Re: master 3fdc36e: Error out on --with-harfbuzz without HarfBuzz support

Re: master 3fdc36e: Error out on --with-harfbuzz without HarfBuzz support

[Andreas Schwab]

On Nov 17 2019, Lars Ingebrigtsen wrote:

> diff --git a/configure.ac b/configure.ac
> index 3b6a2a6..0976b66 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -3469,6 +3469,11 @@ if test "${HAVE_X11}" = "yes" && test "${HAVE_FREETYPE}" = "yes" \
>    fi
>  fi
>  
> +# The user explicitly asked for HarfBuzz, but it won't be used.
> +if test "${with_harfbuzz}" == "yes" && test "$HAVE_HARFBUZZ" != "yes"; then
> +   AC_MSG_ERROR([HarfBuzz wanted, but won't be used.  Maybe some library files are missing?]);
> +fi

That gives a spurious error when configuring with --without-x.

Also, test a == b is a bashism.  The string compare operator is =.

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510  2552 DF73 E780 A9DA AEC1
"And now for something completely different."

Re: master 3fdc36e: Error out on --with-harfbuzz without HarfBuzz support

[Lars Ingebrigtsen]

Andreas Schwab <schwab@linux-m68k.org> writes:

> That gives a spurious error when configuring with --without-x.

I've now removed the check.  I assumed that the with_harfbuzz variable
was only set to "yes" if the user had explicitly said "--with-harfbuzz",
but that is not true -- that's the default value of the variable?
Because:

OPTION_DEFAULT_ON([harfbuzz],[don't use HarfBuzz for text shaping])

Is there a way to check whether a user has explicitly said --with-foo?

> Also, test a == b is a bashism.  The string compare operator is =.

Ah, thanks.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no