db8c7deb30
Fixed #3232 (Check if container is modified inside BOOST_FOREACH)
2011-10-23 13:07:43 +02:00
6047a28902
Fixed #3185 (Add three more exit functions)
2011-10-23 12:23:53 +02:00
54b3d72ee3
Added Token::nextArgument()
2011-10-23 11:23:48 +02:00
b7ab1e7d7e
Fixed #3237 (Bug in parser of class operator functions)
2011-10-23 10:25:14 +02:00
bf2759118b
Added support for C++11's cend/crend/cbegin/crbegin functions in some STL checks
...
Patch from: PKEuS <philipp.kloke@web.de>
2011-10-22 23:38:03 +03:00
3568b5a841
Improved memset with 0 bytes check. TODO done (unit test).
...
Patch from: PKEuS <philipp.kloke@web.de>
2011-10-22 23:34:10 +03:00
91c6608175
Improve fix of #934 (be careful of macros code).
2011-10-22 20:43:42 +02:00
20179673ce
Fixed #934 (new check: missuse of std::cout)
2011-10-22 17:12:52 +02:00
12f6ce46f8
Add more warnings and fix the ones reported by them.
2011-10-22 15:05:43 +02:00
3d8fa2f76e
Fixed #3123 ([False Positive] Shadowed enum)
2011-10-22 13:43:53 +02:00
32e20d4e31
made 'variableHidingTypedef' inconclusive. See ticket #3199
2011-10-22 13:26:33 +02:00
2b54f00c87
Fixed #3238 (Crash in obsolete functions check.)
2011-10-22 12:36:45 +02:00
ccf087d2ea
checkMathFunctions: Refactoring the check
2011-10-22 12:31:47 +02:00
55d9f0873a
Fix #3243 (Improve non reentrant function check)
...
- Use std::map instead of linear std::list walk
and run fast tests like tok->isName() first.
Global speed up is 4.8% (profiled with google-perftools)
- Catch function invocations in global namespace
and ignore other namespaces except "std".
std::localtime() and others are also non-thread safe on POSIX.
Note: The check matches f.e. also on "std::getrpcbyname()",
but that would result in a compile error anyway.
No need to have an extra "std::xxxxx" whitelist.
- Remove double listed "rand" and "getrpcbyname" function names
2011-10-22 11:54:52 +02:00
a9d2d45fbc
astyle formatting
2011-10-22 11:11:54 +02:00
8157799e10
Fixed #3204 (Refactor standards support in Settings)
2011-10-22 10:34:04 +02:00
cde95d09a3
Merge pull request #41 from gansb/master
...
std::istream used but not declared.
2011-10-22 00:54:04 -07:00
ab827e3631
Merge pull request #40 from joshbeck/master
...
Added option to set #ifdef configuration check limit
2011-10-22 00:51:05 -07:00
b332ea8222
Fixed #3204 (Refactor standards support in Settings)
2011-10-22 09:45:48 +02:00
3f25bd9530
Qualify fill_n properly.
...
Relying on ADL isn't a good idea as it's not always well implemented.
2011-10-20 09:02:20 +02:00
463b105d98
Add condition for HP-UX.
2011-10-20 09:01:58 +02:00
7d6b76e1d1
Include <iosfwd> for std::istream.
2011-10-20 08:46:29 +02:00
41d9daa03d
Fixed ticket #3133 (Improve Check: Found obsolete function)
...
- add check for std::gets
- improve check when multiple obsolete functions are used
- remove false positive (declared functions)
2011-10-19 20:21:50 +02:00
9257e82475
Improve tokenizer: don't confuse between a label and the case/default statements.
2011-10-19 18:49:02 +02:00
882b0c15a0
Improve tokenizing of "do{ ... } while(0)".
...
The simplification will be the same for every combination, except for the 'continue' and the 'break'. In that case, keep the block braces.
2011-10-19 18:20:29 +02:00
c9ee19c654
Complete 'disable the rules by default' for other platforms.
2011-10-19 15:14:39 +02:00
4f256ef314
Runastyle + removal of useless conditions after previous commit.
2011-10-19 14:48:08 +02:00
73f0e2c5b9
Fixed ticket #3227 (Improve tokenizer: labels are not handled correctly after a case or another label instruction)
...
