Re: [boost] Re: bounds checking in array.hpp

From: Jeremy Siek (jsiek_at_[hidden])
Date: 2000-12-15 11:28:44

Why not assert()? I don't know... its so... old school ;) And
std::range_error is just itching to get used.

Just for fun, here's an alternate definition that's more useful when
debuging (at least in gdb). Stops execution (via a seg fault) and lets you
see the current environment, stack, etc.

#define BOOST_ASSERT_RANGE(expr) \
  do if (!expr) { \
    int* x = 0; \
    *x = 0; \
  } while (0)

On Fri, 15 Dec 2000, DEAN STURTEVANT wrote:

dean> --- In boost_at_[hidden], Jeremy Siek <jsiek_at_l...> wrote:
dean> > Should we have a boost macro for adding in this kind of assertion?
dean> > 
dean> > #define BOOST_ASSERT_RANGE(expr) \
dean> >   if (expr) \
dean> >     throw std::range_error( \
dean> >       __FILE__ + std::string(":") \
dean> >       + boost::lexical_cast<std::string>(__LINE__) + \
dean> >       std::string(": range error") )
dean> > 
dean> 
dean> Well, ignoring the fact that the test should be on !(expr) rather 
dean> than expr and that the construction is syntactically incorrect (You 
dean> can't go "string" + std::string("str")), I was thinking of something 
dean> more lightweight. What is wrong with assert()?
dean> 
dean> 
dean> 
dean> 
dean> 
dean> 

----------------------------------------------------------------------
 Jeremy Siek                        www: http://www.lsc.nd.edu/~jsiek/
 Ph.D. Candidate                  email: jsiek_at_[hidden]
 Univ. of Notre Dame         work phone: [phone]
----------------------------------------------------------------------