Merge lp:~vorlon/process-cpp/explicit-gcc-version into lp:process-cpp

[Marcus Tomlinson]

Bump major revision and so name to account for toolchain update.

Bump major revision and so name to account for toolchain update.

[Marcus Tomlinson (marcustomlinson)]

Steve:

This branch duplicates changes already appearing in lp:~thomas-voss/process-cpp/bump-so-name-and-major-version

Could you please, rebase this branch so that your diff shows only the additions you've made.

review: Needs Fixing

[Marcus Tomlinson (marcustomlinson)]

Otherwise, is it not cleaner to just have tvoss update his branch with this change?

[PS Jenkins bot (ps-jenkins)]

FAILED: Continuous integration, rev:54
No commit message was specified in the merge proposal. Click on the following link and set the commit message (if you want a jenkins rebuild you need to trigger it yourself):
https://code.launchpad.net/~vorlon/process-cpp/explicit-gcc-version/+merge/226569/+edit-commit-message

http://jenkins.qa.ubuntu.com/job/process-cpp-ci/68/
Executed test runs:
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-amd64-ci/7
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-armhf-ci/7
        deb: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-armhf-ci/7/artifact/work/output/*zip*/output.zip
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-i386-ci/7

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/process-cpp-ci/68/rebuild

review: Needs Fixing (continuous-integration)

[Steve Langasek (vorlon)]

Marcus,

"prerequisite" implies that the other branch should be merged independently and that my branch should be merged afterwards. As I described at <https://code.launchpad.net/~thomas-voss/process-cpp/explicit-gcc-version/+merge/224755>, which I have marked as rejected, Thomas's branch is broken and causes a build failure on one architecture and can therefore not be landed. There is no reason for these branches to be landed serially when the first one is broken.

As for asking tvoss to include the changes in his own branch, this is part of a landing of 17 distinct branches that have been pending since the 27th of June and are blocking a toolchain update for Utopic. As I'm working on getting the landing unstuck by clearing the build failures, it would not make sense to wait until Thomas can merge these changes into his branch when I can just raise a new MP myself.

[PS Jenkins bot (ps-jenkins)]

PASSED: Continuous integration, rev:54
http://jenkins.qa.ubuntu.com/job/process-cpp-ci/69/
Executed test runs:
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-amd64-ci/8
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-armhf-ci/8
        deb: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-armhf-ci/8/artifact/work/output/*zip*/output.zip
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-i386-ci/8

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/process-cpp-ci/69/rebuild

review: Approve (continuous-integration)

[Marcus Tomlinson (marcustomlinson)]

> Marcus,
>
> "prerequisite" implies that the other branch should be merged independently
> and that my branch should be merged afterwards. As I described at
> <https://code.launchpad.net/~thomas-voss/process-cpp/explicit-gcc-
> version/+merge/224755>, which I have marked as rejected, Thomas's branch is
> broken and causes a build failure on one architecture and can therefore not be
> landed. There is no reason for these branches to be landed serially when the
> first one is broken.

Ah, forgive me for finding this arrangement of branches confusing. So there is a 3rd branch with the exact same stuff in it? Perhaps you should reject lp:~thomas-voss/process-cpp/bump-so-name-and-major-version as well. Its difficult to tell which branch is the master.

>
> As for asking tvoss to include the changes in his own branch, this is part of
> a landing of 17 distinct branches that have been pending since the 27th of
> June and are blocking a toolchain update for Utopic. As I'm working on
> getting the landing unstuck by clearing the build failures, it would not make
> sense to wait until Thomas can merge these changes into his branch when I can
> just raise a new MP myself.

I've clearly intruded on something you're in control of. Sorry.

review: Abstain

[Steve Langasek (vorlon)]

Aha, thanks, I didn't even notice that the branch you mentioned wasn't the same one I had superseded the MP for! Yes, I'll mark that other MP rejected as well now.

[Marcus Tomlinson (marcustomlinson)]

Resubmitted as originally intended. Sorry again about the prerequisite mixup.

[PS Jenkins bot (ps-jenkins)]

PASSED: Continuous integration, rev:54
http://jenkins.qa.ubuntu.com/job/process-cpp-ci/70/
Executed test runs:
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-amd64-ci/9
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-armhf-ci/9
        deb: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-armhf-ci/9/artifact/work/output/*zip*/output.zip
    SUCCESS: http://jenkins.qa.ubuntu.com/job/process-cpp-utopic-i386-ci/9

Click here to trigger a rebuild:
http://s-jenkins.ubuntu-ci:8080/job/process-cpp-ci/70/rebuild

review: Approve (continuous-integration)

[Marcus Tomlinson (marcustomlinson)]

fwiw, this looks good to me.

review: Approve

[Thomas Voß (thomas-voss)]

LGTM.

review: Approve

=== modified file 'CMakeLists.txt'
--- CMakeLists.txt	2014-05-21 16:56:20 +0000
+++ CMakeLists.txt	2014-07-12 22:57:38 +0000
@@ -56,9 +56,9 @@
   SET(CMAKE_SHARED_LINKER_FLAGS "${CMAKE_SHARED_LINKER_FLAGS} -ftest-coverage -fprofile-arcs" )
 ENDIF(cmake_build_type_lower MATCHES coverage)
 
