This is the mail archive of the mailing list for the glibc project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Switch IDNA implementation to libidn2 [BZ #19728] [BZ #19729] [BZ #22247]

On 02/02/2018 04:41 AM, Florian Weimer wrote:
> This is the patch rebased to 2.28.  It still depends on the
> allocate_once patch (which has been reviewed).
> I'd especially appreciate a review of the makefile/build system changes.

If the default is to build with an external libidn2 if possible,
shouldn't there be a configure switch to build --without-libidn2 (or
--en/disable-idn, etc.)?

I've only skimmed over the patch, and ran configure/make, which seems to
have gone fine (but that's without `make check', and I haven't reviewed
output in any detail, other than to make sure `grep idna' showed something).

> idna-2.28.patch


> diff --git a/manual/contrib.texi b/manual/contrib.texi
> index 0da6dcbb59..2be9122469 100644
> --- a/manual/contrib.texi
> +++ b/manual/contrib.texi
> @@ -178,7 +178,7 @@ software floating-point support and for
>  his direction as part of @theglibc{} steering committee.
>  @item
> -Simon Josefsson for the @code{libidn} add-on.
> +Simon Josefsson for the @code{libidn} add-on (which has been removed again).

Might it be kinder to say something like, "albeit eventually removed",
perhaps with, "...after N years"?

> diff --git a/resolv/tst-resolv-ai_idn-nolibidn2.c b/resolv/tst-resolv-ai_idn-nolibidn2.c
> new file mode 100644
> index 0000000000..8f30da8741
> --- /dev/null
> +++ b/resolv/tst-resolv-ai_idn-nolibidn2.c
> +/* Tests for getnameinfo.  */
> +static void
> +gni_tests (void)
> +{
> +  /* All non-IDN an IDN results are the same due to lack of libidn2



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]