* Code changes for Token::mImpl optimisation
* Added new TokenImpl optimisation
Moving members to the TokenImpl struct reduces the size of the Token class, which is a fairly significant optimisation. In my testing on Windows with 32-bit Release-PCRE, this change reduced the size of the Token class from 108 bits to 52 bits and reduced run-time of my test case by around 20%.
* Several optimisations
Deleted some code that ran very slowly and did nothing, as there is no need to change a Token's string to null if you are about to delete it.
Added a frontToken to simplifyCalculations to reduce the amount of work it has to do on already-simplified calculations.
Moved template removal to the end of the list as this reduces redundant iteration and saves time.
* Added tok argument to simplifyCalculations
This means callers can avoid unnecessary work if they know which tokens have already been simplified. Passing nullptr indicates the original behaviour (starting from the front of the list).
* Removed mention of member from another change
* Re-added and optimised some code deleted in error
Changing mTemplateInstantiations to a vector avoids the high cost of doing repeated linear searches. Changing how the code iterates through the array was necessary because the vector can be resized at several points during the loop, which breaks existing references and iterators.
* Changed mTemplateInstantiations to a vector
This is an optimisation that makes repeated linear searches of this collection significantly faster.
Also added a copy constructor to TokenAndName so code can make copies of these objects to keep a reference if a vector gets resized.
* A cleaner optimisation to removing template tokens
This reverts the previous change to made mInstantiatedTemplates a vector and the iterator changes to support this, and makes mTypesUsedInTemplateInstantiation so the eraseTokens logic can be unified.
* Reverted vector to list
Also made mTypesUsedInTemplateInstantiation a vector of TokenAndName objects so it can share the same logic as the other members.
* Added member for template simplifier pointer
This can be used more efficiently than marking Tokens with a flag and then searching through all templates to find the one that matches.
* Turned loop inside out
This means we only have to iterate through the std::list once. std::list is very expensive to iterate through.
* Latest code from danmar and fixed optimisations
In particular I have optimised simplifying template instantiation names as this was incredibly slow because of the number of times it had to iterate through the template instantiation list. Previous optimisations to this weren't very effective and broke some edge cases.
* Added changes from danmar
Made mExplicitInstantiationsToDelete a vector of TokenAndName to be consistent with the rest of the members, which are cleaned up very efficiently.
* Tokens can have many templateSimplifierPointers
* templateSimplifierPointers must be kept in sync
* Fixed#8693 (Template specialization: Constructor detected as normal function (functionStatic error))
Refactor template simplifier to remove the existing full specialization
function expandSpecialized and allow full specializations to use the
existing function expandTemplate. The function expandTemplate was
modified to either expand the template like it originally did by copying
it or to modify the template in place. Both instantiated and
uninstantiated full specializations are modified in place. This also
fixes#8692 and probably other related tickets as well.
The function simplifyTemplates now tries twice to simplify templates so
more templates can be simplified. We should try as many times as
necessary to find all possible templates. We can't do that now because
uninstantiated templates are left unchanged. It is relatively straight
forward to have the new code also expand in place uninstantiated
templates with their symbolic types but namespaces are not handled
properly (ticket #8671) and it would introduce regressions.
* Fix travis warnings.
* Add tests for invalid ranges
* Refactor loadLibErrors
This reduces the amount of code slightly and will simplify adding
more tests.
* Handle empty valid field
Before this change, the sequence <valid></valid> in a config file would
result in a segmentation fault. Now an empty field results in the error
message:
cppcheck: Failed to load library configuration file 'mycfg.cfg'. Bad attribute value '""'
* Add support for valid for floating point arguments
Previously, it was not possible to add valid ranges to floating point
arguments since it only handled integers. This made ranges not work well
for floating point arguments since arguments were cast to integers
before the ranges were handled.
Fix this by using doubles instead of integers if the argument is a float.
Add some tests for this and make sure errors are printed with enough
precision (somewhat arbitrarily chosen).
Note that it is still only possible to add integer ranges (i.e. -1:1).
* Add support for floats in configuration valid range
Now that it is possible to handle decimal arguments, there is no reason
to not allow non-integer ranges. Take care to not allow broken
configurations.
* Move check to within if-clause
* Move asin{,f,l} and acos{,f,l} input checks to config file
* Added declaration for deletePrevious function
* Added definition for deletePrevious function
* Fixed crash from deleteThis invalidating pointers
The crash was caused by deleteThis() invalidating the pointer to a constant variable usage. This happened when a usage followed an assignment. This fixes bug #8579.
* Added tokensFront to match tokensBack
This means deletePrevious can set the list's front if necessary.
* Initialised tokensFront in appropriate places
* Switched to using default Token constructor
* Switched to using Token default constructor
* Switched to using default constructor for Token
* Added missing argument to Token constructor
* Changed to use default constructor for Tokens
* Switched to using default constructor for Tokens
* Switched to using default constructor for Token
* Added new test for deleting front Token
Also made sure to use the correct constructor for Token in other tests.