Moved the labels simplification after combining the tokens and moved the case/default simplification inside the labels.
2011-10-19 14:20:09 +02:00
56561835f8
Add option to set #ifdef configuration check limit
2011-10-18 19:52:55 -05:00
f4cdf57955
Fixed ticket #3229 (Segfault with goto):
...
Don't handle labels, gotos, returns and indentations inside the round braces. This is applied not only for simplifyGoto, but also to simplifyFlowControl.
Rewrite enum badly handled by astyle inside mathlib.cpp and move simplifyFlowControl at the end of the simplifyTokenList in order to catch more dead code, not related to the ticket.
2011-10-18 23:56:35 +02:00
dc15641954
Fixed #3223 (Improve check: Check more STL algorithms for missmatching containers check)
2011-10-18 21:55:41 +02:00
ea02bd905a
Fixed #3225 (Boolean comparison with string literals)
2011-10-18 21:37:03 +02:00
3d18fdfa3f
Fixed #3220 (False positive: possible null pointer dereference: 'SwDoc *pDoc = NULL; pDoc->do_something();')
2011-10-18 19:34:14 +02:00
03ff25f152
gcc: fixed compiler warnings/errors
2011-10-17 20:32:59 +02:00
bf0386c938
Preprocessor: Started working on new preprocessing that uses -D information better. Ticket: #2131
2011-10-17 20:12:46 +02:00
de8eb48a5c
Remove redundant function findmatch inside tokenize.cpp.
...
There's Token::findmatch which is already used inside other Tokenizer functions.
2011-10-17 13:46:27 +02:00
c3caade3ca
- Fixed mispelled function name;
...
changed variable name inside simplifyFlowControl for consistency;
improved simplifyFlowControl to handle better this kind of code:
"return; { { } { label : ; ok ( ) ; } }"->"return ; { { label: ok ( ) ; } }".
2011-10-17 02:16:49 +02:00
4705026242
fixed a few doxygen warnings
2011-10-16 17:16:07 +02:00
1a710e5761
fixed regression on linux due to missing include
2011-10-16 14:16:35 +02:00
f8074b71d2
Fixed #2470 (False negative: Possible null pointer dereference (C++0x keyword 'nullptr'))
2011-10-16 12:54:58 +02:00
32496fb009
Token: Updated comments
2011-10-16 08:25:11 +02:00
a52b73f9f9
Fix #3208 (Simplify pointer to standard type, C only)
...
The symbol database is unavailable during token simplification
and &data[0] might return something completely different for C++.
Moved code_is_c() from checkOther to Tokenizer.
2011-10-16 08:09:57 +02:00
1ec32e27db
Borland: Fixed compiler errors
2011-10-16 07:52:54 +02:00
d5883ef0c4
astyle formatting
2011-10-16 07:50:34 +02:00
71a1d98693
Fixed ##3211 (Crash in gitHEAD when arglist count is smaller than format string)
2011-10-16 07:06:18 +02:00
8afc1b6f2d
Changed the name of the removal of dead code after flow control statements for consistency (simplifyDeadCode is too generalized as name)
2011-10-15 12:45:48 +02:00
57ead6988e
Fixed ticket #3138 (Tokenizer: remove unreachable code below goto inside a namespace|class|struct block)
2011-10-15 11:35:45 +02:00
f95b692a69
Fixed ticket #3113 (complete simplification of dead code after control flow statements)
2011-10-15 01:34:07 +02:00
7824e5c0f5
Fixed #3210 (STL check: Add support for reverse iterator)
2011-10-14 19:54:20 +02:00
7ae39f13cc
Fixed #3198 (Add check for readlink())
2011-10-14 19:45:51 +02:00
4fb7865f49
fix typos using codespell
...
Codespell is available at http://git.profusion.mobi/cgit.cgi/lucas/codespell/ .
2011-10-13 22:04:41 +02:00
6f8e42a5af
changed the astyle formatting flags
2011-10-13 20:53:06 +02:00
b792efb761
Fix ticket 3140 and 3137 (Improve Tokenizer: Remove dead for loop - simplify while(0) better.
...
