On 19-Mar-09, at 8:20 PM, Christian Biere wrote: > Matthew Lye wrote: >> There's an oracle? > > Maybe. Just google it. > >> The linker problem should only be a nuisance for people who've >> enabled >> the bug reporting system included in the Xcode bundle. There's >> probably a way to disable it per session, somewhere off in >> Onlinedokumentistan; I'll let you know if it turns up. > > I take it the modifications to Configure didn't make any difference?
No, sorry. The abort-signaling errors seem to occur only when a function is assumed to be being implicitly declared. That is, when 'invent.h' is not found the linker fails gracefully even though getinvent() is being implicitly (of course) declared. Similarly, the test for posix_fadvise() fails gracefully because none of the identifiers (such as POSIX_FADV_NORMAL) in the code are declared, an obvious error. However, the test for a function NOT declared in an existing header file fails ungracefully, producing an abort signal. There is a cc flag in the Darwin version of gcc, '-Werror-implicit- function-declaration', which causes all implicit declarations to be treated as errors rather than warnings. Unfortunately one must then include 'stdlib.h' to use exit, but this seems to only be once, in 'try.c'. Treating implicit function declarations as errors seems to solve every incidence of non-graceful termination, pre-empting it with graceful termination, except for one. When the use of the locale_charset() function left undeclared by the (existing) header file 'libcharset.h' is tested, the implicit declaration goes unnoticed. I'm not sure why. ------------------------------------------------------------------------------ Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are powering Web 2.0 with engaging, cross-platform capabilities. Quickly and easily build your RIAs with Flex Builder, the Eclipse(TM)based development software that enables intelligent coding and step-through debugging. Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com _______________________________________________ gtk-gnutella-devel mailing list gtk-gnutella-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/gtk-gnutella-devel
