, later), Function (FuncId, later), Language keyword, Name (unknown identifier)
2016-04-22 06:02:54 +02:00
eNumber , eString , eChar , eBoolean , eLiteral , eEnumerator , // Literals: Number, String, Character, Boolean, User defined literal (C++11), Enumerator
2012-04-26 23:04:55 +02:00
eArithmeticalOp , eComparisonOp , eAssignmentOp , eLogicalOp , eBitOp , eIncDecOp , eExtendedOp , // Operators: Arithmetical, Comparison, Assignment, Logical, Bitwise, ++/--, Extended
eBracket , // {, }, <, >: < and > only if link() is set. Otherwise they are comparison operators.
2019-07-05 12:30:42 +02:00
eLambda , // A function without a name
2019-09-04 08:07:30 +02:00
eEllipsis , // "..."
2012-04-23 21:05:26 +02:00
eOther ,
eNone
} ;
2018-05-25 07:15:05 +02:00
explicit Token ( TokensFrontBack * tokensFrontBack = nullptr ) ;
2009-01-03 21:29:20 +01:00
~ Token ( ) ;
2009-06-14 08:55:23 +02:00
2014-04-02 18:04:20 +02:00
template < typename T >
2014-11-20 14:20:09 +01:00
void str ( T & & s ) {
2018-06-16 23:03:15 +02:00
mStr = s ;
2018-12-21 13:51:45 +01:00
mImpl - > mVarId = 0 ;
2014-04-02 18:04:20 +02:00
update_property_info ( ) ;
}
2008-12-18 22:28:57 +01:00
2011-10-23 21:21:42 +02:00
/**
2011-10-23 22:46:55 +02:00
* Concatenate two ( quoted ) strings . Automatically cuts of the last / first character .
* Example : " hello " " world " - > " hello world " . Used by the token simplifier .
2011-10-23 21:21:42 +02:00
*/
2009-03-28 20:33:55 +01:00
void concatStr ( std : : string const & b ) ;
2014-11-20 14:20:09 +01:00
const std : : string & str ( ) const {
2018-06-16 23:03:15 +02:00
return mStr ;
2009-01-05 16:49:57 +01:00
}
2008-12-18 22:28:57 +01:00
/**
2019-07-13 15:47:53 +02:00
* Unlink and delete the next ' count ' tokens .
2008-12-18 22:28:57 +01:00
*/
2019-07-14 12:22:33 +02:00
void deleteNext ( nonneg int count = 1 ) ;
2008-12-18 22:28:57 +01:00
2018-05-25 07:15:05 +02:00
/**
2019-07-13 15:47:53 +02:00
* Unlink and delete the previous ' count ' tokens .
2018-05-25 07:15:05 +02:00
*/
2019-07-14 12:22:33 +02:00
void deletePrevious ( nonneg int count = 1 ) ;
2018-05-25 07:15:05 +02:00
2014-12-27 10:53:26 +01:00
/**
* Swap the contents of this token with the next token .
*/
void swapWithNext ( ) ;
2008-12-18 22:28:57 +01:00
/**
2014-04-27 09:32:02 +02:00
* @ return token in given index , related to this token .
2008-12-18 22:28:57 +01:00
* For example index 1 would return next token , and 2
* would return next from that one .
*/
2009-01-03 21:29:20 +01:00
const Token * tokAt ( int index ) const ;
2014-11-20 14:20:09 +01:00
Token * tokAt ( int index ) {
2015-08-27 16:14:33 +02:00
return const_cast < Token * > ( const_cast < const Token * > ( this ) - > tokAt ( index ) ) ;
2013-01-03 22:26:20 +01:00
}
2008-12-18 22:28:57 +01:00
2011-11-11 21:55:37 +01:00
/**
2014-04-27 09:32:02 +02:00
* @ return the link to the token in given index , related to this token .
2011-11-11 21:55:37 +01:00
* For example index 1 would return the link to next token .
*/
const Token * linkAt ( int index ) const ;
2014-11-20 14:20:09 +01:00
Token * linkAt ( int index ) {
2015-08-27 16:14:33 +02:00
return const_cast < Token * > ( const_cast < const Token * > ( this ) - > linkAt ( index ) ) ;
2013-01-03 22:26:20 +01:00
}
2011-11-11 21:55:37 +01:00
2014-04-27 09:32:02 +02:00
/**
* @ return String of the token in given index , related to this token .
2014-07-11 09:33:09 +02:00
* If that token does not exist , an empty string is being returned .
2014-04-27 09:32:02 +02:00
*/
2011-11-14 09:16:47 +01:00
const std : : string & strAt ( int index ) const ;
2008-12-18 22:28:57 +01:00
2008-12-23 22:51:54 +01:00
/**
* Match given token ( or list of tokens ) to a pattern list .
*
* Possible patterns
* " someRandomText " If token contains " someRandomText " .
* @ note Use Match ( ) if you want to use flags in patterns
*
* The patterns can be also combined to compare to multiple tokens at once
* by separating tokens with a space , e . g .
* " ) void { " will return true if first token is ' ) ' next token
* is " void " and token after that is ' { ' . If even one of the tokens does
* not match its pattern , false is returned .
*
* @ param tok List of tokens to be compared to the pattern
* @ param pattern The pattern against which the tokens are compared ,
* e . g . " const " or " ) void { " .
* @ return true if given token matches with given pattern
* false if given token does not match with given pattern
*/
2020-05-26 20:13:56 +02:00
template < size_t count >
static bool simpleMatch ( const Token * tok , const char ( & pattern ) [ count ] ) {
return simpleMatch ( tok , pattern , count - 1 ) ;
}
static bool simpleMatch ( const Token * tok , const char pattern [ ] , size_t pattern_len ) ;
2008-12-22 00:28:09 +01:00
2008-12-18 22:28:57 +01:00
/**
* Match given token ( or list of tokens ) to a pattern list .
*
* Possible patterns
2010-03-13 22:16:06 +01:00
* - " %any% " any token
2015-12-31 01:25:36 +01:00
* - " %assign% " a assignment operand
2010-03-13 22:16:06 +01:00
* - " %bool% " true or false
2012-11-04 16:34:32 +01:00
* - " %char% " Any token enclosed in & apos ; - character .
2012-12-01 01:31:35 +01:00
* - " %comp% " Any token such that isComparisonOp ( ) returns true .
2013-03-01 11:43:59 +01:00
* - " %cop% " Any token such that isConstOp ( ) returns true .
2015-12-31 01:25:36 +01:00
* - " %name% " any token which is a name , variable or type e . g . " hello " or " int "
* - " %num% " Any numeric token , e . g . " 23 "
* - " %op% " Any token such that isOp ( ) returns true .
2010-10-25 03:14:21 +02:00
* - " %or% " A bitwise - or operator ' | '
* - " %oror% " A logical - or operator ' | | '
2015-12-31 01:25:36 +01:00
* - " %type% " Anything that can be a variable type , e . g . " int " , but not " delete " .
* - " %str% " Any token starting with & quot ; - character ( C - string ) .
* - " %var% " Match with token with varId > 0
* - " %varid% " Match with parameter varid
2010-03-13 22:16:06 +01:00
* - " [abc] " Any of the characters ' a ' or ' b ' or ' c '
* - " int|void|char " Any of the strings , int , void or char
* - " int|void|char| " Any of the strings , int , void or char or empty string
* - " !!else " No tokens or any token that is not " else " .
* - " someRandomText " If token contains " someRandomText " .
2008-12-18 22:28:57 +01:00
*
2014-01-01 20:46:00 +01:00
* multi - compare patterns such as " int|void|char " can contain % % or % , % % oror % and % % op %
2015-12-31 01:25:36 +01:00
* it is recommended to put such an % % cmd % as the first pattern .
2012-12-01 02:16:03 +01:00
* For example : " %var%|%num%|) " means yes to a variable , a number or ' ) ' .
*
2008-12-18 22:28:57 +01:00
* The patterns can be also combined to compare to multiple tokens at once
* by separating tokens with a space , e . g .
* " ) const|void { " will return true if first token is ' ) ' next token is either
* " const " or " void " and token after that is ' { ' . If even one of the tokens does not
* match its pattern , false is returned .
*
* @ param tok List of tokens to be compared to the pattern
2008-12-23 22:51:54 +01:00
* @ param pattern The pattern against which the tokens are compared ,
* e . g . " const " or " ) const|volatile| { " .
2014-01-01 20:46:00 +01:00
* @ param varid if % % varid % is given in the pattern the Token : : varId
2012-12-01 02:16:03 +01:00
* will be matched against this argument
2008-12-18 22:28:57 +01:00
* @ return true if given token matches with given pattern
* false if given token does not match with given pattern
*/
2019-07-14 12:22:33 +02:00
static bool Match ( const Token * tok , const char pattern [ ] , nonneg int varid = 0 ) ;
2008-12-18 22:28:57 +01:00
2009-08-30 13:07:10 +02:00
/**
2014-04-27 09:32:02 +02:00
* @ return length of C - string .
2009-08-30 13:07:10 +02:00
*
2014-01-01 20:46:00 +01:00
* Should be called for % % str % % tokens only .
2009-08-30 13:07:10 +02:00
*
* @ param tok token with C - string
* */
2019-07-14 12:22:33 +02:00
static nonneg int getStrLength ( const Token * tok ) ;
2014-08-01 13:12:18 +02:00
Set correct type and size of string and char literals (#2275)
* Set correct type and size of string and char literals
Use that string and char literal tokens store the prefix. This makes
it possible to distinghuish between different type of string literals
(i.e., utf8 encoded strings, utf16, wide strings, etc) which have
different type.
