This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: [v3] basic_string::reserve() shrink-to-fit?
- From: Neil Ferguson <nferguso at eso dot org>
- To: Paolo Carlini <pcarlini at unitus dot it>
- Cc: LibStdC++ General List <libstdc++ at gcc dot gnu dot org>, Nathan Myers <ncm at cantrip dot org>
- Date: Thu, 11 Dec 2003 16:25:03 +0100
- Subject: Re: [v3] basic_string::reserve() shrink-to-fit?
- References: <3FD88478.3070309@eso.org> <3FD887C9.1020906@unitus.it>
Paolo Carlini wrote:
Neil Ferguson wrote:
so that __res only needs to be different from capacity(), rather than
greater,
First blush (no testsuite, etc...), seems a good idea to me, Nathan?
Paolo.
Sorry, I forgot to mention that I'd tried the supplied DejaGNU test
suite on it, for i686-pc-linux-gnu. I got some XPASS lines for the
22_locale tests, but nothing for strings, and the following summary:
--
=== libstdc++-v3 Summary ===
# of expected passes 453
# of unexpected successes 26
# of expected failures 2
--
I haven't written a test case that specifically exercises my change,
though, as I haven't had time to learn how yet. This is the first time
I've meddled with libstdc++ :-)
Neil.