8c32cfe853
Python 3 directly decodes the text when it is read(). If there is any invalid UTF-8 character in the text an exception is thrown (IIRC it is UnicodeDecodeError). Opening the file with `error='ignore'` avoids throwing an exception and just ignores the invalid character. Since this is only possible since Python version 3 there must be extra code for older versions. The test script has been enhanced. It now also uses a package which contains a file with at least one invalid UTF-8 character. |
||
---|---|---|
.. | ||
test | ||
triage | ||
ci.py | ||
clang-ast.cpp | ||
compare-ast-clang-and-cppcheck.py | ||
compare.cs | ||
daca-test-patch.sh | ||
daca2-download.py | ||
daca2-getpackages.py | ||
daca2-logs2git.sh | ||
daca2-report.py | ||
dmake.cpp | ||
dmake.sln | ||
dmake.vcxproj | ||
donate-cpu-server.py | ||
donate-cpu.py | ||
extract_and_run_more_tests.sh | ||
extracttests.py | ||
generate_and_run_more_tests.sh | ||
generate_cfg_tests.cpp | ||
git-pre-commit-cppcheck | ||
listErrorsWithoutCWE.py | ||
matchcompiler.py | ||
parse-glibc.py | ||
pr.py | ||
readme.md | ||
reduce.cpp | ||
reduce.py | ||
run-coverity.sh | ||
run_more_tests.sh | ||
test_matchcompiler.py | ||
test_showtimetop5.sh | ||
testrunnerify_code.sh | ||
times-tags.sh | ||
times-vs.py | ||
times.c | ||
times.sh | ||
trac-keywords.py |
readme.md
Cppcheck developer and build tools
* tools/matchcompiler.py
The matchcompiler.py is a build script that performs a few code transformations to .cpp files under the lib directory. These transformations are related to the use of Token::Match()
function and are intended to improve code performance. The transformed files are saved on the build directory. This tool is silently used when building the code with SRCDIR=build
, that is:
$ cd path/to/cppcheck
$ make SRCDIR=build
Here is a simple example of the matchcompiler.py optimization. Suppose there is a file example.cpp under lib/:
// lib/example.cpp
void f1() {
Token::Match(tok, "abc");
}
void f2() {
const char *abc = "abc";
Token::Match(tok, abc);
}
If you manually run matchcompiler.py from the main directory:
$ cd path/to/cppcheck
$ python tools/matchcompiler.py
A file example.cpp will be generated on the build directory:
// build/example.cpp
#include "token.h"
#include "errorlogger.h"
#include <string>
#include <cstring>
static const std::string matchStr1("abc");
// pattern: abc
static bool match1(const Token* tok) {
if (!tok || !(tok->str()==matchStr1)/* abc */)
return false;
return true;
}
void f1() {
match1(tok);
}
void f2() {
const char *abc = "abc";
Token::Match(tok, abc);
}
From this we can see that the usage of Token::Match()
in f1()
has been optimized, whereas the one in f2()
couldn't be optimized (the string wasn't inline on the Token::Match()
call). The developer doesn't need to use this tool during development but should be aware of these optimizations. Building with this optimization, cppcheck can get a boost of 2x of speed-up.
* tools/dmake.cpp
Automatically generates the main Makefile
for Cppcheck (the main Makefile
should not be modified manually). To build and run the dmake
tool execute:
$ cd path/to/cppcheck
$ make dmake
$ ./dmake
* tools/reduce.cpp
Cppcheck tool that reduces code for a hang/false positive. To build the tool run:
$ cd path/to/cppcheck
$ make reduce
* tools/times.sh
Script to generate a times.log
file that contains timing information of the last 20 revisions.