-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Richi Plana wrote: > Would it make sense to address these individual packages and convert all > gethostby*(3) function calls to getaddrinfo(3)? That's the idea. > One approach I can think of is to go ahead and make these patches, apply > them to Fedora packages and submit them upstream to their corresponding > maintainers. If it gets applied upstream, the package maintainers can > rebase and drop the patch. If upstream says "no", then the patches can > be dropped and the software reverted. There is no reason to drop them. Using the gethostby* functions is simply wrong. Just like any other bug fix the patch should be carried until upstream takes it in one form or another. > I'm just worrying about the different platforms these software are > expected to run on. According to gethostbyname(3), the function conforms > to POSIX as well as 4.3BSD so certain platforms might not have support > for getaddrinfo(). What's the proper way of addressing these? > Preprocessor directives? autoconf - -- ➧ Ulrich Drepper ➧ Red Hat, Inc. ➧ 444 Castro St ➧ Mountain View, CA ❖ -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) iD8DBQFHEFC22ijCOnn/RHQRAmhfAJ0ePG3sA7SC/OIFN2UYn2Y025n2WwCgpOgY xgS/O6RhWDhi4hlXUVMCszI= =HcQd -----END PGP SIGNATURE----- -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
