[Buildroot] [PATCH v2] azmq: Add new package

Arnout Vandecappelle arnout at mind.be
Thu Aug 3 07:12:04 UTC 2017



On 03-08-17 01:45, RJ Ascani wrote:
> Hi Arnout,
> 
> Thank you so much for the review. I'm not in any rush, so making it
> into 2017.08 is not a necessity.
> 
> On Wed, Aug 2, 2017 at 3:29 PM, Arnout Vandecappelle <arnout at mind.be> wrote:
>>  Hi RJ,
>>
>>  I'm not sure if we can still get this in 2017.08 since I have a few comments
>> that I can't fix up myself...
>>
>> On 28-07-17 20:04, RJ Ascani wrote:
>>> The azmq library provides Boost Asio style bindings for ZeroMQ. This
>>> library is built on top of ZeroMQ's standard C interface and is intended
>>> to work well with C++ applications which use the Boost libraries in
>>> general, and Asio in particular.
>>>
>>> Signed-off-by: RJ Ascani <rj.ascani at gmail.com>
>>
>>  We require the Signed-off-by to be a full, real name because it's a legal
>> statement. It is a short way for you to assert that you are entitled to
>> contribute the patch under buildroot's GPL license.  See
>> http://elinux.org/Developer_Certificate_Of_Origin for more details.
>>
>>  The Author should be the same name, of course.
>>
> 
> Just to be clear, it's my initials that are the issue? I've gone by RJ
> my entire life so I didn't realize that would be an issue.

 In that case it's OK - it looked like an alias/nick, that's all.


>> [snip]
>>> +From d960f93fb52e373f76cfedbee5f4fb0c5276856a Mon Sep 17 00:00:00 2001
>>> +From: RJ Ascani <rj.ascani at gmail.com>
>>> +Date: Thu, 20 Jul 2017 15:53:40 -0700
>>> +Subject: [PATCH] Conditionally disable test and doc builds
>>
>>  What is the upstream status of this patch? Please send it upstream if you
>> didn't already do so.
>>
> 
> Actually, I probably should have elaborated a bit more in the
> explanation here. The reason I added this patch was actually because
> there is a bug in building the tests under buildroot. Upstream is
> missing a library that needs to be linked in their test
> CMakeLists.txt. Since buildroot already forcibly tries to disable docs
> and tests for cmake packages, I assumed it would be better to just
> have the tests not even be built for buildroot. I'm not certain I
> could get those conditionals into upstream, but I'm fairly certain I
> could get the test bug patched upstream. Would it be preferable to: 1)
> patch the actual test bug and let the tests build or 2) disable the
> building of the tests entirely?

 Disabling building of tests is preferable, yes. I don't see why upstream
wouldn't agree with a patch that makes that optional - especially if the name is
standard, like you mention below. So please send it upstream (after checking
that the defaults for both these symbols is ON).

>>> +
>>> +Signed-off-by: RJ Ascani <rj.ascani at gmail.com>
>>> +---
>>> + CMakeLists.txt | 12 +++++++++---
>>> + 1 file changed, 9 insertions(+), 3 deletions(-)
>>> +
>>> +diff --git a/CMakeLists.txt b/CMakeLists.txt
>>> +index 5b5b9cc..97006a4 100644
>>> +--- a/CMakeLists.txt
>>> ++++ b/CMakeLists.txt
>>> +@@ -49,7 +49,9 @@ include_directories(${Boost_INCLUDE_DIRS}
>>> +                     ${ZeroMQ_INCLUDE_DIRS}
>>> +                     ${PROJECT_SOURCE_DIR})
>>> +
>>> +-enable_testing()
>>> ++if (BUILD_TESTING)
>>
>>  I would expect that upstream would require BUILD_TESTING to be set to ON by
>> default. Also, I think BUILD_TESTS is more common, but I have no data to back
>> that up.
> 
> BUILD_TESTING is an actually a built-in for CTest (CMake's testing module).

 OK, I didn't know that, thanks!

[snip]
>>
>>  So I would expect that we would need to pass an explicit -DBUILD_TESTING=OFF
>> and -DBUILD_DOCS=OFF
>>
> 
> Those are already added to the cmake configure command from
> package/pkg-cmake.mk. Are you saying we should also add them as an
> explicit option in AZMQ_CONF_OPTS?

 D'oh, I actually knew that, silly me!

 So the patch is OK then, I've applied to the next branch unchanged.

 Could you still send a patch that adds you to the DEVELOPERS file? That way,
you'll get a personal mail when this package breaks in the autobuilders.

 Regards,
 Arnout


-- 
Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF



More information about the buildroot mailing list