This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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]

[PATCH] PR c++/43890


Hello,

This fix a 4.6 regression, coming from the fix for PR c++/25811.
We need to check recursively for a user-provided constructor.

I've taken the opportunity to change the error message from:
"uninitialized const member in %q#T using %<new%>" to  "uninitialized
const member in %q#T using %<new%>"whithout new-initializer" which
sounds better to me.

(One failure in g++.dg/warn/string1.C but i can't believe this is mine)
Tested x86_64-unknown-linux-gnu. OK for trunk ?


gcc/testsuite/ChangeLog:

2010-04-26  Fabien Chêne  <fabien.chene@gmail.com>
	PR c++/43890
	* g++.dg/init/pr43890.C: New.

gcc/cp/ChangeLog:

2010-04-26  Fabien Chêne  <fabien.chene@gmail.com>
	PR c++/43890
	* init.c (diagnose_uninitialized_cst_or_ref_member): check for
	user-provided constructor while recursing.
-- 
Fab

Attachment: pr43890.patch
Description: Binary data


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