← 输入框|github.com|刷新
Skip to content

Add cxx feature cxx_noexcept - #3

Closed
hean01 wants to merge 1 commit into
solvespace:masterfrom
hean01:hean01/fixcompile
Closed

Add cxx feature cxx_noexcept#3
hean01 wants to merge 1 commit into
solvespace:masterfrom
hean01:hean01/fixcompile

Conversation

@hean01

@hean01 hean01 commented Feb 15, 2016

Copy link
Copy Markdown

Fixes build with gcc

@whitequark

whitequark commented Feb 15, 2016

Copy link
Copy Markdown
Contributor

CI builds with gcc-4.6 succeed. What gcc version and what compilation problem do you have?

@hean01

hean01 commented Feb 15, 2016

Copy link
Copy Markdown
Author

This is on FC23, gcc (GCC) 5.3.1 20151207 (Red Hat 5.3.1-2)

The errors are hundreds of lines but see the top of output down below...

Cmake will add -std=gnu++11, when defining compiler features cxx_noexcept, which is the correct way of doing this.

[ 15%] Built target slvs
[ 17%] Building CXX object src/CMakeFiles/solvespace.dir/gtk/gtkmain.cpp.o
In file included from /usr/include/c++/5.3.1/type_traits:35:0,
                 from /usr/include/sigc++-2.0/sigc++/visit_each.h:22,
                 from /usr/include/sigc++-2.0/sigc++/functors/slot.h:6,
                 from /usr/include/sigc++-2.0/sigc++/signal_base.h:27,
                 from /usr/include/sigc++-2.0/sigc++/signal.h:8,
                 from /usr/include/sigc++-2.0/sigc++/sigc++.h:86,
                 from /usr/include/glibmm-2.4/glibmm/iochannel.h:30,
                 from /usr/include/glibmm-2.4/glibmm/main.h:26,
                 from /home/henrik/Development/upstream/solvespace.git/src/gtk/gtkmain.cpp:18:
/usr/include/c++/5.3.1/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support for the ISO C++ 2011 standard. This support is currently experimental, and must be enabled with the -std=c++11 or -std=gnu++11 compiler options.
 #error This file requires compiler and library support for the \
  ^
In file included from /usr/include/glibmm-2.4/glibmm/main.h:23:0,
                 from /home/henrik/Development/upstream/solvespace.git/src/gtk/gtkmain.cpp:18:
/usr/include/glibmm-2.4/glibmm/refptr.h:83:70: error: ‘nullptr’ was not declared in this scope
     static const bool value = sizeof(check(static_cast<T_CppObject*>(nullptr))) == sizeof(big);
                                                                      ^
/usr/include/glibmm-2.4/glibmm/refptr.h:86:34: error: ‘Glib::RefPtr::is_allowed_type::value’ is not a type
   static_assert(is_allowed_type::value,
                                  ^
/usr/include/glibmm-2.4/glibmm/refptr.h:87:5: error: expected identifier before string constant
     "Glib::RefPtr must not be used with this class.");
     ^
/usr/include/glibmm-2.4/glibmm/refptr.h:87:5: error: expected ‘,’ or ‘...’ before string constant
/usr/include/glibmm-2.4/glibmm/refptr.h:87:53: error: ISO C++ forbids declaration of ‘static_assert’ with no type [-fpermissive]
     "Glib::RefPtr must not be used with this class.");
                                                     ^
/usr/include/glibmm-2.4/glibmm/refptr.h:95:17: error: expected ‘;’ at end of member declaration
   inline RefPtr() noexcept;
                 ^
/usr/include/glibmm-2.4/glibmm/refptr.h:95:19: error: ‘noexcept’ does not name a type
   inline RefPtr() noexcept;

@whitequark

whitequark commented Feb 15, 2016

Copy link
Copy Markdown
Contributor

Oh yes, I see, glibmm upgraded to requiring C++11 instead of C++0x.

Cmake will add -std=gnu++11, when defining compiler features cxx_noexcept, which is the correct way of doing this.

It's not. If I want C++11, I should request C++11, not some random feature that carries no semantic meaning as to why it's requested. It's failing on static_assert, not noexcept, isn't it? CMake's solution is stupid.

Ok, I will look into bumping the requirement. SolveSpace on non-Linux systems already carried the C++11 requirement for a while so it should be OK; mostly it's C++0x just for Travis CI, which has an old gcc for some reason.

@hean01

hean01 commented Feb 15, 2016

Copy link
Copy Markdown
Author

It's not. If I want C++11, I should request C++11, not some random feature that carries no semantic meaning as to why it's requested. It's failing on static_assert, not noexcept, isn't it? CMake's solution is stupid.

Ah, that is true. I got confused and cmake is cmake :). Bumping up the requirement for SolveSpace is the proper way of doing it.

@hean01 hean01 closed this Feb 15, 2016
@whitequark

whitequark commented Feb 15, 2016

Copy link
Copy Markdown
Contributor

Oh no, you aren't confused, this is indeed what the CMake manual recommends. But it's still stupid.

@whitequark

whitequark commented Feb 15, 2016

Copy link
Copy Markdown
Contributor

I've fixed this and updated the for-upstream branch.

@michthom michthom mentioned this pull request Nov 5, 2016
Closed
@wpwrak wpwrak mentioned this pull request Feb 3, 2017
Closed
@softechnoconrajat softechnoconrajat mentioned this pull request Dec 4, 2018
Closed
mdecourse referenced this pull request in mdecourse/solvespace Dec 13, 2019
@KmolYuan
Merge pull request #3 from solvespace/master
cfd630f 
Merge from origin master.
@phkahler phkahler mentioned this pull request May 1, 2020
Closed
@phkahler phkahler mentioned this pull request May 12, 2020
Closed
@ktrace ktrace mentioned this pull request May 14, 2022
Closed
@BoykoNeov BoykoNeov mentioned this pull request Aug 11, 2026
Merged
@ruevs ruevs mentioned this pull request Aug 25, 2026
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@hean01 @whitequark