Re: [gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-29 Thread Michał Górny 
On Tue, 29 Nov 2016 09:46:18 +0100
Michał Górny  wrote:

> On Mon, 28 Nov 2016 06:42:44 -0500
> Mike Pagano  wrote:
> 
> > On 11/27/2016 09:06 PM, Mike Gilbert wrote:  
> > > On Sun, Nov 27, 2016 at 7:23 PM, Mike Pagano  wrote:  
> > >   
> > >> If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
> > >> inherited from eutils.eclass is undefined.
> > >> See bug #579188
> > > 
> > > kernel-2.eclass currently calls die in global scope for EAPI 6, so
> > > this change has no real effect.
> > > 
> > > I would suggest modernizing the eclass before allowing it to be used
> > > with EAPI 6. For example, patches should be applied in src_prepare,
> > > not src_unpack.
> > > 
> > 
> > Is this change not a step in the right direction? Or should it just be
> > one big bang?
> > All or nothing?  
> 
> While I don't mean incremental changes, I'd really prefer if they made

s/mean/mind/.


-- 
Best regards,
Michał Górny



pgp_GoxXSx1pG.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-29 Thread Mike Pagano 


On 11/29/2016 03:46 AM, Michał Górny wrote:
> On Mon, 28 Nov 2016 06:42:44 -0500
> Mike Pagano  wrote:
> 
>> On 11/27/2016 09:06 PM, Mike Gilbert wrote:
>>> On Sun, Nov 27, 2016 at 7:23 PM, Mike Pagano  wrote:  
 If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
 inherited from eutils.eclass is undefined.
 See bug #579188  
>>>
>>> kernel-2.eclass currently calls die in global scope for EAPI 6, so
>>> this change has no real effect.
>>>
>>> I would suggest modernizing the eclass before allowing it to be used
>>> with EAPI 6. For example, patches should be applied in src_prepare,
>>> not src_unpack.
>>>   
>>
>> Is this change not a step in the right direction? Or should it just be
>> one big bang?
>> All or nothing?
> 
> While I don't mean incremental changes, I'd really prefer if they made
> any sense. Calling eapply_user in src_unpack() is invalid, per PMS:
> 
> | In EAPIs where it is supported, eapply_user must be called once
> | in the src_prepare phase.
> 

Your last sentence is constructive criticism. I'll fix this.

-- 
Mike Pagano
Gentoo Developer - Kernel Project
Gentoo Sources - Lead
E-Mail : mpag...@gentoo.org
GnuPG FP   : 52CC A0B0 F631 0B17 0142 F83F 92A6 DBEC 81F2 B137
Public Key :
http://http://pgp.mit.edu/pks/lookup?search=0x92A6DBEC81F2B137=index




signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-29 Thread Michał Górny 
On Mon, 28 Nov 2016 06:42:44 -0500
Mike Pagano  wrote:

> On 11/27/2016 09:06 PM, Mike Gilbert wrote:
> > On Sun, Nov 27, 2016 at 7:23 PM, Mike Pagano  wrote:  
> >> If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
> >> inherited from eutils.eclass is undefined.
> >> See bug #579188  
> > 
> > kernel-2.eclass currently calls die in global scope for EAPI 6, so
> > this change has no real effect.
> > 
> > I would suggest modernizing the eclass before allowing it to be used
> > with EAPI 6. For example, patches should be applied in src_prepare,
> > not src_unpack.
> >   
> 
> Is this change not a step in the right direction? Or should it just be
> one big bang?
> All or nothing?

While I don't mean incremental changes, I'd really prefer if they made
any sense. Calling eapply_user in src_unpack() is invalid, per PMS:

| In EAPIs where it is supported, eapply_user must be called once
| in the src_prepare phase.

-- 
Best regards,
Michał Górny



pgpyK9DhrPl8u.pgp
Description: OpenPGP digital signature

Re: [gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-28 Thread Mike Gilbert 
On Mon, Nov 28, 2016 at 6:42 AM, Mike Pagano  wrote:
>
>
> On 11/27/2016 09:06 PM, Mike Gilbert wrote:
>> On Sun, Nov 27, 2016 at 7:23 PM, Mike Pagano  wrote:
>>> If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
>>> inherited from eutils.eclass is undefined.
>>> See bug #579188
>>
>> kernel-2.eclass currently calls die in global scope for EAPI 6, so
>> this change has no real effect.
>>
>> I would suggest modernizing the eclass before allowing it to be used
>> with EAPI 6. For example, patches should be applied in src_prepare,
>> not src_unpack.
>>
>
>
> Is this change not a step in the right direction? Or should it just be
> one big bang?
> All or nothing?

A full patch series would make more sense than piecemeal commits I think.

Re: [gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-28 Thread Mike Pagano 


On 11/27/2016 09:06 PM, Mike Gilbert wrote:
> On Sun, Nov 27, 2016 at 7:23 PM, Mike Pagano  wrote:
>> If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
>> inherited from eutils.eclass is undefined.
>> See bug #579188
> 
> kernel-2.eclass currently calls die in global scope for EAPI 6, so
> this change has no real effect.
> 
> I would suggest modernizing the eclass before allowing it to be used
> with EAPI 6. For example, patches should be applied in src_prepare,
> not src_unpack.
> 


Is this change not a step in the right direction? Or should it just be
one big bang?
All or nothing?

-- 
Mike Pagano
Gentoo Developer - Kernel Project
Gentoo Sources - Lead
E-Mail : mpag...@gentoo.org
GnuPG FP   : 52CC A0B0 F631 0B17 0142 F83F 92A6 DBEC 81F2 B137
Public Key :
http://http://pgp.mit.edu/pks/lookup?search=0x92A6DBEC81F2B137=index




signature.asc
Description: OpenPGP digital signature

Re: [gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-27 Thread Mike Gilbert 
On Sun, Nov 27, 2016 at 7:23 PM, Mike Pagano  wrote:
> If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
> inherited from eutils.eclass is undefined.
> See bug #579188

kernel-2.eclass currently calls die in global scope for EAPI 6, so
this change has no real effect.

I would suggest modernizing the eclass before allowing it to be used
with EAPI 6. For example, patches should be applied in src_prepare,
not src_unpack.

[gentoo-dev] [PATCH 1/1] kernel-2.eclass: Support EAPI 6 when applying user patches.

2016-11-27 Thread Mike Pagano 
If kernel-2_src_unpack() is called in EAPI6, epatch_user() function
inherited from eutils.eclass is undefined.
See bug #579188

---
 eclass/kernel-2.eclass | 8 ++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/eclass/kernel-2.eclass b/eclass/kernel-2.eclass
index c171940..3f5fb3b 100644
--- a/eclass/kernel-2.eclass
+++ b/eclass/kernel-2.eclass
@@ -1260,8 +1260,12 @@ kernel-2_src_unpack() {
# we run misc `make` functions below
[[ $(type -t kernel-2_hook_premake) == "function" ]] &&
kernel-2_hook_premake

-   debug-print "Doing epatch_user"
-   epatch_user
+   debug-print "Applying user patches if they exist"
+   # apply any user patches
+case ${EAPI:-0} in
+0|1|2|3|4|5) epatch_user ;;
+6) eapply_user ;;
+esac

debug-print "Doing unpack_set_extraversion"

-- 
2.7.3



signature.asc
Description: OpenPGP digital signature