// Something to this effect
if(memAllocated == NULL)
return MEMORY_ALLOCATION_FAILURE;
Author
To see if something allocated memory using new would this work:
New is suppossed to throw an exception, but it doesn''t in MSVC (it does in Borland). MSVC just returns NULL if it fails. I don''t know about GCC, I haven''t tried.
[Resist Windows XP''s Invasive Production Activation Technology!]
[Resist Windows XP''s Invasive Production Activation Technology!]
GCC also throws an exception, but doesn''t return NULL.
[Resist Windows XP''s Invasive Production Activation Technology!]
[Resist Windows XP''s Invasive Production Activation Technology!]
I wrote a header like this to do new for me, that way it will be accurate under MSVC as well as other compilers (also, it is really nice for finding memory leaks 
). It doesn''t work under older Borland''s, unfortunetly. You can tell from the #if that I''m expecting newer MSVC''s to actually throw exceptions...
[Resist Windows XP''s Invasive Production Activation Technology!]
). It doesn''t work under older Borland''s, unfortunetly. You can tell from the #if that I''m expecting newer MSVC''s to actually throw exceptions... #ifndef INC_RUNE_0016xx_MASTER_COMMON_MEM_HEADER#define INC_RUNE_0016xx_MASTER_COMMON_MEM_HEADER#if (!defined(_MSC_VER) || _MSC_VER > 0x000004B0) #define NEW_USES_EXCEPTIONS #include <exception>#endiftemplate <class T> inline T *rNew(void) { #ifdef NEW_USES_EXCEPTIONS T *Get; try { Get = new T; } catch(exception) { return NULL; } return Get; #else return new T; #endif}template <class T> inline T *rNew(unsigned int Count) { #ifdef NEW_USES_EXCEPTIONS T *Get; try { Get = new T[Count]; } catch(exception) { return NULL; } return Get; #else return new T[Count]; #endif}template <class T> inline void rDelete(T *What) { delete What;}template <class T> inline void rDeleteArr(T *What) { delete [] What;}#ifdef NEW_USES_EXCEPTIONS #undef NEW_USES_EXCEPTIONS#endif#endif [Resist Windows XP''s Invasive Production Activation Technology!]
Ok, that''s the most my code has ever been mangled by this board, a new record!
[Resist Windows XP''s Invasive Production Activation Technology!]
[Resist Windows XP''s Invasive Production Activation Technology!]
quote:Original post by Null and Void
Ok, that''s the most my code has ever been mangled by this board, a new record!
Well, edit it then.
quote:Original post by Darkor
Well, edit it then.
Would that help? No, I don''t think so. The code I pasted looked fine, it is their script with warps it. You paste some code here, and see what happens to it if it''s in the source tags.
[Resist Windows XP''s Invasive Production Activation Technology!]
quote:Original post by Null and Void
Would that help? No, I don''t think so. The code I pasted looked fine, it is their script with warps it. You paste some code here, and see what happens to it if it''s in the source tags.
You have to insert some extra blank lines, or lines with only whitspace in them, for the source-boxes to look right. In particular most preprocessor directives usually make their terminating newline disappear, but having an extra blank line after every #include, #ifdef, etc. does the trick.
In reference to the original question: There is a version of new that is guaranteed to *not* throw any exception if it fails, and simply return NULL instead. To use it pass std::nothrow as a parameter to new.
e.g.
e.g.
using std::nothrow;void foobar(){ int *ptr = new (nothrow) int;}; This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
