In article <u64q4872w.fsf@boost-consulting.com>, David Abrahams <dave@boost-consulting.com> wrote:
Ben Artin <macdev@artins.org> writes:
I think that this particular nit is hardly worth the amount of discussion it's generated
Exactly.
I am happy to drop this, but I will give you my opinion, since I assume that you care enough to hear it once, and then just want everyone to move on to more important matters.
which is why Daryle and I believe that the change should be made now to make subsequent modifications easier.
Do you have any idea what code you're asking to have changed?
Yes.
Do you realize that the code being argued over was never checked in?
Yes.
The current test is (boost/python/detail/config.hpp):
# if !defined(_WIN32) && !defined(__CYGWIN__) \ && !defined(BOOST_PYTHON_USE_GCC_SYMBOL_VISIBILITY)
\ && BOOST_WORKAROUND(__GNUC__, >= 3) && (__GNUC_MINOR__ >=5 || __GNUC__ > 3) # define BOOST_PYTHON_USE_GCC_SYMBOL_VISIBILITY 1 # endif
Do you have a problem with that?
I do; that is *really* hard to read, IMO. It's clever, sure, but it's hard to read.
I can imagine that the following would be a very minor improvement, but it hardly seems to be worth the bits we're spending on it.
# if !defined(_WIN32) && !defined(__CYGWIN__) \ && !defined(BOOST_PYTHON_USE_GCC_SYMBOL_VISIBILITY)
\ && (BOOST_WORKAROUND(__GNUC__, == 3) && __GNUC_MINOR__ >=5 \ || BOOST_WORKAROUND(__GNUC__, > 3)) # define BOOST_PYTHON_USE_GCC_SYMBOL_VISIBILITY 1 # endif
It would be an improvement, it would be minor, I wouldn't have brought it up as I think it's a pretty small fish to fry, but Daryle did, and now that he has, my opinion is that he does have a good point and that the right thing is to make this code easier to read and (possibly) maintain. You are the maintainer, the final decision is yours, and I am going to turn my attention back to bigger things in my life. Ben -- I changed my name: <http://periodic-kingdom.org/People/NameChange.php>