[borland] Which version of bcc32 and of Boost do you use?
Hallo, I'd be interested in knowing which version of Boost do people use with which version of the Borland/CodeGear compilers. The reason I'm asking relates to the state of Boost regression tests for the version currently being developed, as can be seen here: http://beta.boost.org/development/tests/trunk/developer/summary.html As you can see the situation is not very good. Moreover, while I still have to apply the latest bcbboost patches for 5.9.2, 5.6.4 and 5.8.2 should already have the same level of support they had in Boost 1.34 . Unless there's a substantial demand it looks highly likely that support for at least BCB6 will have to be dropped or, more precisely, be left to rot. Contributions are an even more effective way to avoid it :-) Cheers, Nicola -- Nicola.Musatti <at> gmail <dot> com Home: http://nicola.musatti.googlepages.com/home Blog: http://wthwdik.wordpress.com/
Nicola Musatti <Nicola.Musatti <at> gmail.com> writes:
Hallo, I'd be interested in knowing which version of Boost do people use with which version of the Borland/CodeGear compilers. The reason I'm asking relates to the state of Boost regression tests for the version currently being developed, as can be seen here: http://beta.boost.org/development/tests/trunk/developer/summary.html
We're stuck at 1_30_0 with BCB6. Did try moving to 1_32_0 but the (admittedly small) effort of maintaining 1_32_0 with spirit 1_6 was deemed unnecessarily error-prone in a multideveloper environment. Similarly, we don't use any non-header-only libraries; the configuration pain is not worth the effort when we have a fair code library of our own which has some overlap with boost. Since we've got to change compiler anyway we're moving to C# for the bulk of our work.
Simon Carter wrote:
Nicola Musatti <Nicola.Musatti <at> gmail.com> writes:
Hallo, I'd be interested in knowing which version of Boost do people use with which version of the Borland/CodeGear compilers. The reason I'm asking relates to the state of Boost regression tests for the version currently being developed, as can be seen here: http://beta.boost.org/development/tests/trunk/developer/summary.html
We're stuck at 1_30_0 with BCB6. Did try moving to 1_32_0 but the (admittedly small) effort of maintaining 1_32_0 with spirit 1_6 was deemed unnecessarily error-prone in a multideveloper environment. Similarly, we don't use any non-header-only libraries; the configuration pain is not worth the effort when we have a fair code library of our own which has some overlap with boost.
Since we've got to change compiler anyway we're moving to C# for the bulk of our work.
I am always amused when a C++ programmer, who is going to be using .Net, says they have to change compilers and then mentions they will be using C#. Are you aware that a C++ implementation for .Net already exists called C++/CLI and that you can directly used Boost code using C++/CLI in mixed mode programming under .Net ?
From: Edward Diener
Simon Carter wrote:
We're stuck at 1_30_0 with BCB6.
Since we've got to change compiler anyway we're moving to C# for the bulk of our work.
I am always amused when a C++ programmer, who is going to be using .Net, says they have to change compilers and then mentions they will be using C#. Are you aware that a C++ implementation for .Net already exists called C++/CLI and that you can directly used Boost code using C++/CLI in mixed mode programming under .Net ?
For most programmers who are moving to .Net, you are correct. However in this particular case, Simon would /still/ have to change compiler - from BCB6 to VS2005. Given that, it is not unreasonable to decide to go the whole hog and go to C#. (It's not what I would do, but ...)
participants (4)
-
Edward Diener -
Martin Bonner -
Nicola Musatti -
Simon Carter