When the tokens holding the string and character values have the correct
type, it is possible to improve Token::getStrSize() to give the correct
result for all string types. Previously, it would return the number of
characters in the string, i.e., it would give the wrong size unless
the type of the string was char*.
Since strings now can have different size (in number of bytes) and
length (in number of elements), add a new helper function that returns
the number of characters. Checkers have been updated to use the correct
functions.
Having the size makes it possible to find more problems with prefixed
strings, and to reduce false positives, for example in the buffer
overflow checker.
Also, improve the stringLiteralWrite error message to also print the
prefix of the string (if there is one).
* Add comment and update string length
2019-10-20 07:11:57 +02:00
/**
* @ return array length of C - string .
*
* Should be called for % % str % % tokens only .
*
* @ param tok token with C - string
* */
static nonneg int getStrArraySize ( const Token * tok ) ;
2014-08-01 13:12:18 +02:00
/**
* @ return sizeof of C - string .
*
* Should be called for % % str % % tokens only .
*
* @ param tok token with C - string
Set correct type and size of string and char literals (#2275)
* Set correct type and size of string and char literals
Use that string and char literal tokens store the prefix. This makes
it possible to distinghuish between different type of string literals
(i.e., utf8 encoded strings, utf16, wide strings, etc) which have
different type.
When the tokens holding the string and character values have the correct
type, it is possible to improve Token::getStrSize() to give the correct
result for all string types. Previously, it would return the number of
characters in the string, i.e., it would give the wrong size unless
the type of the string was char*.
Since strings now can have different size (in number of bytes) and
length (in number of elements), add a new helper function that returns
the number of characters. Checkers have been updated to use the correct
functions.
Having the size makes it possible to find more problems with prefixed
strings, and to reduce false positives, for example in the buffer
overflow checker.
Also, improve the stringLiteralWrite error message to also print the
prefix of the string (if there is one).
* Add comment and update string length
2019-10-20 07:11:57 +02:00
* @ param settings Settings
2014-08-01 13:12:18 +02:00
* */
Set correct type and size of string and char literals (#2275)
* Set correct type and size of string and char literals
Use that string and char literal tokens store the prefix. This makes
it possible to distinghuish between different type of string literals
(i.e., utf8 encoded strings, utf16, wide strings, etc) which have
different type.
When the tokens holding the string and character values have the correct
type, it is possible to improve Token::getStrSize() to give the correct
result for all string types. Previously, it would return the number of
characters in the string, i.e., it would give the wrong size unless
the type of the string was char*.
Since strings now can have different size (in number of bytes) and
length (in number of elements), add a new helper function that returns
the number of characters. Checkers have been updated to use the correct
functions.
Having the size makes it possible to find more problems with prefixed
strings, and to reduce false positives, for example in the buffer
overflow checker.
Also, improve the stringLiteralWrite error message to also print the
prefix of the string (if there is one).
* Add comment and update string length
2019-10-20 07:11:57 +02:00
static nonneg int getStrSize ( const Token * tok , const Settings * const ) ;
2009-08-30 13:07:10 +02:00
2013-01-13 20:52:38 +01:00
/**
2014-04-27 09:32:02 +02:00
* @ return char of C - string at index ( possible escaped " \\ n " )
2013-01-13 20:52:38 +01:00
*
2014-01-01 20:46:00 +01:00
* Should be called for % % str % % tokens only .
2013-01-13 20:52:38 +01:00
*
* @ param tok token with C - string
* @ param index position of character
* */
2019-07-13 16:06:24 +02:00
static std : : string getCharAt ( const Token * tok , MathLib : : bigint index ) ;
2013-01-13 20:52:38 +01:00
2015-10-04 19:42:58 +02:00
const ValueType * valueType ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mValueType ;
2015-10-04 19:42:58 +02:00
}
void setValueType ( ValueType * vt ) ;
2015-10-05 10:12:30 +02:00
const ValueType * argumentType ( ) const {
const Token * top = this ;
while ( top & & ! Token : : Match ( top - > astParent ( ) , " ,|( " ) )
top = top - > astParent ( ) ;
2018-12-21 13:51:45 +01:00
return top ? top - > mImpl - > mValueType : nullptr ;
2015-10-05 10:12:30 +02:00
}
2015-08-14 20:46:13 +02:00
Token : : Type tokType ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType ;
2009-08-23 17:17:57 +02:00
}
2015-08-14 20:46:13 +02:00
void tokType ( Token : : Type t ) {
2018-06-16 16:40:02 +02:00
mTokType = t ;
2017-10-15 01:27:47 +02:00
2018-06-16 16:40:02 +02:00
const bool memoizedIsName = ( mTokType = = eName | | mTokType = = eType | | mTokType = = eVariable | |
mTokType = = eFunction | | mTokType = = eKeyword | | mTokType = = eBoolean | |
mTokType = = eEnumerator ) ; // TODO: "true"/"false" aren't really a name...
2017-10-15 01:27:47 +02:00
setFlag ( fIsName , memoizedIsName ) ;
2018-06-16 16:40:02 +02:00
const bool memoizedIsLiteral = ( mTokType = = eNumber | | mTokType = = eString | | mTokType = = eChar | |
mTokType = = eBoolean | | mTokType = = eLiteral | | mTokType = = eEnumerator ) ;
2017-10-15 01:27:47 +02:00
setFlag ( fIsLiteral , memoizedIsLiteral ) ;
2011-03-08 02:04:25 +01:00
}
2018-05-29 11:46:07 +02:00
void isKeyword ( const bool kwd ) {
2014-06-04 22:33:08 +02:00
if ( kwd )
2017-10-15 01:27:47 +02:00
tokType ( eKeyword ) ;
2018-06-16 16:40:02 +02:00
else if ( mTokType = = eKeyword )
2017-10-15 01:27:47 +02:00
tokType ( eName ) ;
2014-06-04 22:33:08 +02:00
}
2014-11-20 14:20:09 +01:00
bool isKeyword ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eKeyword ;
2014-06-04 22:33:08 +02:00
}
2014-11-20 14:20:09 +01:00
bool isName ( ) const {
2017-10-15 01:27:47 +02:00
return getFlag ( fIsName ) ;
2009-08-23 17:17:57 +02:00
}
2019-08-05 07:18:06 +02:00
bool isNameOnly ( ) const {
return mFlags = = fIsName & & mTokType = = eName ;
}
2012-06-21 19:00:53 +02:00
bool isUpperCaseName ( ) const ;
2014-11-20 14:20:09 +01:00
bool isLiteral ( ) const {
2017-10-15 01:27:47 +02:00
return getFlag ( fIsLiteral ) ;
2013-05-03 06:50:48 +02:00
}
2014-11-20 14:20:09 +01:00
bool isNumber ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eNumber ;
2011-03-08 02:04:25 +01:00
}
2016-04-22 06:02:54 +02:00
bool isEnumerator ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eEnumerator ;
2016-04-22 06:02:54 +02:00
}
2014-11-20 14:20:09 +01:00
bool isOp ( ) const {
2013-02-28 21:50:29 +01:00
return ( isConstOp ( ) | |
isAssignmentOp ( ) | |
2018-06-16 16:40:02 +02:00
mTokType = = eIncDecOp ) ;
2013-02-28 21:50:29 +01:00
}
2014-11-20 14:20:09 +01:00
bool isConstOp ( ) const {
2011-04-09 15:54:36 +02:00
return ( isArithmeticalOp ( ) | |
2018-06-16 16:40:02 +02:00
mTokType = = eLogicalOp | |
mTokType = = eComparisonOp | |
mTokType = = eBitOp ) ;
2011-04-08 19:40:22 +02:00
}
2014-11-20 14:20:09 +01:00
bool isExtendedOp ( ) const {
2013-02-28 21:50:29 +01:00
return isConstOp ( ) | |
2018-06-16 16:40:02 +02:00
mTokType = = eExtendedOp ;
2011-04-08 19:40:22 +02:00
}
2014-11-20 14:20:09 +01:00
bool isArithmeticalOp ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eArithmeticalOp ;
2013-02-28 21:50:29 +01:00
}
2014-11-20 14:20:09 +01:00
bool isComparisonOp ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eComparisonOp ;
2012-11-13 18:30:33 +01:00
}
2014-11-20 14:20:09 +01:00
bool isAssignmentOp ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eAssignmentOp ;
2011-04-09 20:36:05 +02:00
}
2014-11-20 14:20:09 +01:00
bool isBoolean ( ) const {
2018-06-16 16:40:02 +02:00
return mTokType = = eBoolean ;
2011-03-08 02:04:25 +01:00
}
2020-11-26 17:34:42 +01:00
bool isIncDecOp ( ) const {
return mTokType = = eIncDecOp ;
}
2018-07-13 18:52:03 +02:00
bool isBinaryOp ( ) const {
return astOperand1 ( ) ! = nullptr & & astOperand2 ( ) ! = nullptr ;
}
bool isUnaryOp ( const std : : string & s ) const {
return s = = mStr & & astOperand1 ( ) ! = nullptr & & astOperand2 ( ) = = nullptr ;
2018-07-13 16:40:32 +02:00
}
2015-05-24 17:02:00 +02:00
bool isUnaryPreOp ( ) const ;
2012-04-23 21:05:26 +02:00
2014-11-20 14:20:09 +01:00
unsigned int flags ( ) const {
2018-06-16 16:14:34 +02:00
return mFlags ;
2014-05-06 06:35:48 +02:00
}
2018-05-29 11:46:07 +02:00
void flags ( const unsigned int flags_ ) {
2018-06-16 16:14:34 +02:00
mFlags = flags_ ;
2014-05-06 06:35:48 +02:00
}
2014-11-20 14:20:09 +01:00
bool isUnsigned ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsUnsigned ) ;
