Mailing-List: contact users-help@groovy.apache.org; run by ezmlm
Precedence: bulk
Reply-To: users@groovy.apache.org
MIME-Version: 1.0
In-Reply-To: <583376C8.7090101@gmx.org>
References: <5C676E6359909E478C7B811BDB48CA35613416@CWWAPP478.windstream.com> <583376C8.7090101@gmx.org>
From: Keegan Witt <keeganwitt@gmail.com>
Date: Mon, 21 Nov 2016 22:08:10 -0500
Message-ID: <CAL2qGUGB_jRs9vt+0904u5PH1_hGzL17Ewhp2soZT9B2GcY9iw@mail.gmail.com>
Subject: Re: GMavenPlus compiles stubs
To: users@groovy.apache.org
Content-Type: multipart/alternative; boundary=94eb2c0880e6460cd70541db1643
archived-at: Tue, 22 Nov 2016 03:08:36 -0000

--94eb2c0880e6460cd70541db1643
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Jochen is right, Groovy compilation happens after stubgen and javac, hence
it'll replace those previously generated classes.

Why shouldn't javac compile the stubs?  Wouldn't javac need the classes so
that it has something to compile against (since Java is statically
compiled) so that circular compilation works?  GMaven worked this way also.

-Keegan

On Nov 21, 2016 4:36 PM, "Jochen Theodorou" <blackdrag@gmx.org> wrote:

> On 21.11.2016 23:08, Winnebeck, Jason wrote:
>
>> I recently encountered an error from javac compiling a stub that was
>> otherwise valid from Groovy itself when I enabled generate-stubs task. I
>> resolved the error properly, but I wonder, should javac be compiling
>> stubs when using gmavenplus?
>>
>
> afaik no
>
> I thought it just referred to them but did
>> not compile them. If it=E2=80=99s compiling the stubs, am I at risk of h=
aving
>> the stub replace the Groovy-generated code in the final jar?
>>
>
> since groovy compilation happens after the groovy compiler will overwrite
> those classes. So no danger from that front
>
> bye Jochen
>

--94eb2c0880e6460cd70541db1643
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><p dir=3D"ltr">Jochen is right, Groovy compilation happens=
 after stubgen and javac, hence it&#39;ll replace those previously generate=
d classes.</p>
<p dir=3D"ltr">Why shouldn&#39;t javac compile the stubs?=C2=A0 Wouldn&#39;=
t javac need the classes so that it has something to compile against (since=
 Java is statically compiled) so that circular compilation works?=C2=A0 GMa=
ven worked this way also.</p>
<p dir=3D"ltr">-Keegan</p>
<div class=3D"gmail_extra"><br><div class=3D"gmail_quote">On Nov 21, 2016 4=
:36 PM, &quot;Jochen Theodorou&quot; &lt;<a href=3D"mailto:blackdrag@gmx.or=
g" target=3D"_blank">blackdrag@gmx.org</a>&gt; wrote:<br type=3D"attributio=
n"><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left=
:1px #ccc solid;padding-left:1ex">On 21.11.2016 23:08, Winnebeck, Jason wro=
te:<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
I recently encountered an error from javac compiling a stub that was<br>
otherwise valid from Groovy itself when I enabled generate-stubs task. I<br=
>
resolved the error properly, but I wonder, should javac be compiling<br>
stubs when using gmavenplus?<br>
</blockquote>
<br>
afaik no<br>
<br>
<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">
I thought it just referred to them but did<br>
not compile them. If it=E2=80=99s compiling the stubs, am I at risk of havi=
ng<br>
the stub replace the Groovy-generated code in the final jar?<br>
</blockquote>
<br>
since groovy compilation happens after the groovy compiler will overwrite t=
hose classes. So no danger from that front<br>
<br>
bye Jochen<br>
</blockquote></div></div>
</div>

--94eb2c0880e6460cd70541db1643--