🚀 go-pugleaf

In article <HARALD.95Nov1105205@moon.bion.kth.se>,
Harald Winroth <harald@bion.kth.se> wrote:
>I have a question regarding the proposed string class. In the April WP 
[...text deleted]
>would append "123" to str. Is that interpretation correct? If so, I think

It is that way in Plauger's "The Draft Standard C++ Library", and I'm
pretty sure it's one of the things that didn't change in the April
draft.

Speaking as someone that's written an implementation of (and uses
pretty heavily) this string class interface, it hasn't been an issue
at all. I think it's because that all of the functions that take char*
work in a similar fashion, and it's simple enough (and perhaps a bit
more obvious code-wise) to add the offset to the char* pointer if
that's what is desired. (Admittedly it seemed a little disturbing when
I was initially implementing it, but it just hasn't been a problem in
actual use.)

But I would like to argue for removing the position offset from
append/assign/insert etc. The same thing can be accomplished with
using substr(), and for a reasonably efficient implementation there
would only be a little extra overhead--not to mention that it makes
the resulting code more readable. The reasoning for the current design
may have been that since the offset case is efficient for char*, it
should be equally efficient for instances of string as well.
						Bob
---
[ comp.std.c++ is moderated.  Submission address: std-c++@ncar.ucar.edu.
  Contact address: std-c++-request@ncar.ucar.edu.  The moderation policy
  is summarized in http://dogbert.lbl.gov/~matt/std-c++/policy.html. ]