Another jdk9 patch :)

[Jiri Vanek]

On 01/13/2016 01:36 PM, Fridrich Strba wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On 13/01/16 13:24, Jiri Vanek wrote:
>> Have it really to be sun.applet? Is new modular system preventing
>> to have it in different package?
>
> I just kidnapped the previous macro that was testing for a generic

Sure, I just saw the notes around the rm- rf a bit lower, so I thought its worthy to try in defualt 
package.

> class. But since it does not have a behaviour in case that class is
> not found, but just bombs with AC_MSG_ERROR, I modified it to look for
> the first location first and if not found for the second
>
>> Once this class is not found What is output of the macro (when used
>> by configure). Simple "no" or the stacktrace?
>
> The Class.forName("sun.security.util.HexDumpEncoder")) simply throws
> again the ClassNotFound exception and the result will be no and it
> will stop the configure with the AC_MSG_ERROR.

good.

>
>>> -import sun.misc.HexDumpEncoder;
>> I would put comment there
>>> +import sun.misc.*;
>> and there, expalining why "*" is used. Most IDes is marking it as
>> warning, so there is risk of evaluating thsoe stars and so
>> reintroduce the bug.
>>
>>> +import sun.security.util.*;
>
> Yeah, let me comment there. It was basically the easiest way to handle
> that without having to know in the particular java file which full
> name the HexDumpEncoder has. I will comment about that.

Yes, thats what I have guessed. Nice workaround :)
>
> Cheers
>
> Fridrich
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v2
>
> iEYEARECAAYFAlaWROAACgkQu9a1imXPdA+5KQCfdfBbP6Z52KyYMtAB5TGQCc3Y
> ubAAnj4TyQ5e81ZyX7+McRon4CbD6thP
> =7CBO
> -----END PGP SIGNATURE-----
>