sorry for the slow replies, but i was down for a couple of days. the compatibility changes for windows are far from being finished. the NS_EXPORT situation is a mess, not being able to write to the log file in an error situation is not really an option.
Rather than commenting out error messages, the better option is probably to move the ns.h include to the end and add warning lines, since reentrant.c contains no NS_EXTERN etc. i would not be surprised, if the same problem shows under windows as well in some modules. -g Am 01.10.14 21:12, schrieb Andrew Piskorski: > On Wed, Sep 17, 2014 at 06:50:08PM -0400, Andrew Piskorski wrote: > >> include/ns.h has this: >> >> #define NS_EXTERN extern NS_EXPORT >> >> But I believe that is NOT relevent here because master.c, thread.h, >> and nsthread.h (correctly) do NOT include ns.h. >> >> Instead, I think it is the separate definition of NS_EXTERN in >> nsthread.h that matters when building nsthread. > Gustaf, on 9/27 you changed nsthread/reentrant.c to import ns.h. It > never did that before, and now it imports ns.h before thread.h and > nsthread.h. It looks like you did that in order to call Ns_Log in > reentrant.c. > > However, I suspect that ALSO changed the definition of NS_EXTERN to > the one in ns.h rather than nsthread.h. Was that an unintentional? > Defining NS_EXTERN in two separate places like that is kind of > confusing. Can/should we merge them into one somehow? > > I bring this up because after merging all your latest changes into my > fork, the nsthread Windows build (which worked before) now fails like > this: > > Creating library nsthread.lib and object nsthread.exp > reentrant.o : error LNK2019: unresolved external symbol Ns_Log referenced > in function ns_asctime > nsthread.dll : fatal error LNK1120: 1 unresolved externals > > > changeset: 2884:950a8cebabb8 > user: Gustaf Neumann <neum...@wu-wien.ac.at> > date: Sat Sep 27 22:17:55 2014 +0200 > files: include/nsthread.h nsd/exec.c nsthread/reentrant.c > description: > - setting WINVERSION to Vista to allow usage of InetNtop() > - make code more robust against previous defines > - use the right pid_t type for win > - add shutdown flags > - use *_s function in win-specific reentrant code > -- Univ.Prof. Dr. Gustaf Neumann WU Vienna Institute of Information Systems and New Media Welthandelsplatz 1, A-1020 Vienna, Austria ------------------------------------------------------------------------------ Meet PCI DSS 3.0 Compliance Requirements with EventLog Analyzer Achieve PCI DSS 3.0 Compliant Status with Out-of-the-box PCI DSS Reports Are you Audit-Ready for PCI DSS 3.0 Compliance? Download White paper Comply to PCI DSS 3.0 Requirement 10 and 11.5 with EventLog Analyzer http://pubads.g.doubleclick.net/gampad/clk?id=154622311&iu=/4140/ostg.clktrk _______________________________________________ naviserver-devel mailing list naviserver-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/naviserver-devel
