CMake Announcement from Boost Steering Committee?

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
16 messages Options
Reply | Threaded
Open this post in threaded view
|

CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
Hello Boost builders

Someone brings me up the annoucement from the Steering Committee to choose CMake as the open source build system for Boost -- reproduced partly below:

"Therefore, we, the Steering Committee, announce to the Boost community our desire and intent to move Boost’s build system to CMake for users and developers alike. We are soliciting comments and proposals from the community to guide the process and the goals. Our desire is that the community can come to consensus by the end of the calendar year with a vision of supporting users and developers."

I am wondering if any alternative choice as Gradle has been considered instead of choosing CMake?

Thank for any view on this subject.

Francis ANDRE

PS: I looked for Gradle in the Boost archive on Nabble : Found 0 matching posts for gradle in Boost




_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
(resent from the correct address)


Hi Francis,

On 13 Aug 2017 10:16 pm, "Francis ANDRE via Boost-build" <[hidden email]> wrote:

I am wondering if any alternative choice as Gradle has been considered instead of choosing CMake?

Thank for any view on this subject.

Francis ANDRE

PS: I looked for Gradle in the Boost archive on Nabble : Found 0 matching posts for gradle in Boost


Thanks for the pointer, but no thanks.

I definitely wouldn't want to install a JVM just to build boost.

Andrea (a boost user)


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
In reply to this post by Boost - Build mailing list
What is the fate of existing Boost Build Jam scripts and infrastructure? We have chosen Boost Build over CMake years ago and have made a considerable investment in creating our software build infrastructure using JAM scripts. What is to be?

Many thanks,

Doug C.

On Aug 13, 2017, at 11:09 AM, Francis ANDRE via Boost-build wrote:

Hello Boost builders

Someone brings me up the annoucement from the Steering Committee to choose CMake as the open source build system for Boost -- reproduced partly below:

"Therefore, we, the Steering Committee, announce to the Boost community our desire and intent to move Boost’s build system to CMake for users and developers alike. We are soliciting comments and proposals from the community to guide the process and the goals. Our desire is that the community can come to consensus by the end of the calendar year with a vision of supporting users and developers."

I am wondering if any alternative choice as Gradle has been considered instead of choosing CMake?

Thank for any view on this subject.

Francis ANDRE

PS: I looked for Gradle in the Boost archive on Nabble : Found 0 matching posts for gradle in Boost



_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
In reply to this post by Boost - Build mailing list
On 13 Aug 2017 11:29 pm, "Francis ANDRE" <[hidden email]> wrote:

Le 13/08/2017 à 22:58, Andrea Bocci a écrit :
I definitely wouldn't want to install a JVM just to build boost.
If I make a POC of building Boost for VS2015 with debug/release, shared/static in 4 days, would you reconsider your position?

Hi Francis,
I'm not sure what a POC is - and as I signed I am just a user here.

From my point of view, it is not a matter of how quickly Boost could be ported to build with Gradle or any other Java-based build system: the downside would simply be the added dependency on Java.

Ciao,
Andrea (a boost user)


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
In reply to this post by Boost - Build mailing list
On Aug 13, 2017 3:15 PM, "Francis ANDRE via Boost-build" <[hidden email]> wrote:
I am wondering if any alternative choice as Gradle has been considered instead of choosing CMake?

Thank for any view on this subject.

This list is for discussion of the Boost Build program. Not for discussion of what build system Boost C++ Libraries uses. Perhaps you want to ask in the developers list.


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
In reply to this post by Boost - Build mailing list
On 8/13/2017 6:11 PM, Douglas Capeci via Boost-build wrote:
> What is the fate of existing Boost Build Jam scripts and infrastructure?
> We have chosen Boost Build over CMake years ago and have made a
> considerable investment in creating our software build infrastructure
> using JAM scripts. What is to be?

Please do not top post.

