Commit Graph

918 Commits

Author SHA1 Message Date
Edoardo Prezioso b2f8161862 Rewrite fix of ticket #3415 (Segmentation fault in new check for uninitialized variables):
new simplify: 'code ; ({ some_code ; });' to 'code ; some_code ;';
remove the sanity check in checkunusedvar, so when it crashes again it's easy to spot a bigger problem;
'internal' reported 'Token::Match' with simple pattern.

'Tokenizer::tokenize()' refactorings:
make one loop with many simple simplifications instead of many loops with one simplification.
some style changes.
2011-12-28 16:55:59 +01:00
Reijo Tomperi 1afd923d34 Fix bug in C# detection code. The test had a bug also, so it was not detected by the test. 2011-12-27 22:50:56 +02:00
Jonathan Neuschäfer 7a5627029f tokenizer: proper checks for Java and C# 2011-12-27 18:01:39 +01:00
Daniel Marjamäki 91874214d8 Refactoring: Added utility function that determine if scope ends with a call to a noreturn function 2011-12-27 18:00:12 +01:00
Daniel Marjamäki 987392f254 Tokenizer: Better handling of array of function pointers 2011-12-27 11:56:40 +01:00
Daniel Marjamäki 9593e14d72 Removed debug variables 2011-12-27 09:27:46 +01:00
Edoardo Prezioso 293437bdfc Fixed ticket #3416 (segmentation fault of cppcheck) 2011-12-26 03:21:23 +01:00
Reijo Tomperi e112bfdd47 Refactoring: Rename some variables and funtions in tokenizer.
Small logic change also for count variable to make new name more logical.
2011-12-26 00:06:27 +02:00
Reijo Tomperi 347982a347 Refactoring: Extract several methods from tokenize() 2011-12-24 23:23:08 +02:00
Reijo Tomperi 8e6ebd4a22 Refactoring: Extract method: simplifyDoublePlusAndDoubleMinus() 2011-12-24 22:51:55 +02:00
Reijo Tomperi 6155932207 Refactoring: Rename some variables with too short names, such as "s". 2011-12-24 00:25:03 +02:00
Reijo Tomperi 00cbf02d84 Refactoring: Rename variable used -> templateInstantiations 2011-12-24 00:03:03 +02:00
Reijo Tomperi 70c4bb54b4 Refactoring: Split long function in tokenizer.
simplifyTemplatesInstantiate()
2011-12-22 23:13:45 +02:00
Reijo Tomperi 1c12d04d0d Refactoring: Make a 300 line function a little smaller.
simplifyTemplatesInstantiate()
2011-12-22 22:39:15 +02:00
Daniel Marjamäki 2bb5de4c89 Fixed #3425 (False positive: Null pointer dereference (pointer is checked in macro)) 2011-12-18 13:33:23 +01:00
Daniel Marjamäki 83cdf734fe Tokenizer: The tokenizer didn't handle '0x0E-7' correctly 2011-12-18 07:37:20 +01:00
Daniel Marjamäki d21ca7881c Fixed #3424 (Tokenizer: doesn't tokenize if-if-else-else correctly) 2011-12-17 19:05:14 +01:00
Daniel Marjamäki 6a4b1127aa astyle formatting 2011-12-14 17:07:06 +01:00
Edoardo Prezioso 40aa326065 Tokenizer: remove redundant 'done' condition, related to ticket #3409.
Add more possible letters to simplifyInitVar patterns.
Replace 'tokAt(1)' with 'next()' and 'linkAt(1)' with 'next()->link()'.
Little optimizing improvements to simplifyAsm.
2011-12-13 22:41:38 +01:00
PKEuS c9f5117cf5 Fixed #3407 (False positive: (inconclusive) Found duplicate branches for if and else. (inline assembler)) 2011-12-13 21:42:38 +01:00
Edoardo Prezioso 81a2e62abd Tokenizer::simplifyGoto: fix another memory read error reported by Valgrind.
Also improvements to Tokenizer are done.
2011-12-12 22:03:25 +01:00
Edoardo Prezioso 79b82f115f Tokenizer: some changes to line numbers of some tokens:
simplifyAsm: change line number of newly added 'asm ( )' in order to be the same as next ';'.
simplifyIfAddBraces: Change line number of newly added '}' in order to be the same as next 'else', except for '{ ; } else'.
2011-12-12 20:55:25 +01:00
Edoardo Prezioso 626f13ae7e Tokenizer: Fix another Valgrind error in simplifyErrNoInWhile. Add more jumping patterns to simplifyIfAddBraces and simplifyDoWhileAddBraces. 2011-12-12 13:15:35 +01:00
Edoardo Prezioso 99d8ce3732 Fix Valgrind error in arraySize():
if there's code like: 'int [ ] a = { 1 , 5 , }', 'end' is last '}', when 'tok2' arrives to second ',' and cppcheck finds out that next token to 'tok2' is '}', remove this ',' with 'tok2->deleteThis()' but it causes (maybe) a memory corruption to 'end' which will gave problems later because 'tok' will be assigned to it or its next token.
2011-12-12 02:23:49 +01:00
Edoardo Prezioso 52620e6493 Tokenizer::simplifyDoWhileAddBraces:
1)rewrite fix for ticket #988 (just don't simplify inside macro parenthesis);