2010-03-28 15:56:13 +02:00
}
2018-05-29 11:46:07 +02:00
void isUnsigned ( const bool sign ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsUnsigned , sign ) ;
2010-03-28 15:56:13 +02:00
}
2014-11-20 14:20:09 +01:00
bool isSigned ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsSigned ) ;
2010-03-28 15:56:13 +02:00
}
2018-05-29 11:46:07 +02:00
void isSigned ( const bool sign ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsSigned , sign ) ;
2010-03-28 15:56:13 +02:00
}
2014-11-20 14:20:09 +01:00
bool isPointerCompare ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsPointerCompare ) ;
2011-07-07 15:14:33 +02:00
}
2018-05-29 11:46:07 +02:00
void isPointerCompare ( const bool b ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsPointerCompare , b ) ;
2011-07-07 15:14:33 +02:00
}
2014-11-20 14:20:09 +01:00
bool isLong ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsLong ) ;
2010-03-28 15:56:13 +02:00
}
2014-11-20 14:20:09 +01:00
void isLong ( bool size ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsLong , size ) ;
2010-03-28 15:56:13 +02:00
}
2014-11-20 14:20:09 +01:00
bool isStandardType ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsStandardType ) ;
}
2018-05-29 11:46:07 +02:00
void isStandardType ( const bool b ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsStandardType , b ) ;
2012-04-16 15:28:38 +02:00
}
2014-11-20 14:20:09 +01:00
bool isExpandedMacro ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsExpandedMacro ) ;
2011-12-18 13:33:23 +01:00
}
2018-05-29 11:46:07 +02:00
void isExpandedMacro ( const bool m ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsExpandedMacro , m ) ;
2011-12-18 13:33:23 +01:00
}
2014-12-24 10:35:40 +01:00
bool isCast ( ) const {
return getFlag ( fIsCast ) ;
2014-12-23 16:16:14 +01:00
}
2014-12-24 10:35:40 +01:00
void isCast ( bool c ) {
setFlag ( fIsCast , c ) ;
2014-12-23 16:16:14 +01:00
}
2014-11-20 14:20:09 +01:00
bool isAttributeConstructor ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsAttributeConstructor ) ;
2013-08-30 06:27:46 +02:00
}
2018-05-29 11:46:07 +02:00
void isAttributeConstructor ( const bool ac ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsAttributeConstructor , ac ) ;
2013-08-30 06:27:46 +02:00
}
2014-11-20 14:20:09 +01:00
bool isAttributeDestructor ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsAttributeDestructor ) ;
2014-03-14 05:40:17 +01:00
}
2018-05-29 11:46:07 +02:00
void isAttributeDestructor ( const bool value ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsAttributeDestructor , value ) ;
2014-03-14 05:40:17 +01:00
}
2014-11-20 14:20:09 +01:00
bool isAttributeUnused ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsAttributeUnused ) ;
2013-08-30 06:27:46 +02:00
}
2014-11-20 14:20:09 +01:00
void isAttributeUnused ( bool unused ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsAttributeUnused , unused ) ;
2013-08-30 06:27:46 +02:00
}
2014-11-20 14:20:09 +01:00
bool isAttributeUsed ( ) const {
2014-08-06 11:13:58 +02:00
return getFlag ( fIsAttributeUsed ) ;
}
2018-05-29 11:46:07 +02:00
void isAttributeUsed ( const bool unused ) {
2014-08-06 11:13:58 +02:00
setFlag ( fIsAttributeUsed , unused ) ;
}
2014-11-20 14:20:09 +01:00
bool isAttributePure ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsAttributePure ) ;
2014-03-14 05:40:17 +01:00
}
2018-05-29 11:46:07 +02:00
void isAttributePure ( const bool value ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsAttributePure , value ) ;
2014-03-14 05:40:17 +01:00
}
2014-11-20 14:20:09 +01:00
bool isAttributeConst ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsAttributeConst ) ;
2014-03-14 05:40:17 +01:00
}
2014-11-20 14:20:09 +01:00
void isAttributeConst ( bool value ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsAttributeConst , value ) ;
2014-03-14 05:40:17 +01:00
}
2014-12-24 12:50:51 +01:00
bool isAttributeNoreturn ( ) const {
return getFlag ( fIsAttributeNoreturn ) ;
}
2018-05-29 11:46:07 +02:00
void isAttributeNoreturn ( const bool value ) {
2014-12-24 12:50:51 +01:00
setFlag ( fIsAttributeNoreturn , value ) ;
}
2014-11-20 14:20:09 +01:00
bool isAttributeNothrow ( ) const {
2014-05-06 06:35:48 +02:00
return getFlag ( fIsAttributeNothrow ) ;
2014-04-20 20:40:55 +02:00
}
2018-05-29 11:46:07 +02:00
void isAttributeNothrow ( const bool value ) {
2014-05-06 06:35:48 +02:00
setFlag ( fIsAttributeNothrow , value ) ;
2014-04-20 20:40:55 +02:00
}
2016-09-05 17:27:12 +02:00
bool isAttributePacked ( ) const {
return getFlag ( fIsAttributePacked ) ;
}
2018-05-29 11:46:07 +02:00
void isAttributePacked ( const bool value ) {
2016-09-05 17:27:12 +02:00
setFlag ( fIsAttributePacked , value ) ;
}
2018-05-29 21:43:56 +02:00
bool isAttributeNodiscard ( ) const {
return getFlag ( fIsAttributeNodiscard ) ;
}
void isAttributeNodiscard ( const bool value ) {
setFlag ( fIsAttributeNodiscard , value ) ;
}
2020-09-06 11:02:22 +02:00
bool isAttributeMaybeUnused ( ) const {
return getFlag ( fIsAttributeMaybeUnused ) ;
2020-04-12 20:35:54 +02:00
}
2020-09-06 11:02:22 +02:00
void isAttributeMaybeUnused ( const bool value ) {
setFlag ( fIsAttributeMaybeUnused , value ) ;
2020-04-12 20:35:54 +02:00
}
2019-07-12 11:09:24 +02:00
void setCppcheckAttribute ( TokenImpl : : CppcheckAttributes : : Type type , MathLib : : bigint value ) {
mImpl - > setCppcheckAttribute ( type , value ) ;
}
bool getCppcheckAttribute ( TokenImpl : : CppcheckAttributes : : Type type , MathLib : : bigint * value ) const {
return mImpl - > getCppcheckAttribute ( type , value ) ;
}
2020-08-22 11:37:44 +02:00
bool hasCppcheckAttributes ( ) const {
return nullptr ! = mImpl - > mCppcheckAttributes ;
}
2017-11-17 22:10:39 +01:00
bool isControlFlowKeyword ( ) const {
return getFlag ( fIsControlFlowKeyword ) ;
}
2015-07-22 13:57:14 +02:00
bool isOperatorKeyword ( ) const {
return getFlag ( fIsOperatorKeyword ) ;
2015-07-22 13:48:30 +02:00
}
2018-05-29 11:46:07 +02:00
void isOperatorKeyword ( const bool value ) {
2015-07-22 13:57:14 +02:00
setFlag ( fIsOperatorKeyword , value ) ;
2015-07-22 13:48:30 +02:00
}
2015-08-26 13:13:45 +02:00
bool isComplex ( ) const {
return getFlag ( fIsComplex ) ;
}
2018-05-29 11:46:07 +02:00
void isComplex ( const bool value ) {
2015-08-26 13:13:45 +02:00
setFlag ( fIsComplex , value ) ;
}
2016-04-22 06:02:54 +02:00
bool isEnumType ( ) const {
return getFlag ( fIsEnumType ) ;
}
2018-05-29 11:46:07 +02:00
void isEnumType ( const bool value ) {
2016-04-22 06:02:54 +02:00
setFlag ( fIsEnumType , value ) ;
}
2018-12-04 16:52:41 +01:00
bool isAtAddress ( ) const {
return getFlag ( fAtAddress ) ;
}
void isAtAddress ( bool b ) {
setFlag ( fAtAddress , b ) ;
}
2019-08-05 07:18:06 +02:00
bool isIncompleteVar ( ) const {
return getFlag ( fIncompleteVar ) ;
}
void isIncompleteVar ( bool b ) {
setFlag ( fIncompleteVar , b ) ;
}
2019-08-16 07:56:39 +02:00
bool isConstexpr ( ) const {
return getFlag ( fConstexpr ) ;
}
void isConstexpr ( bool b ) {
setFlag ( fConstexpr , b ) ;
}
2019-10-08 17:48:09 +02:00
bool isExternC ( ) const {
return getFlag ( fExternC ) ;
}
void isExternC ( bool b ) {
setFlag ( fExternC , b ) ;
}
2020-09-09 16:22:36 +02:00
bool isSplittedVarDeclComma ( ) const {
return getFlag ( fIsSplitVarDeclComma ) ;
2020-09-06 11:02:22 +02:00
}
2020-09-09 16:22:36 +02:00
void isSplittedVarDeclComma ( bool b ) {
setFlag ( fIsSplitVarDeclComma , b ) ;
}
bool isSplittedVarDeclEq ( ) const {
return getFlag ( fIsSplitVarDeclEq ) ;
}
void isSplittedVarDeclEq ( bool b ) {
setFlag ( fIsSplitVarDeclEq , b ) ;
2020-09-06 11:02:22 +02:00
}
2018-03-12 12:49:27 +01:00
2018-05-02 20:55:11 +02:00
bool isBitfield ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mBits > 0 ;
2018-05-02 20:55:11 +02:00
}
unsigned char bits ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mBits ;
2018-05-02 20:55:11 +02:00
}
2020-05-15 09:02:59 +02:00
std : : set < TemplateSimplifier : : TokenAndName * > * templateSimplifierPointers ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mTemplateSimplifierPointers ;
2018-09-01 11:26:10 +02:00
}
2018-12-21 13:51:45 +01:00
void templateSimplifierPointer ( TemplateSimplifier : : TokenAndName * tokenAndName ) {
2020-05-15 09:02:59 +02:00
if ( ! mImpl - > mTemplateSimplifierPointers )
mImpl - > mTemplateSimplifierPointers = new std : : set < TemplateSimplifier : : TokenAndName * > ;
mImpl - > mTemplateSimplifierPointers - > insert ( tokenAndName ) ;
2018-09-01 11:26:10 +02:00
}
2018-05-29 11:46:07 +02:00
void setBits ( const unsigned char b ) {
2018-12-21 13:51:45 +01:00
mImpl - > mBits = b ;
2018-05-02 20:55:11 +02:00
}
Set correct type and size of string and char literals (#2275)
* Set correct type and size of string and char literals
Use that string and char literal tokens store the prefix. This makes
it possible to distinghuish between different type of string literals
(i.e., utf8 encoded strings, utf16, wide strings, etc) which have
different type.