-set(PROCESS_CPP_VERSION_MAJOR 1)
+set(PROCESS_CPP_VERSION_MAJOR 2)
 set(PROCESS_CPP_VERSION_MINOR 0)
-set(PROCESS_CPP_VERSION_PATCH 1)
+set(PROCESS_CPP_VERSION_PATCH 0)
 
 include(CTest)
 

=== modified file 'debian/changelog'
--- debian/changelog	2014-05-22 09:49:51 +0000
+++ debian/changelog	2014-07-12 22:57:38 +0000
@@ -1,3 +1,9 @@
+process-cpp (2.0.0) UNRELEASED; urgency=medium
+
+  * Bump major revision and so name to account for toolchain update.
+
+ -- Thomas Voß <[e-mail]>  Fri, 27 Jun 2014 09:49:30 +0200
+
 process-cpp (1.0.1+14.10.20140522-0ubuntu1) utopic; urgency=medium
 
   [ Pete Woods ]

=== modified file 'debian/control'
--- debian/control	2014-03-24 17:12:32 +0000
+++ debian/control	2014-07-12 22:57:38 +0000
@@ -6,6 +6,10 @@
                doxygen,
                google-mock,
                graphviz,
+# We rely on C++11 features, and to prevent from ABI breaks
+# in libstdc++ causing us issues, we explicitly select a G++
+# version.
+               g++-4.9,
                libboost-dev,
                libboost-iostreams-dev,
                libboost-system-dev,
@@ -19,7 +23,7 @@
 Vcs-Bzr: https://code.launchpad.net/~phablet-team/content-hub/trunk
 Vcs-Browser: https://bazaar.launchpad.net/~phablet-team/process-cpp/trunk/files
 
-Package: libprocess-cpp1
+Package: libprocess-cpp2
 Architecture: any
 Multi-Arch: same
 Pre-Depends: ${misc:Pre-Depends},
@@ -35,7 +39,7 @@
 Architecture: any
 Multi-Arch: same
 Recommends: libprocess-cpp-doc,
-Depends: libprocess-cpp1 (= ${binary:Version}),
+Depends: libprocess-cpp2 (= ${binary:Version}),
          ${misc:Depends},
          libproperties-cpp-dev,
 Description: C++11 library for handling processes - dev headers and libraries

=== renamed file 'debian/libprocess-cpp1.install' => 'debian/libprocess-cpp2.install'
=== renamed file 'debian/libprocess-cpp1.symbols' => 'debian/libprocess-cpp2.symbols'
--- debian/libprocess-cpp1.symbols	2014-05-22 09:49:50 +0000
+++ debian/libprocess-cpp2.symbols	2014-07-12 22:57:38 +0000
@@ -1,4 +1,4 @@
-libprocess-cpp.so.1 libprocess-cpp1 #MINVER#
+libprocess-cpp.so.2 libprocess-cpp2 #MINVER#
  (c++)"core::posix::Signalable::send_signal(core::posix::Signal, std::error_code&)@Base" 1.0.0+14.04.20140326.2
  (c++)"core::posix::Signalable::send_signal_or_throw(core::posix::Signal)@Base" 1.0.0+14.04.20140326.2
  (c++)"core::posix::ChildProcess::DeathObserver::create_once_with_signal_trap(std::shared_ptr<core::posix::SignalTrap>)@Base" 1.0.0+14.04.20140326.2
@@ -8,7 +8,6 @@
  (c++)"core::posix::ChildProcess::invalid()@Base" 0.0.1+14.04.20140311
  (c++)"core::posix::ChildProcess::wait_for(core::posix::wait::Flags const&)@Base" 0.0.1+14.04.20140311
  (c++)"core::posix::ChildProcess::~ChildProcess()@Base" 1.0.0+14.04.20140326.2
- (c++|arch=ppc64el)"core::posix::ChildProcess::ChildProcess(core::posix::ChildProcess const&)@Base" 1.0.0+14.04.20140326.2
  (c++)"core::posix::this_process::cin()@Base" 1.0.0+14.04.20140326.2
  (c++)"core::posix::this_process::env::get_or_throw(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)@Base" 1.0.0+14.04.20140326.2
  (c++)"core::posix::this_process::env::set_or_throw(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)@Base" 1.0.0+14.04.20140326.2

=== modified file 'debian/rules'
--- debian/rules	2013-10-21 12:07:28 +0000
+++ debian/rules	2014-07-12 22:57:38 +0000
@@ -3,5 +3,15 @@
 
 export DPKG_GENSYMBOLS_CHECK_LEVEL=4
 
+include /usr/share/dpkg/default.mk
+
+# Explicitly selecting a G{CC,++}-version here to avoid accidental
+# ABI breaks introduced by toolchain updates.
+export CC=$(DEB_HOST_GNU_TYPE)-gcc-4.9
+export CXX=$(DEB_HOST_GNU_TYPE)-g++-4.9
+
 %:
 	dh $@ --fail-missing
+
+override_dh_auto_configure:
+	dh_auto_configure -- -DCMAKE_C_COMPILER=$(CC) -DCMAKE_CXX_COMPILER=$(CXX)