[detail] boost/detail/utf8_codecvt_facet.hpp broken by recent change
boost/detail/utf8_codecvt_facet.hpp fails to compile for STL's MinGW distro from https://nuwen.net/mingw.html The error message is: ..\..\../boost/detail/utf8_codecvt_facet.hpp:117:29: error: function 'boost::filesystem::detail::utf8_codecvt_facet::~utf8_codecvt_facet()' definition is marked dllimport virtual BOOST_UTF8_DECL ~utf8_codecvt_facet(){} The problem (on develop) is caused by Robert Ramey's Aug 13 commit 73c12230c31a851c15ecb9a649468425ebd92f8e "adjustment to avoid linking problem: vtable not found found for ~utf8_codecvt_facet.hpp" which added "BOOST_UTF8_DECL" to line 117. This is breaking my local filesystem tests for GCC/MinGW. If the change is reverted, the tests start passing again. Robert, could you please revert the change? I can also to it, but thought you needed to be alerted first so you can revisit whatever you were trying to fix. Thanks, --Beman
On 8/25/17 6:09 AM, Beman Dawes via Boost wrote:
boost/detail/utf8_codecvt_facet.hpp fails to compile for STL's MinGW distro from https://nuwen.net/mingw.html
The error message is:
..\..\../boost/detail/utf8_codecvt_facet.hpp:117:29: error: function 'boost::filesystem::detail::utf8_codecvt_facet::~utf8_codecvt_facet()' definition is marked dllimport virtual BOOST_UTF8_DECL ~utf8_codecvt_facet(){}
The problem (on develop) is caused by Robert Ramey's Aug 13 commit 73c12230c31a851c15ecb9a649468425ebd92f8e "adjustment to avoid linking problem: vtable not found found for ~utf8_codecvt_facet.hpp" which added "BOOST_UTF8_DECL" to line 117.
This is breaking my local filesystem tests for GCC/MinGW. If the change is reverted, the tests start passing again.
Robert, could you please revert the change? I can also to it, but thought you needed to be alerted first so you can revisit whatever you were trying to fix.
OK I see this. What I don't see is why it fails which I think is important. I'd like a couple of days to try to understand why this is happening. Robert Ramey
Thanks,
--Beman
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
On 8/25/17 9:18 AM, Robert Ramey via Boost wrote:
On 8/25/17 6:09 AM, Beman Dawes via Boost wrote:
boost/detail/utf8_codecvt_facet.hpp fails to compile for STL's MinGW distro from https://nuwen.net/mingw.html
The error message is:
..\..\../boost/detail/utf8_codecvt_facet.hpp:117:29: error: function 'boost::filesystem::detail::utf8_codecvt_facet::~utf8_codecvt_facet()' definition is marked dllimport virtual BOOST_UTF8_DECL ~utf8_codecvt_facet(){}
The problem (on develop) is caused by Robert Ramey's Aug 13 commit 73c12230c31a851c15ecb9a649468425ebd92f8e "adjustment to avoid linking problem: vtable not found found for ~utf8_codecvt_facet.hpp" which added "BOOST_UTF8_DECL" to line 117.
This is breaking my local filesystem tests for GCC/MinGW. If the change is reverted, the tests start passing again.
Robert, could you please revert the change? I can also to it, but thought you needed to be alerted first so you can revisit whatever you were trying to fix.
OK I see this. What I don't see is why it fails which I think is important. I'd like a couple of days to try to understand why this is happening.
OK - I took a short look at this. It's clear that I can't figure it out in a reasonable time so the best would be to back out the change. Feel free to do it. Robert Ramey
On Fri, Aug 25, 2017 at 12:39 PM, Robert Ramey via Boost < boost@lists.boost.org> wrote:
On 8/25/17 9:18 AM, Robert Ramey via Boost wrote:
On 8/25/17 6:09 AM, Beman Dawes via Boost wrote:
boost/detail/utf8_codecvt_facet.hpp fails to compile for STL's MinGW distro from https://nuwen.net/mingw.html
The error message is:
..\..\../boost/detail/utf8_codecvt_facet.hpp:117:29: error: function 'boost::filesystem::detail::utf8_codecvt_facet::~utf8_codecvt_facet()' definition is marked dllimport virtual BOOST_UTF8_DECL ~utf8_codecvt_facet(){}
The problem (on develop) is caused by Robert Ramey's Aug 13 commit 73c12230c31a851c15ecb9a649468425ebd92f8e "adjustment to avoid linking problem: vtable not found found for ~utf8_codecvt_facet.hpp" which added "BOOST_UTF8_DECL" to line 117.