When the tokens holding the string and character values have the correct
type, it is possible to improve Token::getStrSize() to give the correct
result for all string types. Previously, it would return the number of
characters in the string, i.e., it would give the wrong size unless
the type of the string was char*.
Since strings now can have different size (in number of bytes) and
length (in number of elements), add a new helper function that returns
the number of characters. Checkers have been updated to use the correct
functions.
Having the size makes it possible to find more problems with prefixed
strings, and to reduce false positives, for example in the buffer
overflow checker.
Also, improve the stringLiteralWrite error message to also print the
prefix of the string (if there is one).
* Add comment and update string length
2019-10-20 07:11:57 +02:00
bool isUtf8 ( ) const {
return ( ( ( mTokType = = eString ) & & isPrefixStringCharLiteral ( mStr , ' " ' , " u8 " ) ) | |
( ( mTokType = = eChar ) & & isPrefixStringCharLiteral ( mStr , ' \' ' , " u8 " ) ) ) ;
}
bool isUtf16 ( ) const {
return ( ( ( mTokType = = eString ) & & isPrefixStringCharLiteral ( mStr , ' " ' , " u " ) ) | |
( ( mTokType = = eChar ) & & isPrefixStringCharLiteral ( mStr , ' \' ' , " u " ) ) ) ;
}
bool isUtf32 ( ) const {
return ( ( ( mTokType = = eString ) & & isPrefixStringCharLiteral ( mStr , ' " ' , " U " ) ) | |
( ( mTokType = = eChar ) & & isPrefixStringCharLiteral ( mStr , ' \' ' , " U " ) ) ) ;
}
bool isCChar ( ) const {
return ( ( ( mTokType = = eString ) & & isPrefixStringCharLiteral ( mStr , ' " ' , " " ) ) | |
( ( mTokType = = eChar ) & & isPrefixStringCharLiteral ( mStr , ' \' ' , " " ) & & mStr . length ( ) = = 3 ) ) ;
}
bool isCMultiChar ( ) const {
return ( ( ( mTokType = = eChar ) & & isPrefixStringCharLiteral ( mStr , ' \' ' , " " ) ) & &
( mStr . length ( ) > 3 ) ) ;
}
2018-03-12 12:49:27 +01:00
/**
* @ brief Is current token a template argument ?
*
* Original code :
*
* template < class C > struct S {
* C x ;
* } ;
* S < int > s ;
*
* Resulting code :
*
* struct S < int > {
* int x ; // <- "int" is a template argument
* }
* S < int > s ;
*/
2018-01-11 09:41:22 +01:00
bool isTemplateArg ( ) const {
return getFlag ( fIsTemplateArg ) ;
}
2018-05-29 11:46:07 +02:00
void isTemplateArg ( const bool value ) {
2018-01-11 09:41:22 +01:00
setFlag ( fIsTemplateArg , value ) ;
}
2011-12-18 13:33:23 +01:00
2020-05-26 20:13:56 +02:00
template < size_t count >
static const Token * findsimplematch ( const Token * const startTok , const char ( & pattern ) [ count ] ) {
return findsimplematch ( startTok , pattern , count - 1 ) ;
}
static const Token * findsimplematch ( const Token * const startTok , const char pattern [ ] , size_t pattern_len ) ;
template < size_t count >
static const Token * findsimplematch ( const Token * const startTok , const char ( & pattern ) [ count ] , const Token * const end ) {
return findsimplematch ( startTok , pattern , count - 1 , end ) ;
}
static const Token * findsimplematch ( const Token * const startTok , const char pattern [ ] , size_t pattern_len , const Token * const end ) ;
2019-07-14 12:22:33 +02:00
static const Token * findmatch ( const Token * const startTok , const char pattern [ ] , const nonneg int varId = 0 ) ;
static const Token * findmatch ( const Token * const startTok , const char pattern [ ] , const Token * const end , const nonneg int varId = 0 ) ;
2020-05-26 20:13:56 +02:00
template < size_t count >
static Token * findsimplematch ( Token * const startTok , const char ( & pattern ) [ count ] ) {
return findsimplematch ( startTok , pattern , count - 1 ) ;
}
static Token * findsimplematch ( Token * const startTok , const char pattern [ ] , size_t pattern_len ) {
return const_cast < Token * > ( findsimplematch ( const_cast < const Token * > ( startTok ) , pattern , pattern_len ) ) ;
2013-01-03 22:26:20 +01:00
}
2020-05-26 20:13:56 +02:00
template < size_t count >
static Token * findsimplematch ( Token * const startTok , const char ( & pattern ) [ count ] , const Token * const end ) {
return findsimplematch ( startTok , pattern , count - 1 , end ) ;
2013-01-03 22:26:20 +01:00
}
2020-05-26 20:13:56 +02:00
static Token * findsimplematch ( Token * const startTok , const char pattern [ ] , size_t pattern_len , const Token * const end ) {
return const_cast < Token * > ( findsimplematch ( const_cast < const Token * > ( startTok ) , pattern , pattern_len , end ) ) ;
}
2019-07-14 12:22:33 +02:00
static Token * findmatch ( Token * const startTok , const char pattern [ ] , const nonneg int varId = 0 ) {
2015-08-27 16:14:33 +02:00
return const_cast < Token * > ( findmatch ( const_cast < const Token * > ( startTok ) , pattern , varId ) ) ;
2013-01-03 22:26:20 +01:00
}
2019-07-14 12:22:33 +02:00
static Token * findmatch ( Token * const startTok , const char pattern [ ] , const Token * const end , const nonneg int varId = 0 ) {
2015-08-27 16:14:33 +02:00
return const_cast < Token * > ( findmatch ( const_cast < const Token * > ( startTok ) , pattern , end , varId ) ) ;
2013-01-03 22:26:20 +01:00
}
2008-12-18 22:28:57 +01:00
/**
* Needle is build from multiple alternatives . If one of
* them is equal to haystack , return value is 1. If there
* are no matches , but one alternative to needle is empty
* string , return value is 0. If needle was not found , return
* value is - 1.
*
2017-05-06 11:57:02 +02:00
* @ param tok Current token ( needle )
2009-01-27 20:30:01 +01:00
* @ param haystack e . g . " one|two " or " |one|two "
2014-11-16 19:40:04 +01:00
* @ param varid optional varid of token
2008-12-18 22:28:57 +01:00
* @ return 1 if needle is found from the haystack
* 0 if needle was empty string
* - 1 if needle was not found
*/
2019-07-14 17:31:26 +02:00
static int multiCompare ( const Token * tok , const char * haystack , nonneg int varid ) ;
2008-12-18 22:28:57 +01:00
2019-07-14 12:22:33 +02:00
nonneg int fileIndex ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mFileIndex ;
2017-05-17 14:57:54 +02:00
}
2019-07-14 12:22:33 +02:00
void fileIndex ( nonneg int indexOfFile ) {
2018-12-21 13:51:45 +01:00
mImpl - > mFileIndex = indexOfFile ;
2017-05-17 14:57:54 +02:00
}
2019-07-14 12:22:33 +02:00
nonneg int linenr ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mLineNumber ;
2009-08-23 17:17:57 +02:00
}
2019-07-14 12:22:33 +02:00
void linenr ( nonneg int lineNumber ) {
2018-12-21 13:51:45 +01:00
mImpl - > mLineNumber = lineNumber ;
2009-08-23 17:17:57 +02:00
}
2008-12-18 22:28:57 +01:00
2019-08-18 12:19:05 +02:00
nonneg int column ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mColumn ;
2009-08-23 17:17:57 +02:00
}
2019-08-18 12:19:05 +02:00
void column ( nonneg int c ) {
2018-12-21 13:51:45 +01:00
mImpl - > mColumn = c ;
2009-08-23 17:17:57 +02:00
}
2008-12-18 22:28:57 +01:00
2014-11-20 14:20:09 +01:00
Token * next ( ) const {
2018-06-16 16:16:55 +02:00
return mNext ;
2009-08-23 17:17:57 +02:00
}
2008-12-18 22:28:57 +01:00
/**
* Delete tokens between begin and end . E . g . if begin = 1
* and end = 5 , tokens 2 , 3 and 4 would be erased .