Note: it doesn't remove dead code if there's a label inside. Will be improved in another ticket.
2011-10-13 13:01:24 +02:00
16506f0a8c
Remove redundant simplifyDeadCode call inside the Tokenizer::tokenize and move the relative tests in the file 'testsimplifytokens.cpp'.
2011-10-13 12:26:27 +02:00
82366918ff
Fixed #2627 (switch case fall through: redundant strcpy)
2011-10-13 01:27:22 -07:00
91d7621994
Fixed #3180 (New check: Detect nullpointers given to printf via variable argument list)
2011-10-12 22:11:27 +02:00
74c0468a18
astyle formatting
2011-10-12 22:06:19 +02:00
b73896bcc5
Fixed #3163 (Out of bounds pointer arithmetic not reset)
2011-10-12 20:54:39 +02:00
371871b0b7
Tokenizer::removeMacrosInGlobalScope: Don't remove C/C++ keywords namespace/struct/etc. Ticket: #3193
2011-10-12 19:23:39 +02:00
cf39ece8da
Fixed #3150 (Invalid warning when pointer difference is not adjacent)
2011-10-12 19:01:44 +02:00
ee04c1bfb7
fix #3194 (Parsing glitch causes false 'can be const' error for an 'if' statement)
2011-10-12 09:10:34 -04:00
af6a6f9c2d
run astyle
2011-10-12 09:08:54 -04:00
07d8cd0a12
improved realloc checking '*p = realloc(*p,..'. Ticket: #1791
2011-10-11 22:07:14 +02:00
849bee8437
[PATCH] Detect suspicious use of semicolon after 'if/for/while'
...
statements if they are followed by a {..} block.
Examples are:
for (int i = 0; i < 10; ++i);
{
printf("i)";
}
or
if (i == 100);
{
die("Wrong argument");
}
This new check is active if you enable inconclusive checks.
2011-10-11 08:41:39 +02:00
fcf360825a
Fixed #2955 (New check: Using && instead of &)
2011-10-10 10:11:17 -07:00
742858b67c
Fixed some simple errors in some old commits.
2011-10-10 17:18:47 +02:00
af23a9dcbb
Fixed ticket #3140 (Improve Tokenizer: Remove dead for loop 'for (int i = 0; i < 0; ++i)')
2011-10-10 15:15:45 +02:00
5f9b916dcc
Fixed #2115 (new check: endless loop (unsigned comparison that is always true))
2011-10-09 20:35:46 +02:00
96beb94b21
Fixed #3183 (False positive: (error) syntax error)
2011-10-09 18:42:36 +02:00
175503ef94
Null pointer: improved inconclusive checking in CheckNullPointer::nullPointerAfterLoop
2011-10-09 15:09:02 +02:00
1d9a99012b
Fixed 3133 (Improve Check: Found obsolete function)
...
- add check for std::gets
- improve check when multiple obsolete functions are used
- remove false positive (declared functions)
2011-10-09 10:31:21 +02:00
9270b84d64
More fixes to the label simplifier (related to ticket 3176):
...
1)Simplify the labels if there are combinations of the symbols '*','&','{' and '(' after a '%var% :' and before another '%var%';
2)but do not simplify the label if it's inside an unpreprocessed macro code.
2011-10-08 21:13:53 +02:00
0b09732881
Null pointers: fixed false negatives when there are '(p && ..'
2011-10-08 12:58:10 +02:00
bf25340fd3
Settings: add missing initialisation for member variable c99 (reported by Valgrind)
2011-10-08 10:29:11 +02:00
5c08979920
1.51: Updated versions
2011-10-08 07:45:39 +02:00
37d4a01dcb
Added todo comments to fix false negatives caused by my fix for #3128
2011-10-08 07:39:20 +02:00
e2ec5a127f
Fixed #3128 (False positive: null pointer dereference check does not handle complex boolean logic properly)
2011-10-07 21:08:21 +02:00
a1a839fadc
fix tokenizer and symbol database to not mistake goto statement for variable declaration
2011-10-07 08:11:48 -04:00
7dd001c9a9
run astyle
2011-10-06 21:10:20 -04:00
80df7ea839
Fixed ticket #3176 (cppcheck reports wrong unused variable)
...