* Syntax error
* Replaced tokensFront and tokensBack with a struct
This decreases the size of the Token class for performance purposes.
* Replaced tokensFront and tokensBack with a struct
* Added tokensFrontBack to destructor
* Reworked to use TokensBackFront struct
Also ran astyle.
* Reworked to use TokenList's TokensFrontBack member
* Reworked to use TokensFrontBack struct
* Reworked to use TokensFrontBack struct
* Reworked to work with TokensFrontBack struct
* Removed unnecessary scope operator
* Added missing parentheses
* Fixed syntax error
* Removed unnecessary constructor
* Default constructor now 0-initialises everything
This is safer for not using a temporary TokensFrontBack object, and doesn't use delegating constructors which aren't supported yet.
* Fixed unsafe null check
* Added missing explicit keyword
* Fixing stylistic nits
Removed default constructor as it has been superseded by the single-argument constructor with a default argument value.
Renamed listEnds to tokensFrontBack.
Fixed if statement that was supposed to be adding safety but would actually cause a crash if tokensFrontBack was null.
* Fixing stylistic nits
Removed default constructor and replaced it with a single-argument constructor with a default value.
* Fixing stylistic nits
Renamed _listEnds to _tokensFrontBack.
* Fixing stylistic nits
Renamed _listEnds to _tokensFrontBack.
Three attributes are reset in this member function even if their values
are stored in the same token object.
Move a corresponding sanity check so that less assignments could be
performed eventually.
Link: https://trac.cppcheck.net/ticket/8532
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Before:
[/tmp/test.c:8]: (style) Checking if unsigned variable '.' is less than zero.
[/tmp/test.c:12]: (style) Checking if unsigned variable '.' is less than zero.
After:
[/tmp/test.c:8]: (style) Checking if unsigned variable 'd.n' is less than zero.
[/tmp/test.c:12]: (style) Checking if unsigned variable 'd.n' is less than zero.
- Removed simplifyConst() because it did the opposite of the (superior) simplifyStaticConst()
- Execute simplifyStaticConst() in simplifyTokenList1() - there is no reason to defer it, and it is required to properly parse declarations like "unsigned static int i;"
- Fixed simplifyStaticConst() to handle more patterns. It did not work at the beginning of the token list and for function arguments
- Reimplemented Tokenizer::simplifyStdType()
-> properly support all possible ways to declare integers as requested by the standard, instead of only a few common permutations of "unsigned|signed", "short|char|long|int"
-> Fixed parsing of _Complex/complex types
1) Added global static const std::string emptyString; object:
-> Replaces some static variables in functions which might be not threadsafe
-> Avoids constructor call (std::string::string(""))
-> Even functions that return an empty string in some branches can return by reference now.
Added to config.h to ensure that it is available everywhere
2) Added overloads for TestFixture::assertEquals for the most common use cases:
-> Moves conversion from const char[] to std::string into a function, reducing code duplication in binary.
- Encapsulate parameters passed through AST compiler functions in a struct
-> Reduces maintenance overhead when changing them
-> Contains parameter indicating if the file is C or C++
- Added eKeyword Token type (at the moment, only true for C++ keyword "delete", but should be set for other keywords as well)
-> Performance gain of up to 15% on entire checking time (depends on setup; Result was checked with VS12 (x64), matchcompiled version, ran on tinyxml and cppcheck itself)
- Properly set Token::_type to eVariable, eFunction or eName, depending on _function, _variable or _varId being set.
- Token can't be a variable and a function at a time - put Token::_function and Token::_variable into a union.
- Rename Token::isOp() to Token::isConstOp() (indicating that the operator does _not_ modify the input variables)
- Create new Token::isOp(), returning true also for ++, -- and assignment operators
- Make Token::isExtendedOp() returning also true for all assignment and ++/-- operators
1. fix typos / misspellings
- Fix misspelling within comments, variable/function names, stdout messages
- changes the name of an error code: ```stlBoundries``` changed to ```stlBoundaries```. Alias old name (```stlBoundries```) to the new one.
2. fix gcc v3.4.6 32bit & 64bit warnings
- fixes gcc v3.4.6 warnings, except for those in tinyxml and "-Wmissing-declarations" makefile warnings
- in Preprocessor::handleIncludes(), replace a ```vector <bool>``` with ```stack<bool>``` (see ```vector<bool>``` warning below).