*
* @ param begin Tokens after this will be erased .
* @ param end Tokens before this will be erased .
*/
2009-01-05 16:49:57 +01:00
static void eraseTokens ( Token * begin , const Token * end ) ;
2008-12-18 22:28:57 +01:00
/**
* Insert new token after this token . This function will handle
* relations between next and previous token also .
2010-04-09 21:40:37 +02:00
* @ param tokenStr String for the new token .
2015-12-14 22:04:26 +01:00
* @ param originalNameStr String used for Token : : originalName ( ) .
2012-11-20 02:58:19 +01:00
* @ param prepend Insert the new token before this token when it ' s not
* the first one on the tokens list .
2008-12-18 22:28:57 +01:00
*/
2015-12-14 22:04:26 +01:00
void insertToken ( const std : : string & tokenStr , const std : : string & originalNameStr = emptyString , bool prepend = false ) ;
2013-09-24 06:43:03 +02:00
2014-11-20 14:20:09 +01:00
Token * previous ( ) const {
2018-06-16 16:16:55 +02:00
return mPrevious ;
2009-08-23 17:17:57 +02:00
}
2008-12-18 22:28:57 +01:00
2019-07-14 12:22:33 +02:00
nonneg int varId ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mVarId ;
2009-08-23 17:17:57 +02:00
}
2019-07-14 12:22:33 +02:00
void varId ( nonneg int id ) {
2018-12-21 13:51:45 +01:00
mImpl - > mVarId = id ;
2016-11-26 22:39:47 +01:00
if ( id ! = 0 ) {
2017-10-15 01:27:47 +02:00
tokType ( eVariable ) ;
2016-11-26 22:39:47 +01:00
isStandardType ( false ) ;
} else {
2012-08-26 10:04:22 +02:00
update_property_info ( ) ;
2016-11-26 22:39:47 +01:00
}
2009-08-23 17:17:57 +02:00
}
2008-12-18 22:28:57 +01:00
2020-08-20 18:21:29 +02:00
MathLib : : bigint exprId ( ) const {
return mImpl - > mExprId ;
}
void exprId ( MathLib : : bigint id ) {
mImpl - > mExprId = id ;
}
2008-12-18 22:28:57 +01:00
/**
* For debugging purposes , prints token and all tokens
* followed by it .
* @ param title Title for the printout or use default parameter or 0
* for no title .
*/
2014-02-16 11:47:52 +01:00
void printOut ( const char * title = nullptr ) const ;
2008-12-18 22:28:57 +01:00
2009-11-27 23:04:04 +01:00
/**
* For debugging purposes , prints token and all tokens
* followed by it .
* @ param title Title for the printout or use default parameter or 0
* for no title .
* @ param fileNames Prints out file name instead of file index .
* File index should match the index of the string in this vector .
*/
void printOut ( const char * title , const std : : vector < std : : string > & fileNames ) const ;
2020-05-23 11:30:54 +02:00
/**
* print out tokens
*/
void printLines ( int lines = 5 ) const ;
2009-01-26 23:26:50 +01:00
/**
* Replace token replaceThis with tokens between start and end ,
* including start and end . The replaceThis token is deleted .
2009-06-20 11:54:49 +02:00
* @ param replaceThis This token will be deleted .
2009-01-26 23:26:50 +01:00
* @ param start This will be in the place of replaceThis
* @ param end This is also in the place of replaceThis
*/
static void replace ( Token * replaceThis , Token * start , Token * end ) ;
2020-08-20 18:21:29 +02:00
struct stringifyOptions {
bool varid = false ;
bool exprid = false ;
bool attributes = false ;
bool macro = false ;
bool linenumbers = false ;
bool linebreaks = false ;
bool files = false ;
static stringifyOptions forDebug ( ) {
stringifyOptions options ;
options . attributes = true ;
options . macro = true ;
options . linenumbers = true ;
options . linebreaks = true ;
options . files = true ;
return options ;
}
static stringifyOptions forDebugVarId ( ) {
stringifyOptions options = forDebug ( ) ;
options . varid = true ;
return options ;
}
static stringifyOptions forDebugExprId ( ) {
stringifyOptions options = forDebug ( ) ;
options . exprid = true ;
return options ;
}
} ;
2012-04-16 19:51:07 +02:00
/**
* Stringify a token
* @ param os The result is shifted into that output stream
* @ param varid Print varids . ( Style : " varname@id " )
* @ param attributes Print attributes of tokens like " unsigned " in front of it .
2014-09-29 14:50:00 +02:00
* @ param macro Prints $ in front of the token if it was expanded from a macro .
2012-04-16 19:51:07 +02:00
*/
2020-08-20 18:21:29 +02:00
void stringify ( std : : ostream & os , const stringifyOptions & options ) const ;
2014-09-29 14:50:00 +02:00
void stringify ( std : : ostream & os , bool varid , bool attributes , bool macro ) const ;
2012-04-16 19:51:07 +02:00
/**
* Stringify a list of token , from current instance on .
* @ param varid Print varids . ( Style : " varname@id " )
* @ param attributes Print attributes of tokens like " unsigned " in front of it .
* @ param linenumbers Print line number in front of each line
2014-04-27 09:32:02 +02:00
* @ param linebreaks Insert " \\ n " into string when line number changes
2012-04-16 19:51:07 +02:00
* @ param files print Files as numbers or as names ( if fileNames is given )
* @ param fileNames Vector of filenames . Used ( if given ) to print filenames as strings instead of numbers .
* @ param end Stringification ends before this token is reached . 0 to stringify until end of list .
* @ return Stringified token list as a string
*/
2020-08-20 18:21:29 +02:00
std : : string stringifyList ( const stringifyOptions & options , const std : : vector < std : : string > * fileNames = nullptr , const Token * end = nullptr ) const ;
2017-08-09 20:00:26 +02:00
std : : string stringifyList ( bool varid , bool attributes , bool linenumbers , bool linebreaks , bool files , const std : : vector < std : : string > * fileNames = nullptr , const Token * end = nullptr ) const ;
2012-04-16 19:51:07 +02:00
std : : string stringifyList ( const Token * end , bool attributes = true ) const ;
std : : string stringifyList ( bool varid = false ) const ;
2009-02-13 07:25:29 +01:00
2009-03-13 00:07:05 +01:00
/**
2011-10-16 08:25:11 +02:00
* Remove the contents for this token from the token list .
*
* The contents are replaced with the contents of the next token and
* the next token is unlinked and deleted from the token list .
*
* So this token will still be valid after the ' deleteThis ( ) ' .
2009-03-13 00:07:05 +01:00
*/
void deleteThis ( ) ;
2009-03-13 22:25:56 +01:00
/**
* Create link to given token
2010-06-14 07:54:41 +02:00
* @ param linkToToken The token where this token should link
2009-03-13 22:25:56 +01:00
* to .
*/
2014-11-20 14:20:09 +01:00
void link ( Token * linkToToken ) {
2018-06-16 20:30:09 +02:00
mLink = linkToToken ;
2018-06-16 23:03:15 +02:00
if ( mStr = = " < " | | mStr = = " > " )
2012-08-26 10:23:16 +02:00
update_property_info ( ) ;
2009-08-23 17:17:57 +02:00
}
2009-03-13 22:25:56 +01:00
/**
* Return token where this token links to .
* Supported links are :
* " { " < - > " } "
2011-10-16 08:25:11 +02:00
* " ( " < - > " ) "
* " [ " < - > " ] "
2009-03-13 22:25:56 +01:00
*
* @ return The token where this token links to .
*/
2014-11-20 14:20:09 +01:00
Token * link ( ) const {
2018-06-16 20:30:09 +02:00
return mLink ;
2009-08-23 17:17:57 +02:00
}
2009-03-13 22:25:56 +01:00
2012-08-11 20:47:11 +02:00
/**
* Associate this token with given scope
2012-10-19 11:29:05 +02:00
* @ param s Scope to be associated
2012-08-11 20:47:11 +02:00
*/
2014-11-20 14:20:09 +01:00
void scope ( const Scope * s ) {
2018-12-21 13:51:45 +01:00
mImpl - > mScope = s ;
2012-08-11 20:47:11 +02:00
}
/**
2014-04-27 09:32:02 +02:00
* @ return a pointer to the scope containing this token .
2012-08-11 20:47:11 +02:00
*/
2014-11-20 14:20:09 +01:00
const Scope * scope ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mScope ;
2012-08-11 20:47:11 +02:00
}
2013-01-31 06:41:18 +01:00
/**
* Associate this token with given function
* @ param f Function to be associated
*/
2019-07-05 14:00:59 +02:00
void function ( const Function * f ) ;
2013-01-31 06:41:18 +01:00
/**
2014-04-27 09:32:02 +02:00
* @ return a pointer to the Function associated with this token .