There has been no official word from the Boost Steering Committee
regarding Boost Build. My educated guess is that Boost Build will remain
as a Boost tool, and as an alternative to those library developers who
want to provide jam files along with the required CMake support, when
Boost adopts CMake as their primary build tool. I would expect this to
happen precisely because of programmers like you who have an investment
in Boost Build jamfiles. But if this did happen you would have to expect
most, if not all, new libraries would only support CMake and that
current Boost libraries might well drop Boost Build support, since
maintaining two separate build methodologies is an extra burden for a
library maintainer. But until the Boost Steering Committee makes a
decision regarding this, no one can know for sure what will happen.

>
> Many thanks,
>
> Doug C.
>
> On Aug 13, 2017, at 11:09 AM, Francis ANDRE via Boost-build wrote:
>
>> Hello Boost builders
>>
>> Someone brings me up the annoucement from the Steering Committee to
>> choose CMake as the open source build system for Boost -- reproduced
>> partly below:
>>
>> "Therefore, we, the Steering Committee, announce to the Boost
>> community our desire and intent to move Boost’s build system to
>> CMake for users and developers alike. We are soliciting comments and
>> proposals from the community to guide the process and the goals. Our
>> desire is that the community can come to consensus by the end of the
>> calendar year with a vision of supporting users and developers."
>>
>> I am wondering if any alternative choice as Gradle has been considered
>> instead of choosing CMake?
>>
>> Thank for any view on this subject.
>>
>> Francis ANDRE
>>
>> PS: I looked for Gradle in the Boost archive on Nabble : Found 0
>> matching posts for *gradle* in Boost <http://boost.2283326.n4.nabble.com/>
>>
>>
>>
>> _______________________________________________
>> Unsubscribe & other changes:
>> https://lists.boost.org/mailman/listinfo.cgi/boost-build
>
>
>
> _______________________________________________
> Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
>


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
A sigh of relief in knowing that Boost Build will continue as a supported Boost tool offering.

But why is CMake or another tool being considered to replace the current Boost Library build system? Jam is simple, elegant, easy to learn, and works very nicely for C++ tool enviroments and in MHO is better than CMake and any of its competitors. If it isn't broken why fix it?


On Aug 13, 2017, at 8:08 PM, Edward Diener via Boost-build wrote:

> On 8/13/2017 6:11 PM, Douglas Capeci via Boost-build wrote:
>> What is the fate of existing Boost Build Jam scripts and infrastructure? We have chosen Boost Build over CMake years ago and have made a considerable investment in creating our software build infrastructure using JAM scripts. What is to be?
>
> Please do not top post.
>
> There has been no official word from the Boost Steering Committee regarding Boost Build. My educated guess is that Boost Build will remain as a Boost tool, and as an alternative to those library developers who want to provide jam files along with the required CMake support, when Boost adopts CMake as their primary build tool. I would expect this to happen precisely because of programmers like you who have an investment in Boost Build jamfiles. But if this did happen you would have to expect most, if not all, new libraries would only support CMake and that current Boost libraries might well drop Boost Build support, since maintaining two separate build methodologies is an extra burden for a library maintainer. But until the Boost Steering Committee makes a decision regarding this, no one can know for sure what will happen.
>
>> Many thanks,
>> Doug C.
>> On Aug 13, 2017, at 11:09 AM, Francis ANDRE via Boost-build wrote:
>>> Hello Boost builders
>>>
>>> Someone brings me up the annoucement from the Steering Committee to choose CMake as the open source build system for Boost -- reproduced partly below:
>>>
>>> "Therefore, we, the Steering Committee, announce to the Boost community our desire and intent to move Boost’s build system to CMake for users and developers alike. We are soliciting comments and proposals from the community to guide the process and the goals. Our desire is that the community can come to consensus by the end of the calendar year with a vision of supporting users and developers."
>>>
>>> I am wondering if any alternative choice as Gradle has been considered instead of choosing CMake?
>>>
>>> Thank for any view on this subject.
>>>
>>> Francis ANDRE
>>>
>>> PS: I looked for Gradle in the Boost archive on Nabble : Found 0 matching posts for *gradle* in Boost <http://boost.2283326.n4.nabble.com/>
>>>
>>>
>>>
>>> _______________________________________________
>>> Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
>> _______________________________________________
>> Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
>
>
> _______________________________________________
> Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
On Sun, Aug 13, 2017 at 7:55 PM, Douglas Capeci via Boost-build <[hidden email]> wrote:
A sigh of relief in knowing that Boost Build will continue as a supported Boost tool offering.

I would go farther and say that it's going to get better support now than before. Here's at least my idea of what will happen to Boost Build <https://lists.boost.org/boost-build/2017/07/29461.php>.
 
But why is CMake or another tool being considered to replace the current Boost Library build system? Jam is simple, elegant, easy to learn, and works very nicely for C++ tool enviroments and in MHO is better than CMake and any of its competitors. If it isn't broken why fix it?

First, you would have to ask the proponents of cmake and the Steering Committee about that. Second, can I quote you on that statement for publicity of Boost Build?

--
-- Rene Rivera
-- Grafik - Don't Assume Anything
-- Robot Dreams - http://robot-dreams.net
-- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
You certainly may quote me, and my strong support for keeping Boost Build. 

My group has carefully evaluated other alternatives (one of them being cMake) and we found Boost Build and Jam to be far superior for C++ systems than any of the others. We liked it so much that we have integrated Boost Build and Jam into our Sun Grid Engine compute environment to perform all of our C++ software builds.



On Aug 13, 2017, at 9:38 PM, Rene Rivera via Boost-build wrote:

On Sun, Aug 13, 2017 at 7:55 PM, Douglas Capeci via Boost-build <[hidden email]> wrote:
A sigh of relief in knowing that Boost Build will continue as a supported Boost tool offering.

I would go farther and say that it's going to get better support now than before. Here's at least my idea of what will happen to Boost Build <https://lists.boost.org/boost-build/2017/07/29461.php>.
 
But why is CMake or another tool being considered to replace the current Boost Library build system? Jam is simple, elegant, easy to learn, and works very nicely for C++ tool enviroments and in MHO is better than CMake and any of its competitors. If it isn't broken why fix it?

First, you would have to ask the proponents of cmake and the Steering Committee about that. Second, can I quote you on that statement for publicity of Boost Build?

--
-- Rene Rivera
-- Grafik - Don't Assume Anything
-- Robot Dreams - http://robot-dreams.net
-- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail
_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
In reply to this post by Boost - Build mailing list
On 8/13/2017 8:55 PM, Douglas Capeci via Boost-build wrote:
> A sigh of relief in knowing that Boost Build will continue as a supported Boost tool offering.

I never said that. It is just my guess.

>
> But why is CMake or another tool being considered to replace the current Boost Library build system? Jam is simple, elegant, easy to learn, and works very nicely for C++ tool enviroments and in MHO is better than CMake and any of its competitors. If it isn't broken why fix it?
>
>
> On Aug 13, 2017, at 8:08 PM, Edward Diener via Boost-build wrote:
>
>> On 8/13/2017 6:11 PM, Douglas Capeci via Boost-build wrote:
>>> What is the fate of existing Boost Build Jam scripts and infrastructure? We have chosen Boost Build over CMake years ago and have made a considerable investment in creating our software build infrastructure using JAM scripts. What is to be?
>>
>> Please do not top post.
>>
>> There has been no official word from the Boost Steering Committee regarding Boost Build. My educated guess is that Boost Build will remain as a Boost tool, and as an alternative to those library developers who want to provide jam files along with the required CMake support, when Boost adopts CMake as their primary build tool. I would expect this to happen precisely because of programmers like you who have an investment in Boost Build jamfiles. But if this did happen you would have to expect most, if not all, new libraries would only support CMake and that current Boost libraries might well drop Boost Build support, since maintaining two separate build methodologies is an extra burden for a library maintainer. But until the Boost Steering Committee makes a decision regarding this, no one can know for sure what will happen.
>>
>>> Many thanks,
>>> Doug C.
>>> On Aug 13, 2017, at 11:09 AM, Francis ANDRE via Boost-build wrote:
>>>> Hello Boost builders
>>>>
>>>> Someone brings me up the annoucement from the Steering Committee to choose CMake as the open source build system for Boost -- reproduced partly below:
>>>>
>>>> "Therefore, we, the Steering Committee, announce to the Boost community our desire and intent to move Boost’s build system to CMake for users and developers alike. We are soliciting comments and proposals from the community to guide the process and the goals. Our desire is that the community can come to consensus by the end of the calendar year with a vision of supporting users and developers."
>>>>
>>>> I am wondering if any alternative choice as Gradle has been considered instead of choosing CMake?
>>>>
>>>> Thank for any view on this subject.
>>>>
>>>> Francis ANDRE
>>>>
>>>> PS: I looked for Gradle in the Boost archive on Nabble : Found 0 matching posts for *gradle* in Boost <http://boost.2283326.n4.nabble.com/>

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
In reply to this post by Boost - Build mailing list
On 8/13/17 7:03 PM, Douglas Capeci via Boost-build wrote:
> You certainly may quote me, and my strong support for keeping Boost Build.
>
> My group has carefully evaluated other alternatives (one of them being
> cMake) and we found Boost Build and Jam to be far superior for C++
> systems than any of the others. We liked it so much that we have
> integrated Boost Build and Jam into our Sun Grid Engine compute
> environment to perform all of our C++ software builds.
>
>

Perhaps your group might want to consider contributing to boost build
development, maintenance, testing or documentation.  Since your already
very familiar with it, it would entail only modest expenditure of
resources.  This would also be the one thing that you can do to ensure
it's continue availability.

Robert Ramey

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
Our schedules and resources are tight through the end of the year, but certainly I am open to understanding what is required to keep it going for the long-term, so I can decide what resources on my end could be available for helping out.

The thing that gets my goat here is that I don't see any clear reasoning or explanation as to why the steering committee has started the process to move in the direction of phasing out Boost Build. Doesn't the user community deserve an explanation of this?

DougC

On Aug 14, 2017, at 10:26 AM, Robert Ramey via Boost-build wrote:

> On 8/13/17 7:03 PM, Douglas Capeci via Boost-build wrote:
>> You certainly may quote me, and my strong support for keeping Boost Build.
>> My group has carefully evaluated other alternatives (one of them being cMake) and we found Boost Build and Jam to be far superior for C++ systems than any of the others. We liked it so much that we have integrated Boost Build and Jam into our Sun Grid Engine compute environment to perform all of our C++ software builds.
>
> Perhaps your group might want to consider contributing to boost build development, maintenance, testing or documentation.  Since your already very familiar with it, it would entail only modest expenditure of resources.  This would also be the one thing that you can do to ensure it's continue availability.
>
> Robert Ramey
>
> _______________________________________________
> Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
On Mon, Aug 14, 2017 at 9:42 AM, Douglas Capeci via Boost-build <[hidden email]> wrote:
Our schedules and resources are tight through the end of the year, but certainly I am open to understanding what is required to keep it going for the long-term, so I can decide what resources on my end could be available for helping out.

I'll keep that in mind when asking for help in the future :-)
 
The thing that gets my goat here is that I don't see any clear reasoning or explanation as to why the steering committee has started the process to move in the direction of phasing out Boost Build. Doesn't the user community deserve an explanation of this?

I thought the reason was obvious.. They decided cmake was the "popular" choice. I.e. it'w what the plurality of users want.

--
-- Rene Rivera
-- Grafik - Don't Assume Anything
-- Robot Dreams - http://robot-dreams.net
-- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
I have found that most cMake advocates don't really understand how much better of a world Boost Build offers. They also have a hard time getting out of the dinosaur cave to give up the old arduous ways of maintaining complex Makefile based build systems to simply learn JAM. I would argue that the current cMake product front-end (and that is all it is...) is the result of cMake developers taking a hard and close look at Boost Build and understanding its elegance and superiority.


On Aug 14, 2017, at 10:52 AM, Rene Rivera via Boost-build wrote:

On Mon, Aug 14, 2017 at 9:42 AM, Douglas Capeci via Boost-build <[hidden email]> wrote:
Our schedules and resources are tight through the end of the year, but certainly I am open to understanding what is required to keep it going for the long-term, so I can decide what resources on my end could be available for helping out.

I'll keep that in mind when asking for help in the future :-)
 
The thing that gets my goat here is that I don't see any clear reasoning or explanation as to why the steering committee has started the process to move in the direction of phasing out Boost Build. Doesn't the user community deserve an explanation of this?

I thought the reason was obvious.. They decided cmake was the "popular" choice. I.e. it'w what the plurality of users want.

--
-- Rene Rivera
-- Grafik - Don't Assume Anything
-- Robot Dreams - http://robot-dreams.net
-- rrivera/acm.org (msn) - grafikrobot/aim,yahoo,skype,efnet,gmail
_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build


_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list
On 8/14/2017 10:10 AM, Douglas Capeci via Boost-build wrote:
> I have found that most cMake advocates don't really understand how much better of a world Boost Build offers. They also have a hard time
> getting out of the dinosaur cave to give up the old arduous ways of maintaining complex Makefile based build systems to simply learn JAM.
> I would argue that the current cMake product front-end (and that is all it is...) is the result of cMake developers taking a hard and
> close look at Boost Build and understanding its elegance and superiority.
I believe that CMake became popular in the open source community precisely because maintaining CMakefiles is very much like maintaining an
autotools system Makefile, plus it also makes Visual C++ projects for native Windows support (cue applause). Switching to cmake, after using
autotools for a long time, doesn't require rethinking what a build system should do.

Boost.Build could probably use a "Why Boost.Build instead of Makefiles?" motivational tutorial. My autotools-fu was never that great to
begin with, so I wouldn't be the right one to write it, that's for sure.

As for existing Boost libs dropping support for Boost.Build, I'm not too concerned. I use Rene's boost-ext.jam extensions file and update it
whenever I need a new lib from Boost, or updating Boost requires necessitates changes to it.

-Matt
_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build
Reply | Threaded
Open this post in threaded view
|

Re: [Bulk] Re: [Bulk] Re: [Bulk] Re: [Bulk] Re: [Bulk] CMake Announcement from Boost Steering Committee?

Boost - Build mailing list

On Aug 14, 2017, at 12:05 PM, Chambers, Matthew via Boost-build wrote:

> On 8/14/2017 10:10 AM, Douglas Capeci via Boost-build wrote:
>> I have found that most cMake advocates don't really understand how much better of a world Boost Build offers. They also have a hard time getting out of the dinosaur cave to give up the old arduous ways of maintaining complex Makefile based build systems to simply learn JAM. I would argue that the current cMake product front-end (and that is all it is...) is the result of cMake developers taking a hard and close look at Boost Build and understanding its elegance and superiority.
> I believe that CMake became popular in the open source community precisely because maintaining CMakefiles is very much like maintaining an autotools system Makefile, plus it also makes Visual C++ projects for native Windows support (cue applause). Switching to cmake, after using autotools for a long time, doesn't require rethinking what a build system should do.
The overcomplexity and slowness that GNU autotools flow has placed on our C++ software build system is exactly why we scrapped it for Boost Build. With Boost Build, our software build system is much smaller, faster, and is easily scalable to a standard compute grid engines. cMake does not offer these benefits that very large (>50M lines of code) C++ software systems demand.

The Boost Library has always been focused to provide the best solutions for C++ developers, and now it seems that this hard-core focus is beginning to diminish by trying to accommodate mixed language IDE tool environments which are really toys in the real-world.
>
> Boost.Build could probably use a "Why Boost.Build instead of Makefiles?" motivational tutorial. My autotools-fu was never that great to begin with, so I wouldn't be the right one to write it, that's for sure.
>
> As for existing Boost libs dropping support for Boost.Build, I'm not too concerned. I use Rene's boost-ext.jam extensions file and update it whenever I need a new lib from Boost, or updating Boost requires necessitates changes to it.
>
> -Matt
> _______________________________________________
> Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build

_______________________________________________
Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build