[release] Boost 1.71.0 Release Candidate 1

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

[release] Boost 1.71.0 Release Candidate 1

Boost - Dev mailing list
The first release candidates for the 1.71.0 release are now available at:

<https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>

The SHA256 checksums are as follows:

7fadf835372325d7a287b1c45310c4f96811eb1e4809933a5a400d529df38d0e
boost_1_71_0_rc1.7z

c857a86fb1223c7a256414151e2d1dcd4691b56197aa1329e087e92502191434
boost_1_71_0_rc1.tar.bz2

07c07cad079fed3f0175cbd7097740a333c2cdc6c899ac433dea87d1224d726d
boost_1_71_0_rc1.tar.gz

bc1674ba3e11635fc2effbddc2ed69aecaced3901f81a6a2aae7b77f34e4cf76
boost_1_71_0_rc1.zip

As always, the release managers would appreciate it if you download the
candidate of your choice and give building it a try. Please report both
success and failure, and anything else that is noteworthy.

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

Geometry Extensions docs missing or link broken (was: [release] Boost 1.71.0 Release Candidate 1)

Boost - Dev mailing list
On Fri, 9 Aug 2019 at 20:44, Michael Caisse via Boost
<[hidden email]> wrote:
>
> The first release candidates for the 1.71.0 release are now available at:
> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
> Please report (...) anything else that is noteworthy.

Page libs/libraries.htm lists "Geometry Extensions" leading to
libs/geometry/extensions/index.html which is missing.

I don't know if that "Geometry Extensions" should be updated or removed,
but I'm raising this issue for Adam or Barend who may know better.

Best regards,
--
Mateusz Loskot, http://mateusz.loskot.net

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

GIL docs missing again (was: [release] Boost 1.71.0 Release Candidate 1)

Boost - Dev mailing list
In reply to this post by Boost - Dev mailing list
On Fri, 9 Aug 2019 at 20:44, Michael Caisse via Boost
<[hidden email]> wrote:
>
> The first release candidates for the 1.71.0 release are now available at:
> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
> [...]
> As always, the release managers would appreciate it if you download the
> candidate of your choice and give building it a try. Please report both
> success and failure, and anything else that is noteworthy.

Despite my attempts [1] & [2] to figure out what is going wrong
that prevents GIL docs from being generated.

The local `cd libs/gil && b2 doc` workflow is fine
- first doxygen generates files in doc/html
- then, sphinx-build generates files in doc/html
for both, master and develop.

Is it possible that this GIL top-level Jamfile in master
https://github.com/boostorg/gil/blob/master/Jamfile
with C++11 checks is the troublemaker here?
i.e. the checks fail on the machine where the Boost docs aregenerated

In GIL develop, after an advisory round from Peter and Rene,
I have removed moved the big-fat content from the top-level
Jamfile to test/Jamfile, here is the change
https://github.com/boostorg/gil/commit/4ed7701b47764edf327e7c82f65cf9815026bdc2

How do I know if that is the issue?
If it is, then shall I cherry-picl this restructuring commit into master?
https://github.com/boostorg/gil/commit/4ed7701b47764edf327e7c82f65cf9815026bdc2


[1] https://lists.boost.org/Archives/boost/2019/07/246658.php
[2] https://github.com/boostorg/gil/issues/351

Best regards
--
Mateusz Loskot, http://mateusz.loskot.net

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

Re: [release] Boost 1.71.0 Release Candidate 1

Boost - Dev mailing list
In reply to this post by Boost - Dev mailing list
On Fri, Aug 9, 2019 at 1:43 PM Michael Caisse via Boost <
[hidden email]> wrote:

> The first release candidates for the 1.71.0 release are now available at:
>
> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
>
> The SHA256 checksums are as follows:
>
> 7fadf835372325d7a287b1c45310c4f96811eb1e4809933a5a400d529df38d0e
> boost_1_71_0_rc1.7z
>
> c857a86fb1223c7a256414151e2d1dcd4691b56197aa1329e087e92502191434
> boost_1_71_0_rc1.tar.bz2
>
> 07c07cad079fed3f0175cbd7097740a333c2cdc6c899ac433dea87d1224d726d
> boost_1_71_0_rc1.tar.gz
>
> bc1674ba3e11635fc2effbddc2ed69aecaced3901f81a6a2aae7b77f34e4cf76
> boost_1_71_0_rc1.zip
>
> As always, the release managers would appreciate it if you download the
> candidate of your choice and give building it a try. Please report both
> success and failure, and anything else that is noteworthy.
>
>
Windows/Visual Studio (2010 - 2019) builds look good. Full build logs can
be found here:
https://gist.github.com/teeks99/12c21a6ea7c7d2e06b53cc9aaa8185e7

toolset     arch    compile Link    Execute
msvc-10.0   32      X       X       X
msvc-10.0   64      X       X       X
msvc-11.0   32      X       X       X
msvc-11.0   64      X       X       X
msvc-12.0   32      X       X       X
msvc-12.0   64      X       X       X
msvc-14.0   32      X       X       X
msvc-14.0   64      X       X       X
msvc-14.1   32      X       X       X
msvc-14.1   64      X       X       X
msvc-14.2   32      X       X       X
msvc-14.2   64      X       X       X

Compile means that the b2 command completed without errors
Link means that visual studio was able to link a sample executable to a
library (libboost_thread-vcXXX-mt[-gd]-1_XX.lib) generated
Execute means that the linked program executed without errors.