2013-01-31 06:41:18 +01:00
*/
2014-11-20 14:20:09 +01:00
const Function * function ( ) const {
2019-07-05 12:30:42 +02:00
return mTokType = = eFunction | | mTokType = = eLambda ? mImpl - > mFunction : nullptr ;
2013-01-31 06:41:18 +01:00
}
/**
* Associate this token with given variable
* @ param v Variable to be associated
*/
2014-11-20 14:20:09 +01:00
void variable ( const Variable * v ) {
2018-12-21 13:51:45 +01:00
mImpl - > mVariable = v ;
if ( v | | mImpl - > mVarId )
2017-10-15 01:27:47 +02:00
tokType ( eVariable ) ;
2018-06-16 16:40:02 +02:00
else if ( mTokType = = eVariable )
2017-10-15 01:27:47 +02:00
tokType ( eName ) ;
2013-01-31 06:41:18 +01:00
}
/**
2014-04-27 09:32:02 +02:00
* @ return a pointer to the variable associated with this token .
2013-01-31 06:41:18 +01:00
*/
2014-11-20 14:20:09 +01:00
const Variable * variable ( ) const {
2018-12-21 13:51:45 +01:00
return mTokType = = eVariable ? mImpl - > mVariable : nullptr ;
2015-08-14 20:46:13 +02:00
}
/**
* Associate this token with given type
* @ param t Type to be associated
*/
2016-04-22 06:02:54 +02:00
void type ( const : : Type * t ) ;
2015-08-14 20:46:13 +02:00
/**
* @ return a pointer to the type associated with this token .
*/
const : : Type * type ( ) const {
2018-12-21 13:51:45 +01:00
return mTokType = = eType ? mImpl - > mType : nullptr ;
2013-01-31 06:41:18 +01:00
}
2019-05-02 11:04:23 +02:00
static const : : Type * typeOf ( const Token * tok ) ;
static std : : pair < const Token * , const Token * > typeDecl ( const Token * tok ) ;
static std : : string typeStr ( const Token * tok ) ;
2016-04-22 06:02:54 +02:00
/**
* @ return a pointer to the Enumerator associated with this token .
*/
const Enumerator * enumerator ( ) const {
2018-12-21 13:51:45 +01:00
return mTokType = = eEnumerator ? mImpl - > mEnumerator : nullptr ;
2016-04-22 06:02:54 +02:00
}
/**
* Associate this token with given enumerator
* @ param e Enumerator to be associated
*/
void enumerator ( const Enumerator * e ) {
2018-12-21 13:51:45 +01:00
mImpl - > mEnumerator = e ;
2016-04-22 06:02:54 +02:00
if ( e )
2017-10-15 01:27:47 +02:00
tokType ( eEnumerator ) ;
2018-06-16 16:40:02 +02:00
else if ( mTokType = = eEnumerator )
2017-10-15 01:27:47 +02:00
tokType ( eName ) ;
2016-04-22 06:02:54 +02:00
}
2009-08-22 16:22:50 +02:00
/**
* Links two elements against each other .
* */
static void createMutualLinks ( Token * begin , Token * end ) ;
2009-09-12 22:54:47 +02:00
/**
* This can be called only for tokens that are strings , else
* the assert ( ) is called . If Token is e . g . ' " hello " ' , this will return
* ' hello ' ( removing the double quotes ) .
* @ return String value
*/
2009-09-13 10:02:23 +02:00
std : : string strValue ( ) const ;
2009-09-12 22:54:47 +02:00
2009-11-27 22:21:13 +01:00
/**
2011-11-11 21:55:37 +01:00
* Move srcStart and srcEnd tokens and all tokens between them
2009-11-27 22:21:13 +01:00
* into new a location . Only links between tokens are changed .
* @ param srcStart This is the first token to be moved
* @ param srcEnd The last token to be moved
* @ param newLocation srcStart will be placed after this token .
*/
static void move ( Token * srcStart , Token * srcEnd , Token * newLocation ) ;
2019-06-21 22:16:23 +02:00
/** Get progressValue (0 - 100) */
2019-07-14 12:22:33 +02:00
nonneg int progressValue ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mProgressValue ;
2010-08-03 16:36:21 +02:00
}
/** Calculate progress values for all tokens */
2014-05-20 21:55:08 +02:00
static void assignProgressValues ( Token * tok ) ;
2010-08-03 16:36:21 +02:00
2011-10-23 11:23:48 +02:00
/**
2014-04-27 09:32:02 +02:00
* @ return the first token of the next argument . Does only work on argument
2014-08-20 14:57:00 +02:00
* lists . Requires that Tokenizer : : createLinks2 ( ) has been called before .
* Returns 0 , if there is no next argument .
2011-10-23 11:23:48 +02:00
*/
2011-12-07 21:15:00 +01:00
Token * nextArgument ( ) const ;
2011-10-23 11:23:48 +02:00
2014-08-20 14:57:00 +02:00
/**
* @ return the first token of the next argument . Does only work on argument
* lists . Should be used only before Tokenizer : : createLinks2 ( ) was called .
* Returns 0 , if there is no next argument .
*/
Token * nextArgumentBeforeCreateLinks2 ( ) const ;
2015-01-03 20:35:33 +01:00
/**
* @ return the first token of the next template argument . Does only work on template argument
* lists . Requires that Tokenizer : : createLinks2 ( ) has been called before .
* Returns 0 , if there is no next argument .
*/
Token * nextTemplateArgument ( ) const ;
2012-04-18 16:02:03 +02:00
/**
* Returns the closing bracket of opening ' < ' . Should only be used if link ( )
* is unavailable .
2013-07-31 10:30:20 +02:00
* @ return closing ' > ' , ' ) ' , ' ] ' or ' } ' . if no closing bracket is found , NULL is returned
2012-04-18 16:02:03 +02:00
*/
2013-07-31 10:30:20 +02:00
const Token * findClosingBracket ( ) const ;
Token * findClosingBracket ( ) ;
2012-04-18 16:02:03 +02:00
2018-10-14 16:57:07 +02:00
const Token * findOpeningBracket ( ) const ;
Token * findOpeningBracket ( ) ;
2013-08-31 06:26:39 +02:00
/**
2014-04-27 09:32:02 +02:00
* @ return the original name .
2013-08-31 06:26:39 +02:00
*/
2014-11-20 14:20:09 +01:00
const std : : string & originalName ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mOriginalName ? * mImpl - > mOriginalName : emptyString ;
2013-08-31 06:26:39 +02:00
}
2017-03-27 18:48:34 +02:00
const std : : list < ValueFlow : : Value > & values ( ) const {
2019-09-20 21:54:30 +02:00
return mImpl - > mValues ? * mImpl - > mValues : TokenImpl : : mEmptyValueList ;
2017-03-27 18:48:34 +02:00
}
2013-08-31 06:26:39 +02:00
/**
* Sets the original name .
*/
2014-04-02 18:04:20 +02:00
template < typename T >
2014-11-20 14:20:09 +01:00
void originalName ( T & & name ) {
2018-12-21 13:51:45 +01:00
if ( ! mImpl - > mOriginalName )
mImpl - > mOriginalName = new std : : string ( name ) ;
2014-06-26 10:57:39 +02:00
else
2018-12-21 13:51:45 +01:00
* mImpl - > mOriginalName = name ;
2013-08-31 06:26:39 +02:00
}
2020-05-10 20:32:59 +02:00
bool hasKnownIntValue ( ) const ;
bool hasKnownValue ( ) const ;
2018-03-24 07:58:37 +01:00
2019-03-06 07:08:36 +01:00
MathLib : : bigint getKnownIntValue ( ) const {
return mImpl - > mValues - > front ( ) . intvalue ;
}
2020-05-10 20:32:59 +02:00
bool isImpossibleIntValue ( const MathLib : : bigint val ) const ;
2014-01-21 16:58:23 +01:00
2020-05-10 20:32:59 +02:00
const ValueFlow : : Value * getValue ( const MathLib : : bigint val ) const ;
const ValueFlow : : Value * getMaxValue ( bool condition ) const ;
const ValueFlow : : Value * getMovedValue ( ) const ;
2016-11-20 15:14:49 +01:00
2014-04-02 06:49:28 +02:00
const ValueFlow : : Value * getValueLE ( const MathLib : : bigint val , const Settings * settings ) const ;
const ValueFlow : : Value * getValueGE ( const MathLib : : bigint val , const Settings * settings ) const ;
2019-07-14 12:22:33 +02:00
const ValueFlow : : Value * getInvalidValue ( const Token * ftok , nonneg int argnr , const Settings * settings ) const ;
2017-04-20 22:14:54 +02:00
2020-05-10 20:32:59 +02:00
const ValueFlow : : Value * getContainerSizeValue ( const MathLib : : bigint val ) const ;
2018-08-10 11:29:16 +02:00
2014-08-04 08:25:10 +02:00
const Token * getValueTokenMaxStrLength ( ) const ;
Set correct type and size of string and char literals (#2275)
* Set correct type and size of string and char literals
Use that string and char literal tokens store the prefix. This makes
it possible to distinghuish between different type of string literals
(i.e., utf8 encoded strings, utf16, wide strings, etc) which have
different type.
When the tokens holding the string and character values have the correct
type, it is possible to improve Token::getStrSize() to give the correct
result for all string types. Previously, it would return the number of
characters in the string, i.e., it would give the wrong size unless
the type of the string was char*.
Since strings now can have different size (in number of bytes) and
length (in number of elements), add a new helper function that returns
the number of characters. Checkers have been updated to use the correct
functions.
