At 17:36 Uhr -0500 02.02.2002, David R. Morrison wrote: >Two more comments about the new zlib > >1) I have only a vague recollection about the earlier discussion of this. >Was zlib originally absent from darwin/OS X, and added in a later revision? >If so, do you need some kind of darwin-version test for the new package?
It was added in 10.1. Yeah we should probably have a depends on the proper darwin/macosx version However, that is currently broken for binary packages. I dunno yet how to fix this. I.e. we have to "teach" dpkg and maybe also apt about these pseudo packages. One way would be to incorporate this simply direct into the source, since I am going to release a new dpkg anytime soon anyway, with some shlibs changes. You may wonder why I do not just fake to dpkg that a package macosx with the correct revision installed. Well, the reason is that a single fink folder might be shared between different systems, and the pseudo packages really should always reflect the current system version. >2) Since Apple is providing libz.h, you don't actually need it in the >new package either, do you? If you leave it out, your package becomes >very similar to one of the new-style library packages I am proposing. >It would, in principle, be accompanied by zlib-dev which would contain >libz.h and the missing symbolic link, but essentially we are letting >Apple provide that. True. > >3) In fact, if we do need a version dependence, you could write zlib-dev >which Conflicts with later versions of darwin/OS X. Then your zlib package >could have a field > BuildDepends: zlib-dev | darwin-version-recent > >(the foo-dev packages should never go in Depends, only in BuildDepends). Yup. Max -- ----------------------------------------------- Max Horn Software Developer email: <mailto:[EMAIL PROTECTED]> phone: (+49) 6151-494890 _______________________________________________ Fink-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/fink-devel
