This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: C++ PATCH: Fix PR 13275
"Zack Weinberg" <zack@codesourcery.com> writes:
| Gabriel Dos Reis <gdr@integrable-solutions.net> writes:
|
| > "Zack Weinberg" <zack@codesourcery.com> writes:
| >
| > | Gabriel Dos Reis <gdr@integrable-solutions.net> writes:
| > |
| > | > It implements something that is not offsetof, it is a real extension
| > | > -- let's call a cat a cat. I'm not sure it is advisable to set the
| > | > precedent where this extension, which does not implement offsetof,
| > | > is threated this way.
| > |
| > | Would you like to submit a patch which does it the way you want it
| > | done?
| >
| > There are at least three issues here:
| >
| > (1) Do we want to hide that extension under the name of implementing
| > offset -- where actually we do not?
|
| *shrug*
|
| > (2) It was objected, that one would be doing ugly things with the
| > grammar, but no preceision was given. I would appreciate to
| > have some outline of such things (so that they may be avoided)
|
| can't comment
|
| > (3) Do we want to that precendent? I may be wrong but my impression
| > is that your question is somewhat rhetorical, and I would love
| > to be wrong there.
|
| not sure what you mean. I'm trying to make a meta-point: given that
| we are in stage 3, and given that we have a real bug which the patch
Note: We never actually implemented an offsetof for C++ since, well,
dark ages.
| does fix, objections of the form "It should be done in a totally
| different way" would carry significantly more weight if they came
I think I get an answer for (3).
| attached to alternative patches. Particularly if the totally
| different way might be too invasive.
-- Gaby