2)use a different organization of the code: start from last token and proceed backwards. This way 'simplifyDoWhileAddBracesHelper' can be called just once, hence the 'Helper' code can be improved and moved in the main function.
2011-12-12 00:20:46 +01:00
Edoardo Prezioso cf574072b6 Tokenizer::arraySize(): improve code structure, fixed 'tok2' for-looping which didn't stop to the ending '}' parenthesis. 2011-12-11 23:13:37 +01:00
Edoardo Prezioso 3c098839d1 simplifyIfAddBraces: Remove restriction for jumping opening parenthesis '(' as a fix to #2873, because even without this the original test case doesn't crash anymore. Add more jumping patterns.
simplifyFunctionParameters: Add more jumping patterns and an observation related to error message for equal parameter names, help needed.
Fix grammar mistake in comment.
2011-12-11 13:07:13 +01:00
Edoardo Prezioso 27801b35eb Other improvements to tokenizer code.
Handle SQL code better, even when there's no ';' after 'EXEC SQL'.
simplifyVarDecl: Check with 'Token::Match' once when you have to check a pattern inside which there should be a number or a variable. Use '%any%' and after that check if that string is a number or a variable name later.
simplifyGoto: process also anonymous structs and unions which should have 'indentspecial' incremented, skip code like 'var = { ... }'.
2011-12-11 02:02:38 +01:00
Edoardo Prezioso 5134964026 Fix code in commit 5f522fb841 and reapply it again. 2011-12-10 23:49:56 +01:00
Daniel Marjamäki 04159b81b8 reverted 5f522fb841, to avoid hang in TestBufferOverrun::arrayInfo 2011-12-10 20:51:36 +01:00
Edoardo Prezioso 5f522fb841 Tokenizer: improve code and remove redundant checks.
In particular: improve 'double sharp' concatenation by handling also code which begins without the [{};]. Do the same with C# code. Also, in the labels simplification function, skip also '{}' when the open bracket is after a '='.
2011-12-10 18:45:27 +01:00
Edoardo Prezioso bf815ac1e4 Improve labels simplification code, remove redundant checking. 2011-12-10 14:13:48 +01:00
Daniel Marjamäki c2e86b867c Fixed #3386 (Syntax Error: 'int a[]={b<c?1:2,3};') 2011-12-10 11:34:27 +01:00
Edoardo Prezioso 7d12951da0 1)Fixed ticket #3184 (Improve Tokenizer: improve simplifyMulAnd to simplify weirder code);
2)Fix a test case inside TestSimplifyTokens::flowControl.
2011-12-09 20:47:51 +01:00
Daniel Marjamäki 8ca8887849 C++ Builder : Fixed compiler error 2011-12-08 22:24:09 +01:00
PKEuS 167a7e3e51 Various code cleanups 2011-12-08 21:28:34 +01:00
PKEuS 1bef8d1247 Tokenizer: Code cleanups 2011-12-08 17:42:26 +01:00
Edoardo Prezioso 149ff355e2 Add more kind of code to process for 'removeExceptionSpecifications' (unions, anonymous unions and structs). 2011-12-08 02:50:05 +01:00
Edoardo Prezioso 4de5d87078 1)Remove newly copy of 'eraseTokens' and replace every occurrence with 'deleteNext' with right parameter.
2)General improvements to code.
2011-12-08 01:44:18 +01:00
PKEuS 1c1c31bea0 Tokenizer: Refactorings and improvements. Better handling of [] 2011-12-07 21:15:00 +01:00
Edoardo Prezioso 612b3c59ca 1)Implement another declaration of 'Token::eraseTokens' with parameters: the starting token and the number of times needed to call 'deleteNext'.
2)Use the overloaded 'eraseTokens' where needed instead of using consecutively 'deleteNext'.
3)Tokenizer: general optimizations about the deletion of some tokens.
4)Run astyle.
2011-12-07 02:29:03 +01:00
Edoardo Prezioso b9224439c1 Add 'exit' and 'abort' to the flow control list in order to simplify code after them. 2011-12-06 22:56:12 +01:00
Edoardo Prezioso eb5fe250ab Partial fix ticket #3385 ('throw' isn't removed if the argument of the function is type 'struct|class'). 2011-12-06 21:02:26 +01:00
Edoardo Prezioso 0504952950 Remove some redundant parenthesis - part 9 2011-12-06 16:53:57 +01:00
Edoardo Prezioso 1b4b8fbefe Tokenizer: use more 'deleteNext' instead of 'eraseTokens' where possible. 2011-12-05 18:52:34 +01:00
Edoardo Prezioso 186ab3645e Oops. I should eventually keep ';', not 'typedef'. 2011-12-05 00:26:08 +01:00
Edoardo Prezioso 65213772df Tokenizer: use 'deleteNext()' or, even better, 'eraseTokens' instead of 'deleteThis()' where possible. 2011-12-05 00:08:50 +01:00
Edoardo Prezioso 9cadd7c17a Remove some redundant semicolons - part 8 2011-12-04 19:11:51 +01:00
Edoardo Prezioso 4946b036ad Remove some redundant semicolons - part 7 2011-12-04 18:08:35 +01:00
Edoardo Prezioso 52a174fa9d Remove some redundant semicolons - part 6 2011-12-04 17:30:25 +01:00