Author: chug
Date: Fri Jun 6 20:53:57 2014
New Revision: 1601010
URL: http://svn.apache.org/r1601010
Log:
PROTON-596: Suppress MSVC warnings
This suppression covers:
>..\..\proton-c\src\object\object.c(232):
warning C4244: 'argument' : conversion from 'ssize_t' to 'int',
possible loss of data
2>..\..\proton-c\src\object\object.c(260):
warning C4267: 'argument' : conversion from 'size_t' to 'int',
possible loss of data
2>..\..\proton-c\src\windows\driver.c(777):
warning C4800: 'int' : forcing value to bool 'true' or 'false'
(performance warning)
6>..\..\..\..\..\tests\tools\apps\c\msgr-recv.c(90):
warning C4996: 'sscanf': This function or variable may be unsafe.
Consider using sscanf_s instead.
To disable deprecation, use _CRT_SECURE_NO_WARNINGS
This covers the majority of the warnings. A typical 64-bit compile has 172
warnings making real problems harder to see. It also follows the
pattern used in Qpid proper.
* C4244 and C4267 could still be real problems.
Try enabling -Wconversion in GCC if you want to be pedantic in Qpid or
Proton.
* C4800 is probably not a risk as it's a fairly idiomatic pattern in C.
* C4996 is MSVC complaining of functions that consume buffers using
{pointer} and not {pointer, size}. They have been the source of seg
faults and innumerable security vulnerabilities throughout the years.
Modified:
qpid/proton/trunk/proton-c/CMakeLists.txt
Modified: qpid/proton/trunk/proton-c/CMakeLists.txt
URL:
http://svn.apache.org/viewvc/qpid/proton/trunk/proton-c/CMakeLists.txt?rev=1601010&r1=1601009&r2=1601010&view=diff
==============================================================================
--- qpid/proton/trunk/proton-c/CMakeLists.txt (original)
+++ qpid/proton/trunk/proton-c/CMakeLists.txt Fri Jun 6 20:53:57 2014
@@ -224,7 +224,13 @@ if (CMAKE_COMPILER_IS_GNUCC)
endif (CMAKE_COMPILER_IS_GNUCC)
if (MSVC)
- set(CMAKE_DEBUG_POSTFIX "d")
+ set(CMAKE_DEBUG_POSTFIX "d")
+ add_definitions(
+ /wd4244
+ /wd4267
+ /wd4800
+ /wd4996
+ )
endif (MSVC)
macro (pn_absolute_install_dir NAME VALUE PREFIX)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
