This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [patch] debug mode

From: "Bo Persson" <bop at gmb dot dk>
To: <libstdc++ at gcc dot gnu dot org>
Date: Fri, 20 Aug 2010 17:08:33 +0200
Subject: Re: [patch] debug mode
References: <4C4F3C54.8020106@free.fr> <4C50900B.6030705@oracle.com> <4C53E2FD.5070206@free.fr> <4C545CB5.4090902@oracle.com> <4C545E6D.5050409@oracle.com> <4C6BD03E.1050706@free.fr> <4C6BDBFB.6000307@oracle.com> <4C6D8985.3010604@free.fr>

François Dumont wrote:

On 08/18/2010 03:11 PM, Paolo Carlini wrote:

About __first and __last, I would suggest
not changing those to __f and __l if at all possible: short names
are always more dangerous vs BADNAMES on some less widespread
machines.

Hi

    Just regarding this point, I grep the lib headers, in bits
sub-folder there are 730 occurrences of __f and about 940 of __l.
So I don't think my usage of it in debug mode will introduce any
regression. Moreover I consider using this technique to limit line
width when I saw some code in debug headers that was already using
__f, __l. Do you still want me to revert this point ?

I guess the limited line length is supposed to improve readability. Obfustacing the names to be able to follow the letter of the law, doesn't seem like an improvement.

Just my 2 cents.

Bo Persson

Follow-Ups:
- Re: [patch] debug mode
  - From: Paolo Carlini

References:
- Re: [patch] debug mode
  - From: François Dumont
- Re: [patch] debug mode
  - From: Paolo Carlini
- Re: [patch] debug mode
  - From: François Dumont

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]