Boost Version 1.47 not building with 2017.
Hello, 1.47 boost is not being built with VS2017. We want to get boost 1.47 built with VS2017, So that we can integrate them with our VS2017 code base. Please guide us, it will be a great help. Thanks, Sarath.
On 28/06/2019 17:30, nunna, sarath babu (GE Healthcare) wrote:
1.47 boost is not being built with VS2017. We want to get boost 1.47 built with VS2017, So that we can integrate them with our VS2017 code base.
You already asked that. The answer has not changed. Boost 1.47 was released in 2011. VS2017 was not released yet. It is not surprising that these do not work together out of the box, especially as VS2017 made breaking changes to their tool locations. Upgrade to a version of Boost that was released after VS2017, and you will have more luck.
On Fri, 28 Jun 2019 at 08:30, nunna, sarath babu (GE Healthcare) via Boost < boost@lists.boost.org> wrote:
1.47 boost is not being built with VS2017. We want to get boost 1.47 built with VS2017, So that we can integrate them with our VS2017 code base.
I would like to get a square peg in a round hole too! Like John Maddock already wrote you, Boost-1.47, released 11 July 2011, pre-dates VS2017, by almost 6 years [year with major upheaval in C++ standards and compiler development]. So there is no way to build a reliable Boost-1.47 with VS2017, released 7 March 2017. Maybe you are using a heavily modified [or certified] Boost and are therefor bound to that version [but then you should use a compiler that matched [and tested] that release (VS2010), because then the bugs (in Boost) will balance out the bugs in the compiler/libs], but otherwise, the best thing is to use the latest and the greatest. degski -- @realdegski https://edition.cnn.com/interactive/2019/06/middleeast/saudi-teen-death-pena... "Anyone who believes that exponential growth can go on forever in a finite world is either a madman or an economist" - Kenneth E. Boulding
On 6/27/19 10:30 PM, nunna, sarath babu (GE Healthcare) via Boost wrote:
Hello,
1.47 boost is not being built with VS2017. We want to get boost 1.47 built with VS2017, So that we can integrate them with our VS2017 code base.
Please guide us, it will be a great help. Note that C++ successfully strives to maintain backwards compatibility. The Boost serialization library is maintained such that it will compile and execute correctly on any C++ compiler which supports C++03 or later. So the behavior you describe would be a bug.
Currently there are no known bugs of nature in the current version of the library. So you should gather your information is a way which would isolate such a bug and report as an issue on Github. Robert Ramey
Thanks, Sarath.
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On Fri, 28 Jun 2019 at 22:48, Robert Ramey via Boost
Note that C++ successfully strives to maintain backwards compatibility.
But OP is talking about forward compatibility [he want to compile Boost-1.47 with VS2017, there is a six year gap, the wrong way], which obviously is never gonna happen. degski -- @realdegski https://edition.cnn.com/interactive/2019/06/middleeast/saudi-teen-death-pena... "Anyone who believes that exponential growth can go on forever in a finite world is either a madman or an economist" - Kenneth E. Boulding
On 6/28/19 2:49 PM, degski via Boost wrote:
On Fri, 28 Jun 2019 at 22:48, Robert Ramey via Boost
wrote: Note that C++ successfully strives to maintain backwards compatibility.
But OP is talking about forward compatibility [he want to compile Boost-1.47 with VS2017, there is a six year gap, the wrong way], which obviously is never gonna happen.
Since C++17 is backward compatible with C++03 - I would be very very surprise if it failed to work. In fact I always test with std=C++03 flag to ensure exactly that.
degski
On Sat, 29 Jun 2019 at 08:41, Robert Ramey via Boost
Since C++17 is backward compatible with C++03
On thing I know, as a matter of fact (not guessing here), is the use of std::auto_ptr throughout that Boost-1.47 release. This thing alone is not a trivial fix. degski -- @realdegski https://edition.cnn.com/interactive/2019/06/middleeast/saudi-teen-death-pena... "Anyone who believes that exponential growth can go on forever in a finite world is either a madman or an economist" - Kenneth E. Boulding
On Sat, 29 Jun 2019 at 08:41, Robert Ramey via Boost
In fact I always test with std=C++03 flag to ensure exactly that.
Yes, you do, and that also fixes the C++17 issue of course, but you did that at the time with VS2010 (or even VS2008) and it succeeded. Boost-1.47 must include a ton of work-arounds to fix the non-compliance of both the pre-processor, the compiler and its STL. Compiling those with VS2017/19 will certainly break some of those (I **am** guessing here), coz the Boost-1.47 code has no knowledge of the now much better compliance of VS. degski -- @realdegski https://edition.cnn.com/interactive/2019/06/middleeast/saudi-teen-death-pena... "Anyone who believes that exponential growth can go on forever in a finite world is either a madman or an economist" - Kenneth E. Boulding
On 29/06/2019 17:41, Robert Ramey wrote:
Since C++17 is backward compatible with C++03 - I would be very very surprise if it failed to work. In fact I always test with std=C++03 flag to ensure exactly that.
We're not even talking about C++17, we're talking about VS2017. Quite apart from anything else already mentioned, the build scripts in Boost 1.47 don't know how to find the existence of the VS2017 compiler, since it didn't exist at the time and it uses a very different installation mechanism from prior versions of VS. (And the VS compiler build scripts were never particularly forward-compatible to begin with.) Since the OP didn't specify which part of the build was failing, let alone any specific error messages, it's just as likely for this to be the problem as any of the other issues. Ultimately, the already-given solution is straightforward: don't do that.
On Wed, 3 Jul 2019 at 04:07, Gavin Lambert via Boost
Ultimately, the already-given solution is straightforward: don't do that.
LOL, +1 degski -- @realdegski https://edition.cnn.com/interactive/2019/06/middleeast/saudi-teen-death-pena... "Anyone who believes that exponential growth can go on forever in a finite world is either a madman or an economist" - Kenneth E. Boulding
participants (4)
-
degski
-
Gavin Lambert
-
nunna, sarath babu (GE Healthcare)
-
Robert Ramey