Tom

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

Re: [release] Boost 1.71.0 Release Candidate 1

Boost - Dev mailing list
In reply to this post by Boost - Dev mailing list
On Fri, Aug 9, 2019 at 11:44 AM Michael Caisse via Boost <
[hidden email]> wrote:

> The first release candidates for the 1.71.0 release are now available at:
>
> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
>
> The SHA256 checksums are as follows:
>
> 7fadf835372325d7a287b1c45310c4f96811eb1e4809933a5a400d529df38d0e
> boost_1_71_0_rc1.7z
>
> c857a86fb1223c7a256414151e2d1dcd4691b56197aa1329e087e92502191434
> boost_1_71_0_rc1.tar.bz2
>
> 07c07cad079fed3f0175cbd7097740a333c2cdc6c899ac433dea87d1224d726d
> boost_1_71_0_rc1.tar.gz
>
> bc1674ba3e11635fc2effbddc2ed69aecaced3901f81a6a2aae7b77f34e4cf76
> boost_1_71_0_rc1.zip
>
> As always, the release managers would appreciate it if you download the
> candidate of your choice and give building it a try. Please report both
> success and failure, and anything else that is noteworthy.
>

I was able to build the libraries successfully with the following
configurations:

Mac OS X 10.14.6, apple-clang 10.01: c++03/11/14/17/2a
Mac OS X 10.14.6, clang-trunk:            c++03/11/14/17/2a
Ubuntu 18.14, gcc 8.3.2:                       c++03/11/14/1z
Ubuntu 18.14, clang-trunk:                    c++03/11/14/17/2a

I should have some more test results later tonight, but the following
libraries failed
many tests on "Mac OS X 10.14.6, apple-clang 10.01: c++03":
    Boost.VMD
    Boost.numeric/ublas (lots of constexpr, and other non-c++03 stuff)
    Boost.Process - lots of non-c++03 stuff

-- Marshall

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

Re: [release] Boost 1.71.0 Release Candidate 1

Boost - Dev mailing list
On 8/9/2019 10:10 PM, Marshall Clow via Boost wrote:

> On Fri, Aug 9, 2019 at 11:44 AM Michael Caisse via Boost <
> [hidden email]> wrote:
>
>> The first release candidates for the 1.71.0 release are now available at:
>>
>> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
>>
>> The SHA256 checksums are as follows:
>>
>> 7fadf835372325d7a287b1c45310c4f96811eb1e4809933a5a400d529df38d0e
>> boost_1_71_0_rc1.7z
>>
>> c857a86fb1223c7a256414151e2d1dcd4691b56197aa1329e087e92502191434
>> boost_1_71_0_rc1.tar.bz2
>>
>> 07c07cad079fed3f0175cbd7097740a333c2cdc6c899ac433dea87d1224d726d
>> boost_1_71_0_rc1.tar.gz
>>
>> bc1674ba3e11635fc2effbddc2ed69aecaced3901f81a6a2aae7b77f34e4cf76
>> boost_1_71_0_rc1.zip
>>
>> As always, the release managers would appreciate it if you download the
>> candidate of your choice and give building it a try. Please report both
>> success and failure, and anything else that is noteworthy.
>>
>
> I was able to build the libraries successfully with the following
> configurations:
>
> Mac OS X 10.14.6, apple-clang 10.01: c++03/11/14/17/2a
> Mac OS X 10.14.6, clang-trunk:            c++03/11/14/17/2a
> Ubuntu 18.14, gcc 8.3.2:                       c++03/11/14/1z
> Ubuntu 18.14, clang-trunk:                    c++03/11/14/17/2a
>
> I should have some more test results later tonight, but the following
> libraries failed
> many tests on "Mac OS X 10.14.6, apple-clang 10.01: c++03":
>      Boost.VMD
>      Boost.numeric/ublas (lots of constexpr, and other non-c++03 stuff)
>      Boost.Process - lots of non-c++03 stuff

VMD needs variadic macro support, which for a very few compiler
implementations means C++11 or higher, or else it fails a test when
compiled at the C++03 level.