Label simplification didn't take count of the '(' parenthesis or the '&' and '*' operators inside or outside the parenthesis after the colon.
2011-10-07 00:54:44 +02:00
fe4ce594ce
Renamed the function 'removeRedundantCodeAfterReturn' with 'simplifyDeadCode'.
...
This is a preparation in order to add new features to this function (see the @todo in the tokenize header file, ticket #3113 and #3175 for details).
2011-10-06 22:40:39 +02:00
73fb6725e6
Readded CheckOther::bitwiseOnBoolean as inconclusive check
2011-10-06 22:01:48 +02:00
307dd00efb
Fixed #3072 (improve check: deprecated function alloca)
2011-10-06 08:10:51 +02:00
09109f19f8
Removed CheckOther::bitwiseOnBoolean check. The reasons can be seen in my comments in ticket #3062 .
2011-10-05 20:46:07 +02:00
494d3af3d1
Fixed #1877 (Be more strict about int vs. bool, part II)
2011-10-05 20:30:36 +02:00
abd2525339
Fixed #3161 (Show buffers size info for snprintf() buffer overruns)
2011-10-05 20:17:57 +02:00
d741f64256
Merge branch 'master' of github.com:danmar/cppcheck
2011-10-05 19:52:49 +02:00
3cfe7ca1a7
Move "information" errors to "style" errors.
...
"information" severity is documented in lib/errorlogger.h as:
Checking information.
Information message about the checking (process) itself. These
messages inform about header files not found etc issues that are
not errors in the code but something user needs to know.
It IS NOT for errors in the code. All the current "information"-
severity errors fit nicely into description of the "style"-
severity.
We definitely need to separate processing information and actual
errors in the code. It is highly confusing for users to mix these
two different things. Hence all current "information" code error
messages are moved to "style" category.
Ticket: #3165 (Stop misusing the 'information' error severity!)
2011-10-05 20:44:00 +03:00
8e90ad106c
Fixed ticket #3148 (Analysis failed on "Modules/mathmodule.c")
...
Fix: Do not handle code when 'return' is inside a macro. This is valid also for 'switch' code inside a macro.
2011-10-05 19:33:46 +02:00
3c370dcfdf
Fixed #3146 "Analysis failed" error on softmagic.c:
...
-Fixed: the code didn't check the rightness of the switch syntax if it was inside another switch;
-Tweaked: removed the two bool variables, so now the conditions take count of indentation unsigned variables.
2011-10-04 12:16:39 +02:00
bc74213036
Fixed #3132 (Analysis failed. If the code is valid then please report this failure.)
2011-10-03 19:38:37 +02:00
417dc1ff2a
Fixed #3159 (Uninitialized variable false positive)
2011-10-02 20:38:58 +02:00
480c403511
Fixed #3125 (FP: Possible null pointer dereference in conditional operator)
2011-10-02 19:27:18 +02:00
b2d0e48a31
Fixed #3126 (false positive: Null pointer dereference)
2011-10-02 18:24:45 +02:00
65b0fb4519
fix #3153 (false positive buffer access out-of-bounds)
2011-09-30 17:28:59 -04:00
272783347b
another variation of false negative from #3149
2011-09-30 16:26:08 -04:00
2193572c9e
test _WIN64 before _WIN32 (64-bit compiler defines both macros)
2011-09-30 14:12:44 -04:00
1dcb8b2382
really fix #3149 (false negative: Technically the member function 'A::f' can be const.)
2011-09-30 11:16:34 -04:00
556d523e4f
fix #3149 (false negative: Technically the member function 'A::f' can be const.)
2011-09-29 20:16:52 -04:00
ac942c6684
fix #3151 (missing portability warning for extra qualifier on destructor)
2011-09-29 19:25:14 -04:00
Thomas Jarosch
PKEuS
Erik Lax
Reijo Tomperi
Edoardo Prezioso
Johan Samuelson
Daniel Marjamäki
Marek Zmysłowski
Benjamin Goose
seb777
Joshua Beck
Ettl Martin
Jonathan Neuschäfer
Zachary Blair
Robert Reif
Stefan Weil
Kimmo Varis