Having the size makes it possible to find more problems with prefixed
strings, and to reduce false positives, for example in the buffer
overflow checker.
Also, improve the stringLiteralWrite error message to also print the
prefix of the string (if there is one).
* Add comment and update string length
2019-10-20 07:11:57 +02:00
const Token * getValueTokenMinStrSize ( const Settings * settings ) const ;
2014-08-03 20:11:22 +02:00
2014-08-05 06:24:23 +02:00
const Token * getValueTokenDeadPointer ( ) const ;
2017-03-27 18:48:34 +02:00
/** Add token value. Return true if value is added. */
bool addValue ( const ValueFlow : : Value & value ) ;
2020-02-01 08:20:08 +01:00
void removeValues ( std : : function < bool ( const ValueFlow : : Value & ) > pred ) {
2019-05-05 09:51:36 +02:00
if ( mImpl - > mValues )
mImpl - > mValues - > remove_if ( pred ) ;
}
2019-07-14 12:22:33 +02:00
nonneg int index ( ) const {
2019-06-21 22:16:23 +02:00
return mImpl - > mIndex ;
}
void assignIndexes ( ) ;
2008-12-18 22:28:57 +01:00
private :
2014-01-21 16:58:23 +01:00
2014-11-20 14:20:09 +01:00
void next ( Token * nextToken ) {
2018-06-16 16:16:55 +02:00
mNext = nextToken ;
2009-08-23 17:17:57 +02:00
}
2014-11-20 14:20:09 +01:00
void previous ( Token * previousToken ) {
2018-06-16 16:16:55 +02:00
mPrevious = previousToken ;
2009-08-23 17:17:57 +02:00
}
2008-12-18 22:28:57 +01:00
2017-08-25 23:30:04 +02:00
/** used by deleteThis() to take data from token to delete */
void takeData ( Token * fromToken ) ;
2009-08-16 22:28:17 +02:00
/**
2012-04-16 15:28:38 +02:00
* Works almost like strcmp ( ) except returns only true or false and
2010-03-13 22:16:06 +01:00
* if str has empty space & apos ; & apos ; character , that character is handled
* as if it were & apos ; \ \ 0 & apos ;
2009-08-16 22:28:17 +02:00
*/
2012-04-16 15:28:38 +02:00
static bool firstWordEquals ( const char * str , const char * word ) ;
2009-08-16 22:28:17 +02:00
/**
* Works almost like strchr ( ) except
2010-03-13 22:16:06 +01:00
* if str has empty space & apos ; & apos ; character , that character is handled
* as if it were & apos ; \ \ 0 & apos ;
2009-08-16 22:28:17 +02:00
*/
static const char * chrInFirstWord ( const char * str , char c ) ;
2018-06-16 23:03:15 +02:00
std : : string mStr ;
2009-08-16 22:28:17 +02:00
2018-06-16 16:16:55 +02:00
Token * mNext ;
Token * mPrevious ;
2018-06-16 20:30:09 +02:00
Token * mLink ;
2012-08-11 20:47:11 +02:00
2014-05-06 06:35:48 +02:00
enum {
fIsUnsigned = ( 1 < < 0 ) ,
fIsSigned = ( 1 < < 1 ) ,
fIsPointerCompare = ( 1 < < 2 ) ,
fIsLong = ( 1 < < 3 ) ,
fIsStandardType = ( 1 < < 4 ) ,
fIsExpandedMacro = ( 1 < < 5 ) ,
2014-12-24 10:35:40 +01:00
fIsCast = ( 1 < < 6 ) ,
2014-12-23 16:16:14 +01:00
fIsAttributeConstructor = ( 1 < < 7 ) , // __attribute__((constructor)) __attribute__((constructor(priority)))
fIsAttributeDestructor = ( 1 < < 8 ) , // __attribute__((destructor)) __attribute__((destructor(priority)))
fIsAttributeUnused = ( 1 < < 9 ) , // __attribute__((unused))
fIsAttributePure = ( 1 < < 10 ) , // __attribute__((pure))
fIsAttributeConst = ( 1 < < 11 ) , // __attribute__((const))
2015-01-09 20:18:09 +01:00
fIsAttributeNoreturn = ( 1 < < 12 ) , // __attribute__((noreturn)), __declspec(noreturn)
fIsAttributeNothrow = ( 1 < < 13 ) , // __attribute__((nothrow)), __declspec(nothrow)
2015-07-22 13:48:30 +02:00
fIsAttributeUsed = ( 1 < < 14 ) , // __attribute__((used))
2016-09-05 17:27:12 +02:00
fIsAttributePacked = ( 1 < < 15 ) , // __attribute__((packed))
2020-09-06 11:02:22 +02:00
fIsAttributeMaybeUnused = ( 1 < < 16 ) , // [[maybe_unsed]]
fIsControlFlowKeyword = ( 1 < < 17 ) , // if/switch/while/...
fIsOperatorKeyword = ( 1 < < 18 ) , // operator=, etc
fIsComplex = ( 1 < < 19 ) , // complex/_Complex type
fIsEnumType = ( 1 < < 20 ) , // enumeration type
fIsName = ( 1 < < 21 ) ,
fIsLiteral = ( 1 < < 22 ) ,
fIsTemplateArg = ( 1 < < 23 ) ,
fIsAttributeNodiscard = ( 1 < < 24 ) , // __attribute__ ((warn_unused_result)), [[nodiscard]]
fAtAddress = ( 1 < < 25 ) , // @ 0x4000
fIncompleteVar = ( 1 < < 26 ) ,
fConstexpr = ( 1 < < 27 ) ,
fExternC = ( 1 < < 28 ) ,
2020-09-09 16:22:36 +02:00
fIsSplitVarDeclComma = ( 1 < < 29 ) , // set to true when variable declarations are split up ('int a,b;' => 'int a; int b;')
fIsSplitVarDeclEq = ( 1 < < 30 ) // set to true when variable declaration with initialization is split up ('int a=5;' => 'int a; a=5;')
2014-05-06 06:35:48 +02:00
} ;
2018-12-21 13:51:45 +01:00
Token : : Type mTokType ;
2018-12-28 12:59:05 +01:00
unsigned int mFlags ;
2018-12-21 13:51:45 +01:00
TokenImpl * mImpl ;
2014-05-06 06:35:48 +02:00
/**
* Get specified flag state .
* @ param flag_ flag to get state of
* @ return true if flag set or false in flag not set
*/
2014-11-20 14:20:09 +01:00
bool getFlag ( unsigned int flag_ ) const {
2018-06-16 16:14:34 +02:00
return ( ( mFlags & flag_ ) ! = 0 ) ;
2014-05-06 06:35:48 +02:00
}
/**
* Set specified flag state .
* @ param flag_ flag to set state
* @ param state_ new state of flag
*/
2014-11-20 14:20:09 +01:00
void setFlag ( unsigned int flag_ , bool state_ ) {
2018-06-16 16:14:34 +02:00
mFlags = state_ ? mFlags | flag_ : mFlags & ~ flag_ ;
2014-05-06 06:35:48 +02:00
}
2010-08-05 21:40:04 +02:00
2011-10-23 20:38:03 +02:00
/** Updates internal property cache like _isName or _isBoolean.
2018-06-16 23:03:15 +02:00
Called after any mStr ( ) modification . */
2011-10-23 20:38:03 +02:00
void update_property_info ( ) ;
2011-11-09 21:45:59 +01:00
/** Update internal property cache about isStandardType() */
void update_property_isStandardType ( ) ;
2012-12-15 20:21:09 +01:00
2019-03-18 06:18:25 +01:00
/** Update internal property cache about string and char literals */
void update_property_char_string_literal ( ) ;
Optimize astStringVerbose() for large arrays (#1815)
Change the astStringVerbose() recursion to extend a string instead of
returning one. This has the benefit that for tokens where the recursion
runs deep (typically large arrays), the time savings can be substantial
(see comments on benchmarks further down).
The reason is that previously, for each token, the astString of its
operands was constructed, and then appended to this tokens astString.
This led to a lot of unnecessary string copying (and with that
allocations). Instead, by passing the string by reference, the number
of temporary strings is greatly reduced.
Another way of seeing it is that previously, the string was constructed
from end to beginning, but now it is constructed from the beginning to
end. There was no notable speedup by preallocating the entire string
using string::reserve() (at least not on Linux).
To benchmark, the changes and master were tested on Linux using the
commands:
make
time cppcheck --debug --verbose $file >/dev/null
i.e., the cppcheck binary was compiled with the settings in the
Makefile. Printing the output to screen or file will of course take
longer time.
In Trac ticket #8355 which triggered this change, an example file from the
Wine repository was attached. Running the above cppcheck on master took
24 minutes and with the changes in this commmit, took 22 seconds.
Another test made was on lib/tokenlist.cpp in the cppcheck repo, which is
more "normal" file. On that file there was no measurable time difference.
A synthetic benchmark was generated to illustrate the effects on dumping
the ast for arrays of different sizes. The generate code looked as
follows:
const int array[] = {...};
with different number of elements. The results are as follows (times are
in seconds):
N master optimized
10 0.1 0.1
100 0.1 0.1
1000 2.8 0.7
2000 19 1.8
3000 53 3.8
5000 350 10
10000 3215 38
As we can see, for small arrays, there is no time difference, but for
large arrays the time savings are substantial.