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

Re: [release] Boost 1.71.0 Release Candidate 1

Boost - Dev mailing list
On Sat, Aug 10, 2019 at 4:05 AM Edward Diener via Boost <
[hidden email]> wrote:

> On 8/9/2019 10:10 PM, Marshall Clow via Boost wrote:
> > On Fri, Aug 9, 2019 at 11:44 AM Michael Caisse via Boost <
> > [hidden email]> wrote:
> >
> >> The first release candidates for the 1.71.0 release are now available
> at:
> >>
> >> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
> >>
> >> The SHA256 checksums are as follows:
> >>
> >> 7fadf835372325d7a287b1c45310c4f96811eb1e4809933a5a400d529df38d0e
> >> boost_1_71_0_rc1.7z
> >>
> >> c857a86fb1223c7a256414151e2d1dcd4691b56197aa1329e087e92502191434
> >> boost_1_71_0_rc1.tar.bz2
> >>
> >> 07c07cad079fed3f0175cbd7097740a333c2cdc6c899ac433dea87d1224d726d
> >> boost_1_71_0_rc1.tar.gz
> >>
> >> bc1674ba3e11635fc2effbddc2ed69aecaced3901f81a6a2aae7b77f34e4cf76
> >> boost_1_71_0_rc1.zip
> >>
> >> As always, the release managers would appreciate it if you download the
> >> candidate of your choice and give building it a try. Please report both
> >> success and failure, and anything else that is noteworthy.
> >>
> >
> > I was able to build the libraries successfully with the following
> > configurations:
> >
> > Mac OS X 10.14.6, apple-clang 10.01: c++03/11/14/17/2a
> > Mac OS X 10.14.6, clang-trunk:            c++03/11/14/17/2a
> > Ubuntu 18.14, gcc 8.3.2:                       c++03/11/14/1z
> > Ubuntu 18.14, clang-trunk:                    c++03/11/14/17/2a
> >
> > I should have some more test results later tonight, but the following
> > libraries failed
> > many tests on "Mac OS X 10.14.6, apple-clang 10.01: c++03":
> >      Boost.VMD
> >      Boost.numeric/ublas (lots of constexpr, and other non-c++03 stuff)
> >      Boost.Process - lots of non-c++03 stuff
>
> VMD needs variadic macro support, which for a very few compiler
> implementations means C++11 or higher, or else it fails a test when
> compiled at the C++03 level.
>

Shouldn't those tests be marked as requiring C++11, then?

-- Marshall

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

Re: [release] Boost 1.71.0 Release Candidate 1

Boost - Dev mailing list
On 8/10/2019 10:35 AM, Marshall Clow via Boost wrote:

> On Sat, Aug 10, 2019 at 4:05 AM Edward Diener via Boost <
> [hidden email]> wrote:
>
>> On 8/9/2019 10:10 PM, Marshall Clow via Boost wrote:
>>> On Fri, Aug 9, 2019 at 11:44 AM Michael Caisse via Boost <
>>> [hidden email]> wrote:
>>>
>>>> The first release candidates for the 1.71.0 release are now available
>> at:
>>>>
>>>> <https://dl.bintray.com/boostorg/release/1.71.0.rc1/source/>
>>>>
>>>> The SHA256 checksums are as follows:
>>>>
>>>> 7fadf835372325d7a287b1c45310c4f96811eb1e4809933a5a400d529df38d0e
>>>> boost_1_71_0_rc1.7z
>>>>
>>>> c857a86fb1223c7a256414151e2d1dcd4691b56197aa1329e087e92502191434
>>>> boost_1_71_0_rc1.tar.bz2
>>>>
>>>> 07c07cad079fed3f0175cbd7097740a333c2cdc6c899ac433dea87d1224d726d
>>>> boost_1_71_0_rc1.tar.gz
>>>>
>>>> bc1674ba3e11635fc2effbddc2ed69aecaced3901f81a6a2aae7b77f34e4cf76
>>>> boost_1_71_0_rc1.zip
>>>>
>>>> As always, the release managers would appreciate it if you download the
>>>> candidate of your choice and give building it a try. Please report both
>>>> success and failure, and anything else that is noteworthy.
>>>>
>>>
>>> I was able to build the libraries successfully with the following
>>> configurations:
>>>
>>> Mac OS X 10.14.6, apple-clang 10.01: c++03/11/14/17/2a
>>> Mac OS X 10.14.6, clang-trunk:            c++03/11/14/17/2a
>>> Ubuntu 18.14, gcc 8.3.2:                       c++03/11/14/1z
>>> Ubuntu 18.14, clang-trunk:                    c++03/11/14/17/2a
>>>
>>> I should have some more test results later tonight, but the following
>>> libraries failed
>>> many tests on "Mac OS X 10.14.6, apple-clang 10.01: c++03":
>>>       Boost.VMD
>>>       Boost.numeric/ublas (lots of constexpr, and other non-c++03 stuff)
>>>       Boost.Process - lots of non-c++03 stuff
>>
>> VMD needs variadic macro support, which for a very few compiler
>> implementations means C++11 or higher, or else it fails a test when
>> compiled at the C++03 level.
>>
>
> Shouldn't those tests be marked as requiring C++11, then?

Most compilers supported variadic macros before the C++11 standard was
created. Therefore the tests should not be marked as requiring C++11 any
more than the library itself requires C++11. The actual code determining
variadic macro support for a compiler implementation, from Paul
Mensonides and tweaked only slightly by me as the maintainer, is in the
Boost PP config.hpp header file, and is pretty conservative. If you find
a compiler/version which supports variadic macros, but which Boost
PP/VMD configures as not supporting variadic macros please create a
Boost PP or Boost VMD issue or PR.


_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost