filesystem.dll built with BOOST_ALL_DYN_LINK
Hello all, I just pulled boost from the CVS repository and built the filesystem library. bjam -d2 --with-filesystem This works and links great on my linux system. When I go over to windows with MSVC 7.1 and try to link I get stuck trying to autolink to the static lib. So It looks to me like the proper solution is to uncomment #define BOOST_ALL_DYN_LINK in the boost/config/user.hpp file. So I do that. Now we are trying to link to the .dll. Good. Only problem is I have two undefined symbols. (demangled) boost::filesystem::path::iterator::increment boost::filesytem::path::iterator::decrement So what did I do wrong? I can't believe I'm the first person to have this problem, but I couldn't find anything in the FAQ or mailing list archives. -- This e-mail and any attachments may contain confidential and privileged information. If you are not the intended recipient, please notify the sender immediately by return e-mail, delete this e-mail and destroy any copies. Any dissemination or use of this information by a person other than the intended recipient is unauthorized and may be illegal. Unless otherwise stated, opinions expressed in this e-mail are those of the author and are not endorsed by the author's employer.
I just pulled boost from the CVS repository and built the filesystem library.
bjam -d2 --with-filesystem
This works and links great on my linux system. When I go over to windows with MSVC 7.1 and try to link I get stuck trying to autolink to the static lib.
So It looks to me like the proper solution is to uncomment #define BOOST_ALL_DYN_LINK in the boost/config/user.hpp file. So I do that. Now we are trying to link to the .dll. Good. Only problem is I have two undefined symbols. (demangled) boost::filesystem::path::iterator::increment boost::filesytem::path::iterator::decrement
So what did I do wrong? I can't believe I'm the first person to have this problem, but I couldn't find anything in the FAQ or mailing list archives.
It looks like a bug, there's no export declaration on the path::iterator nested class, Beman? John.
"John Maddock"
I just pulled boost from the CVS repository and built the filesystem library.
bjam -d2 --with-filesystem
This works and links great on my linux system. When I go over to windows with MSVC 7.1 and try to link I get stuck trying to autolink to the static lib.
So It looks to me like the proper solution is to uncomment #define BOOST_ALL_DYN_LINK in the boost/config/user.hpp file. So I do that. Now we are trying to link to the .dll. Good. Only problem is I have two undefined symbols. (demangled) boost::filesystem::path::iterator::increment boost::filesytem::path::iterator::decrement
So what did I do wrong? I can't believe I'm the first person to have this problem, but I couldn't find anything in the FAQ or mailing list archives.
It looks like a bug, there's no export declaration on the path::iterator nested class, Beman?
I'll look at it tomorrow. Thanks, --Beman
"John Maddock"
I just pulled boost from the CVS repository and built the filesystem library.
bjam -d2 --with-filesystem
This works and links great on my linux system. When I go over to windows with MSVC 7.1 and try to link I get stuck trying to autolink to the static lib.
So It looks to me like the proper solution is to uncomment #define BOOST_ALL_DYN_LINK in the boost/config/user.hpp file. So I do that. Now we are trying to link to the .dll. Good. Only problem is I have two undefined symbols. (demangled) boost::filesystem::path::iterator::increment boost::filesytem::path::iterator::decrement
So what did I do wrong? I can't believe I'm the first person to have this problem, but I couldn't find anything in the FAQ or mailing list archives.
It looks like a bug, there's no export declaration on the path::iterator nested class, Beman?
Exactly. Fixed in CVS. The easy fix was to add BOOST_FILESYSTEM_DECL for the two functions, in both the header and implementation files. I'm adding two dll tests to detect regressions. The two added dll tests are working fine for Intel, Metrowerks, and Microsoft compilers, but are getting runtime failures for Borland and GCC. I'll try to figure out what is happening and get a fix together. Thanks for the bug report, --Beman
participants (3)
-
Beman Dawes -
John Maddock -
Larry E. Ramey