2019-04-30 13:35:48 +02:00
/** Internal helper function to avoid excessive string allocations */
2019-07-14 12:22:33 +02:00
void astStringVerboseRecursive ( std : : string & ret , const nonneg int indent1 = 0 , const nonneg int indent2 = 0 ) const ;
Optimize astStringVerbose() for large arrays (#1815)
Change the astStringVerbose() recursion to extend a string instead of
returning one. This has the benefit that for tokens where the recursion
runs deep (typically large arrays), the time savings can be substantial
(see comments on benchmarks further down).
The reason is that previously, for each token, the astString of its
operands was constructed, and then appended to this tokens astString.
This led to a lot of unnecessary string copying (and with that
allocations). Instead, by passing the string by reference, the number
of temporary strings is greatly reduced.
Another way of seeing it is that previously, the string was constructed
from end to beginning, but now it is constructed from the beginning to
end. There was no notable speedup by preallocating the entire string
using string::reserve() (at least not on Linux).
To benchmark, the changes and master were tested on Linux using the
commands:
make
time cppcheck --debug --verbose $file >/dev/null
i.e., the cppcheck binary was compiled with the settings in the
Makefile. Printing the output to screen or file will of course take
longer time.
In Trac ticket #8355 which triggered this change, an example file from the
Wine repository was attached. Running the above cppcheck on master took
24 minutes and with the changes in this commmit, took 22 seconds.
Another test made was on lib/tokenlist.cpp in the cppcheck repo, which is
more "normal" file. On that file there was no measurable time difference.
A synthetic benchmark was generated to illustrate the effects on dumping
the ast for arrays of different sizes. The generate code looked as
follows:
const int array[] = {...};
with different number of elements. The results are as follows (times are
in seconds):
N master optimized
10 0.1 0.1
100 0.1 0.1
1000 2.8 0.7
2000 19 1.8
3000 53 3.8
5000 350 10
10000 3215 38
As we can see, for small arrays, there is no time difference, but for
large arrays the time savings are substantial.
2019-04-30 13:35:48 +02:00
2012-12-15 20:21:09 +01:00
public :
void astOperand1 ( Token * tok ) ;
void astOperand2 ( Token * tok ) ;
2019-06-16 10:09:38 +02:00
Token * astOperand1 ( ) {
return mImpl - > mAstOperand1 ;
}
2014-11-20 14:20:09 +01:00
const Token * astOperand1 ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mAstOperand1 ;
2012-12-15 20:21:09 +01:00
}
2019-06-16 10:09:38 +02:00
Token * astOperand2 ( ) {
return mImpl - > mAstOperand2 ;
}
2014-11-20 14:20:09 +01:00
const Token * astOperand2 ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mAstOperand2 ;
2012-12-15 20:21:09 +01:00
}
2019-06-16 10:09:38 +02:00
Token * astParent ( ) {
return mImpl - > mAstParent ;
}
2014-11-20 14:20:09 +01:00
const Token * astParent ( ) const {
2018-12-21 13:51:45 +01:00
return mImpl - > mAstParent ;
2013-12-12 15:33:31 +01:00
}
2020-09-07 07:53:41 +02:00
Token * astSibling ( ) {
if ( ! astParent ( ) )
return nullptr ;
if ( this = = astParent ( ) - > astOperand1 ( ) )
return astParent ( ) - > astOperand2 ( ) ;
else if ( this = = astParent ( ) - > astOperand2 ( ) )
return astParent ( ) - > astOperand1 ( ) ;
return nullptr ;
2020-09-07 10:56:02 +02:00
2020-09-07 07:53:41 +02:00
}
const Token * astSibling ( ) const {
if ( ! astParent ( ) )
return nullptr ;
if ( this = = astParent ( ) - > astOperand1 ( ) )
return astParent ( ) - > astOperand2 ( ) ;
else if ( this = = astParent ( ) - > astOperand2 ( ) )
return astParent ( ) - > astOperand1 ( ) ;
return nullptr ;
2020-09-07 10:56:02 +02:00
2020-09-07 07:53:41 +02:00
}
2019-06-16 10:09:38 +02:00
Token * astTop ( ) {
Token * ret = this ;
while ( ret - > mImpl - > mAstParent )
ret = ret - > mImpl - > mAstParent ;
return ret ;
}
2014-11-20 14:20:09 +01:00
const Token * astTop ( ) const {
2012-12-15 20:21:09 +01:00
const Token * ret = this ;
2018-12-21 13:51:45 +01:00
while ( ret - > mImpl - > mAstParent )
ret = ret - > mImpl - > mAstParent ;
2012-12-15 20:21:09 +01:00
return ret ;
}
2018-10-18 12:09:55 +02:00
std : : pair < const Token * , const Token * > findExpressionStartEndTokens ( ) const ;
2013-12-28 11:02:39 +01:00
/**
* Is current token a calculation ? Only true for operands .
* For ' * ' and ' & ' tokens it is looked up if this is a
* dereference or address - of . A dereference or address - of is not
* counted as a calculation .
* @ return returns true if current token is a calculation
*/
2015-09-09 14:46:47 +02:00
bool isCalculation ( ) const ;
2013-12-28 11:02:39 +01:00
2014-11-20 14:20:09 +01:00
void clearAst ( ) {
2018-12-21 13:51:45 +01:00
mImpl - > mAstOperand1 = mImpl - > mAstOperand2 = mImpl - > mAstParent = nullptr ;
2013-11-07 14:38:08 +01:00
}
2017-03-27 18:48:34 +02:00
void clearValueFlow ( ) {
2018-12-21 13:51:45 +01:00
delete mImpl - > mValues ;
mImpl - > mValues = nullptr ;
2017-03-27 18:48:34 +02:00
}
2014-11-20 14:20:09 +01:00
std : : string astString ( const char * sep = " " ) const {
2012-12-15 20:21:09 +01:00
std : : string ret ;
2018-12-21 13:51:45 +01:00
if ( mImpl - > mAstOperand1 )
ret = mImpl - > mAstOperand1 - > astString ( sep ) ;
if ( mImpl - > mAstOperand2 )
ret + = mImpl - > mAstOperand2 - > astString ( sep ) ;
2018-06-16 23:03:15 +02:00
return ret + sep + mStr ;
2012-12-15 20:21:09 +01:00
}
2013-11-02 18:37:35 +01:00
Optimize astStringVerbose() for large arrays (#1815)
Change the astStringVerbose() recursion to extend a string instead of
returning one. This has the benefit that for tokens where the recursion
runs deep (typically large arrays), the time savings can be substantial
(see comments on benchmarks further down).
The reason is that previously, for each token, the astString of its
operands was constructed, and then appended to this tokens astString.
This led to a lot of unnecessary string copying (and with that
allocations). Instead, by passing the string by reference, the number
of temporary strings is greatly reduced.
Another way of seeing it is that previously, the string was constructed
from end to beginning, but now it is constructed from the beginning to
end. There was no notable speedup by preallocating the entire string
using string::reserve() (at least not on Linux).
To benchmark, the changes and master were tested on Linux using the
commands:
make
time cppcheck --debug --verbose $file >/dev/null
i.e., the cppcheck binary was compiled with the settings in the
Makefile. Printing the output to screen or file will of course take
longer time.
In Trac ticket #8355 which triggered this change, an example file from the
Wine repository was attached. Running the above cppcheck on master took
24 minutes and with the changes in this commmit, took 22 seconds.
Another test made was on lib/tokenlist.cpp in the cppcheck repo, which is
more "normal" file. On that file there was no measurable time difference.
A synthetic benchmark was generated to illustrate the effects on dumping
the ast for arrays of different sizes. The generate code looked as
follows:
const int array[] = {...};
with different number of elements. The results are as follows (times are
in seconds):
N master optimized
10 0.1 0.1
100 0.1 0.1
1000 2.8 0.7
2000 19 1.8
3000 53 3.8
5000 350 10
10000 3215 38
As we can see, for small arrays, there is no time difference, but for
large arrays the time savings are substantial.
2019-04-30 13:35:48 +02:00
std : : string astStringVerbose ( ) const ;
2014-02-24 17:22:29 +01:00
2020-05-18 19:31:13 +02:00
std : : string astStringZ3 ( ) const ;
2014-01-17 18:37:49 +01:00
std : : string expressionString ( ) const ;
2020-10-31 18:57:48 +01:00
void printAst ( bool verbose , bool xml , const std : : vector < std : : string > & fileNames , std : : ostream & out ) const ;
2014-01-18 09:58:32 +01:00
2014-07-14 15:51:45 +02:00
void printValueFlow ( bool xml , std : : ostream & out ) const ;
2019-07-31 11:11:01 +02:00
2019-07-31 09:19:27 +02:00
void scopeInfo ( std : : shared_ptr < ScopeInfo2 > newScopeInfo ) ;
std : : shared_ptr < ScopeInfo2 > scopeInfo ( ) const ;
2019-09-02 20:31:01 +02:00
2019-09-03 06:43:08 +02:00
void setCpp11init ( bool cpp11init ) const {
mImpl - > mCpp11init = cpp11init ? TokenImpl : : Cpp11init : : CPP11INIT : TokenImpl : : Cpp11init : : NOINIT ;
}
TokenImpl : : Cpp11init isCpp11init ( ) const {
return mImpl - > mCpp11init ;
}
2008-12-18 22:28:57 +01:00
} ;
2009-07-17 10:49:01 +02:00
/// @}
2013-09-04 20:59:49 +02:00
//---------------------------------------------------------------------------
# endif // tokenH