Rob Stewart wrote:
From: "Reece Dunn" <msclrhd@hotmail.com>
Martin wrote:
2. Substrings ------------- One of the applications I can see for fixed_string is when you want to avoid dynamic allocation but still want to use the std::string interface and algorithms. In that perspective it seems strange that substr returns a std::string instead of a fixed_string.
Agreed.
I agree too. It is just that the current implementation cannot create an instance of boost::char_string (because of the ABC - necessary to allow operations on variable-capacity strings). Also, you cannot get at the fixed_string type/length at compile time so it is not possible to derive the *return type* at compile time.
This is related to the above problem. You cannot create an instance of boost::char_string, etc. and as such cannot implement substr in the usual way (this is the proplem I was having with operator+). The solution would be to defione an alternate substr function in fixed_string, but this would not solve the problem for boost::char_string, etc.
The alternative would be to use a ranged string (pointer to the fixed_string buffer and length), but the problem with this is null-termination.
I do not see an easy way around this without removing the ABC design, since it is not possible to do fixed_string< length() > or something similar.
Couldn't substr() be a pure virtual function in char_string that deferred to the derived class to create a copy of itself? The derived class can just create a duplicate of itself, including the current capacity. That capacity is guarranteed to be sufficient to hold any substring requested.
The return type cannot be boost::fixed_string_base because of the whole ABC issue. Therefore, the return type needs to be a reference or a pointer. It is possible to do this, and have: class fixed_string { inline fixed_string_base & substr_( size_type pos, size_type n ) { return( this_type( *this, pos, n )); } }; This is problematic because of returning the address of a temporary - and leads to unpredictable output. I am looking for a better solution to this. I am also thinking about how to remove the ABC from the implementation using a technique pointed out by someone on this thread (can't remember who): class fs_base { size_t len; szie_t capacity; // needed for buffer-safe operations CharT str[ 1 ]; // string manipulation functions fs_base( size_t c ): capacity( c ){} }; template< size_t n > class fixed_string: public fs_base { CharT data[ n ]; fixed_string(): fs_base( n ){} }; I'll need to look into this, to see what effect this will have, to see if it is a possible replacement for the ABC. What do other people think? Regards, Reece _________________________________________________________________ Use MSN Messenger to send music and pics to your friends http://www.msn.co.uk/messenger