You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@groovy.apache.org by Paul King <pa...@asert.com.au> on 2018/08/12 13:34:06 UTC
ModuleNode and CompuleUnit
Anyone think of a reason we shouldn't allow the CompileUnit to be accessed
from ModuleNode? We already let it be accessed from ClassNode. I was only
thinking read access.
I was contemplating adding NodeMetaDataHandler capability to CompileUnit
for "global" metadata shared across transforms. Thoughts?
Cheers, Paul.
Re: ModuleNode and CompuleUnit
Posted by Paul King <pa...@asert.com.au>.
False alarm. We already allow compile unit to be accessed, it's just called
getUnit in ModuleNode and getCompileUnit in ClassNode.
On Sun, Aug 12, 2018 at 11:34 PM Paul King <pa...@asert.com.au> wrote:
>
> Anyone think of a reason we shouldn't allow the CompileUnit to be accessed
> from ModuleNode? We already let it be accessed from ClassNode. I was only
> thinking read access.
>
> I was contemplating adding NodeMetaDataHandler capability to CompileUnit
> for "global" metadata shared across transforms. Thoughts?
>
> Cheers, Paul.
>
>
Re: ModuleNode and CompuleUnit
Posted by mg <mg...@arscreat.com>.
Alias + deprecate in 2.5.3, delete in 2.5.4 - we need to get up to JDK speed here, guys (& gals)... ;-)
-------- Ursprüngliche Nachricht --------Von: Paul King <pa...@asert.com.au> Datum: 13.08.18 10:56 (GMT+00:00) An: dev@groovy.apache.org Betreff: Re: ModuleNode and CompuleUnit
We could provide an alias I guess.
On Mon, Aug 13, 2018 at 6:09 PM Jochen Theodorou <bl...@gmx.org> wrote:
Am 13.08.2018 um 10:08 schrieb Jochen Theodorou:
>
>
> Am 12.08.2018 um 15:34 schrieb Paul King:
>>
>> Anyone think of a reason we shouldn't allow the CompileUnit to be
>> accessed from ModuleNode? We already let it be accessed from
>> ClassNode. I was only thinking read access.
>>
>> I was contemplating adding NodeMetaDataHandler capability to
>> CompileUnit for "global" metadata shared across transforms. Thoughts?
>
> yeah, can do.I don't see anything speaking against that right now.
>
> bye Jochen
and since it is allowed already (did not see the reply) maybe we should
give that method a better name?
bye Jochen
Re: ModuleNode and CompuleUnit
Posted by Paul King <pa...@asert.com.au>.
We could provide an alias I guess.
On Mon, Aug 13, 2018 at 6:09 PM Jochen Theodorou <bl...@gmx.org> wrote:
>
>
> Am 13.08.2018 um 10:08 schrieb Jochen Theodorou:
> >
> >
> > Am 12.08.2018 um 15:34 schrieb Paul King:
> >>
> >> Anyone think of a reason we shouldn't allow the CompileUnit to be
> >> accessed from ModuleNode? We already let it be accessed from
> >> ClassNode. I was only thinking read access.
> >>
> >> I was contemplating adding NodeMetaDataHandler capability to
> >> CompileUnit for "global" metadata shared across transforms. Thoughts?
> >
> > yeah, can do.I don't see anything speaking against that right now.
> >
> > bye Jochen
>
> and since it is allowed already (did not see the reply) maybe we should
> give that method a better name?
>
> bye Jochen
>
Re: ModuleNode and CompuleUnit
Posted by Jochen Theodorou <bl...@gmx.org>.
Am 13.08.2018 um 10:08 schrieb Jochen Theodorou:
>
>
> Am 12.08.2018 um 15:34 schrieb Paul King:
>>
>> Anyone think of a reason we shouldn't allow the CompileUnit to be
>> accessed from ModuleNode? We already let it be accessed from
>> ClassNode. I was only thinking read access.
>>
>> I was contemplating adding NodeMetaDataHandler capability to
>> CompileUnit for "global" metadata shared across transforms. Thoughts?
>
> yeah, can do.I don't see anything speaking against that right now.
>
> bye Jochen
and since it is allowed already (did not see the reply) maybe we should
give that method a better name?
bye Jochen
Re: ModuleNode and CompuleUnit
Posted by Jochen Theodorou <bl...@gmx.org>.
Am 12.08.2018 um 15:34 schrieb Paul King:
>
> Anyone think of a reason we shouldn't allow the CompileUnit to be
> accessed from ModuleNode? We already let it be accessed from ClassNode.
> I was only thinking read access.
>
> I was contemplating adding NodeMetaDataHandler capability to CompileUnit
> for "global" metadata shared across transforms. Thoughts?
yeah, can do.I don't see anything speaking against that right now.
bye Jochen