This is breaking my local filesystem tests for GCC/MinGW. If the change is reverted, the tests start passing again.
Robert, could you please revert the change? I can also to it, but thought you needed to be alerted first so you can revisit whatever you were trying to fix.
OK I see this. What I don't see is why it fails which I think is important. I'd like a couple of days to try to understand why this is happening.
OK - I took a short look at this. It's clear that I can't figure it out in a reasonable time so the best would be to back out the change. Feel free to do it. http://lists.boost.org/mailman/listinfo.cgi/boost
Done. --Beman
On 8/28/2017 8:03 AM, Beman Dawes via Boost wrote:
On Fri, Aug 25, 2017 at 12:39 PM, Robert Ramey via Boost < boost@lists.boost.org> wrote:
On 8/25/17 9:18 AM, Robert Ramey via Boost wrote:
On 8/25/17 6:09 AM, Beman Dawes via Boost wrote:
boost/detail/utf8_codecvt_facet.hpp fails to compile for STL's MinGW distro from https://nuwen.net/mingw.html
The error message is:
..\..\../boost/detail/utf8_codecvt_facet.hpp:117:29: error: function 'boost::filesystem::detail::utf8_codecvt_facet::~utf8_codecvt_facet()' definition is marked dllimport virtual BOOST_UTF8_DECL ~utf8_codecvt_facet(){}
The problem (on develop) is caused by Robert Ramey's Aug 13 commit 73c12230c31a851c15ecb9a649468425ebd92f8e "adjustment to avoid linking problem: vtable not found found for ~utf8_codecvt_facet.hpp" which added "BOOST_UTF8_DECL" to line 117.
This is breaking my local filesystem tests for GCC/MinGW. If the change is reverted, the tests start passing again.
Robert, could you please revert the change? I can also to it, but thought you needed to be alerted first so you can revisit whatever you were trying to fix.
OK I see this. What I don't see is why it fails which I think is important. I'd like a couple of days to try to understand why this is happening.
OK - I took a short look at this. It's clear that I can't figure it out in a reasonable time so the best would be to back out the change. Feel free to do it. http://lists.boost.org/mailman/listinfo.cgi/boost
Done.
I do not see Robert's change reverted on 'develop'. I am also seeing the same problem you are.
--Beman
On 08/28/17 22:16, Edward Diener via Boost wrote:
On 8/28/2017 8:03 AM, Beman Dawes via Boost wrote:
On Fri, Aug 25, 2017 at 12:39 PM, Robert Ramey via Boost < boost@lists.boost.org> wrote:
OK - I took a short look at this. It's clear that I can't figure it out in a reasonable time so the best would be to back out the change. Feel free to do it. http://lists.boost.org/mailman/listinfo.cgi/boost
Done.
I do not see Robert's change reverted on 'develop'. I am also seeing the same problem you are.
It looks reverted to me: https://github.com/boostorg/detail/commit/782208b04ded0b01733c40dd3af05b81f4...
On 8/28/2017 4:16 PM, Andrey Semashev via Boost wrote:
On 08/28/17 22:16, Edward Diener via Boost wrote:
On 8/28/2017 8:03 AM, Beman Dawes via Boost wrote:
On Fri, Aug 25, 2017 at 12:39 PM, Robert Ramey via Boost < boost@lists.boost.org> wrote:
OK - I took a short look at this. It's clear that I can't figure it out in a reasonable time so the best would be to back out the change. Feel free to do it. http://lists.boost.org/mailman/listinfo.cgi/boost
Done.
I do not see Robert's change reverted on 'develop'. I am also seeing the same problem you are.
It looks reverted to me:
https://github.com/boostorg/detail/commit/782208b04ded0b01733c40dd3af05b81f4...
You are right. I thought the change had been made in serialization.
participants (4)
-
Andrey Semashev -
Beman Dawes -
Edward Diener -
Robert Ramey