- this is the only ```vector<bool>``` in the codebase
- ```vector <bool>``` is actually a case of template specialization, and is not recommended, according to the following links:
http://stackoverflow.com/q/6461487http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2160.htmlhttp://stackoverflow.com/q/670308
- in the codebase before and after this change, testrunner SEGVs in a number of places on gcc v3.4.6, including ```Check::~Check()```, among others
- fc42fc95 fixes this particular runtime issue for DJGPP & __sun
- fix g++ warning:
> lib/checkother.cpp:3779: warning: comparison between signed and unsigned integer expressions
- fix suncc warning (see [everything2](http://everything2.com/title/C%252B%252B%253A+static+extern+%2522C%2522)):
> "lib/checkmemoryleak.cpp", line 578: Warning (Anachronism): Formal argument __compar of type extern "C" int(*)(const void*,const void*) in call to bsearch(const void*, const void*, unsigned long, unsigned long, extern "C" int(*)(const void*,const void*)) is being passed int(*)(const void*,const void*).
- prefer empty() / isEmpty() over "size() > 0" (cases not caught by stlSize)
- fix word misspellings (mostly comments, a few output lines)
- Parenthesis => Parentheses (both variations were used in the codebase)
- fix typo and wording ("never alwayw") in gui/test/data/benchmark/simple.cpp's CheckOther::unsignedPositive():
```
- "An unsigned variable will never alwayw be positive so it is either pointless or "
+ "An unsigned variable can't be negative so it is either pointless or "
```
Use tok->isOp() instead of doing it "manually".
This fixes false positives for the C++ template case since
the brackets in "template < something >" are not comparison operators.
Profiling showed using tok->isOp() is actually faster than before (-O2).
1)The non-const version of Token::tokAt and Token::linkAt now will use a combination of const_cast and static_cast to remove duplicate code.
2)The non-const versions of Token::find(simple)match are added, in order to avoid the usage of const_cast outside the Token code.
3)As a consequence of 2), an useless const_cast applied to a call of Token::findsimplematch in Tokenizer code is removed.
Token::Match returns true if the token is a comparison operator. It's equivalent to matching '<|<=|==|!=|>=|>' and the tests show that the new pattern speeds up the program execution. Added the new pattern in CheckInternal (and also reordered the %cmd% lists) and in TestToken.
If it's true, add a new token before this. Except when this is the first one in the list, because there's not a way to update the list.front (I noticed some Token functions can potentially delete the list.front(), I don't know which effects does it cause).
- Don't simplify wchar_t to unsigned short on Win32 platforms to ensure that checking of strings and nullpointer dereferences works.
- Added it as standard type
- Shortened stringify output - 4 or more subsequent empty lines are abbreviated
- Fixed detection of comparision operators
- Make use of _type: Replaced three asserts by one
- Replace _isNumber, _isName, _isBoolean attributes by a single _type attribute (enum Token::Type), because not two of the old booleans could be true at the same time.
-> Add support for lots of different other kinds of tokens. (More precise checking of token type possible)
-> Replaced instant checking of type for Operators, etc. by a value calculated at creation time. (Faster checking)
- Function that stringifies one token: Token::stringify()
- Functions that stringify a list of tokens: Token::stringifyList()
-- Single and powerful "base" function, used by several "light" functions
Refactorized
- testtokenize.cpp and testsimplifytokens.cpp: Use improved stringification functions instead of several local implementations
- Avoided redundand creation of std::string when using TestTokenizer::tokenizeAndStringify and in cmdlineparser.cpp
- Added support for < and > in nextArgument()
- Made isStandardType inline by writing its implementation to token.h (like other bool is*() functions)
- Made firstWordEquals return bool instead of integer (only 0 and 1 were returned)
- Use std::string::empty() instead of operator==(""); Use character literals instead of string literals for single characters printed.
- Previously MathLib errors did not provide a filename, but after this change at least source file name should be printed
and if token is provided, also line number should be printed.
- Change also Token to use InternalError
- Modify Cppcheck-class to catch InternalError instead of Token
- Run dmake to update Makefile
- Bugfix: tok->stringify(tok) returns now "".
- Removed most of the inline-asm handling in preprocessor; improved the remaining handling of #asm/#endasm: Simplified to "asm(...);" statement
- Bugfix: tok->stringify(tok) returns now "".
- Removed most of the inline-asm handling in preprocessor; improved the remaining handling of #asm/#endasm: Simplified to "asm(...);" statement
rebnridgway
Daniel Marjamäki
Paul Fultz II
IOBYTE
rikardfalkeborn
amai2012
orbitcowboy
Markus Elfring
PKEuS
jrp2014
Simon Martin
orbitcowboy
Florin Iucha
Heiko Eißfeldt
Ayaz Salikhov
Ayaz Salikhov
PKEuS
Matthias Krüger
Frank Zingsheim
Boris Egorov
Lauri Nurmi
Alexander Mai
Philipp Kloke
Dmitry-Me
PKEuS
Simon Martin
Thomas Jarosch
Andrew C. Martin
Erik Lax
Edoardo Prezioso
Reijo Tomperi
acm4me
Ankita-gupta
Daniel Marjamäki
Zachary Blair