Le 07/12/2022 à 17:21, Richard Kimberly Heck a écrit :
On 12/7/22 11:04, Jean-Marc Lasgouttes wrote:
Le 06/12/2022 à 16:17, Jean-Marc Lasgouttes a écrit :
commit f3f478c2fbcf89302ccadeb30aac413a40039d12
Author: Jean-Marc Lasgouttes <lasgout...@lyx.org>
Date: Tue Dec 6 17:12:45 2022 +0100
C++11 compilation fix
for (iter = refs_.begin(); iter != refs_.end(); ++iter) {
// first: plain label name, second: gui name
QString const lab = toqstr((*iter).first);
- refsNames.append(QPair(lab, toqstr((*iter).second)));
+ refsNames.append({lab, toqstr((*iter).second)});
This elicited the answer "Huh?" from Jürgen, but sent to the lyx-cvs
list, so only Riki and I saw it.
Could I have a authoritative answer to the question whether a
constructor like QPair(a, b) can be replaced with {a, b} when the
context is clear enough?
I notice that we use that already, but I may be wrong about when it is
acceptable.
I don't have a strong view about this. It seems to me a case where, if
you're not used to it, it will look weird. That said, the {a,b}
construction is less explicit.
Right, but we are supposed to use something like QPair<SQtring,
QString>(a,b), which is awfully explicit.
JMarc
--
lyx-devel mailing list
lyx-devel@lists.lyx.org
http://lists.lyx.org/mailman/listinfo/lyx-devel
