Re: [PATCH] Tighten length error check in string::replace

[Paolo Carlini <pcarlini at unitus dot it>]

Phil Edwards wrote:

>When I cleaned up and doc'd std_bitset.h, I added some explanatory text
>to the messages in the exceptions.  Previously they were all "bitset"
>which IMHO isn't very helpful to the end user.
>
>So for example, if we know that the code contains
>
>    __throw_length_error("basic_string::replace -- foo, bar, baz, blah");
>
>then the testcase might be able to catch an exception and verify that
>
>    ex.what() == "basic_string::replace -- foo, bar, baz, blah"
>
>This would distinguish it from all the other length_error instances.
>

Interesting. Thanks! I would like to work out something similar for 
basic_string in the near future...

As regards the testcase at hand, however, it is slightly different, 
since the various possible length_error exceptions already happen in 
clearly distinguishable methods, that is basic_string::replace, 
basic_string::_M_replace, basic_string::_S_create.

Ciao, Paolo.