v1: https://www.redhat.com/archives/libvir-list/2017-April/msg01051.html

Changes since v1:

 - Patches 1, 2, 4, and 14 were pushed since they were ACK and "separable"

 - Former patch 3 is now patch 1
   -> Restore the comments for the functions.

 - Former patch 5 is now split into patch 2 and 3
   -> Patch 2 addresses the "if (!obj)" check in virSecretObjListRemove

   -> Patch 3 handles the naming of the variables

 - Former patch 6 is now patch 4
   -> Alterations here are not creating a @def for the express purpose
      of only dereferencing obj->def if that's all that's used. If though
      there is an obj->def->X, then the @def is created

 - Former patch 7 is removed (e.g. keep virSecretObjListGetUUIDs where
   it was). A battle for a different day perhaps.

 - Former patch 8 is now patches 5 and 6
   -> Patch 5 handles just the cleanup path for virSecretObjListGetUUIDs

   -> Patch 6 changes the variable names

 - Former patch 9 is now patches 7 and 8
   -> Patch 7 focuses only on changing the name of @filter to @aclfilter
      (something that wasn't called out specifically in the initial review
       but that I figured I'd do anyway)

   -> Patch 8 does the split and keeps the Callback functions together
      so it doesn't appear that virSecretObjListNumOfSecrets moved. Again
      I'll pick that battle another day. I did change the name of the
      structures to include "virSecret" prefix though.

 - Former patch 10 is now patch 9
   -> This was ACK'd but not "easy" to extract out, so nothing to do here.

 - Former patch 11 is now patch 10
   -> Perhaps merge conflicts were changed here. The algorithm hasn't
      changed though.

 - Former patch 12 is now patch 11
   -> This was ACKd already - just forgot to extract it out for a push.

 - Former patch 13 is removed

So I've tried to go with the spirit of the review even though I disagree
with the don't do the code motion part.  I'll deal with that later when
things really converge and the functions are shorter.

John Ferlan (11):
  secret: Make some virSecretObj* functions static
  secret: Add NULL obj check to virSecretObjListRemove
  secret: Use consistent naming for variables
  secret: Use virSecretDefPtr rather than deref from virSecretObjPtr
  secret: Alter cleanup path for virSecretObjListGetUUIDs
  secret: Change variable names for list traversals
  secret: Rename 'filter' to 'aclfilter'
  secret: Split apart NumOfSecrets and GetUUIDs callback function
  secret: Combine virSecretObjListAdd with Locked function
  secret: Alter FindByUUID to expect the formatted uuidstr
  secret: Generate configDir during driver initialization

 src/conf/virsecretobj.c    | 428 ++++++++++++++++++++++++---------------------
 src/conf/virsecretobj.h    |  54 ++----
 src/secret/secret_driver.c | 170 +++++++++---------
 3 files changed, 328 insertions(+), 324 deletions(-)

-- 
2.9.3

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

On Tue, Apr 25, 2017 at 05:15:24PM -0400, John Ferlan wrote:
> v1: https://www.redhat.com/archives/libvir-list/2017-April/msg01051.html
> 
> Changes since v1:
> 
>  - Patches 1, 2, 4, and 14 were pushed since they were ACK and "separable"
> 
>  - Former patch 3 is now patch 1
>    -> Restore the comments for the functions.
> 
>  - Former patch 5 is now split into patch 2 and 3
>    -> Patch 2 addresses the "if (!obj)" check in virSecretObjListRemove
> 
>    -> Patch 3 handles the naming of the variables
> 
>  - Former patch 6 is now patch 4
>    -> Alterations here are not creating a @def for the express purpose
>       of only dereferencing obj->def if that's all that's used. If though
>       there is an obj->def->X, then the @def is created
> 
>  - Former patch 7 is removed (e.g. keep virSecretObjListGetUUIDs where
>    it was). A battle for a different day perhaps.
> 
>  - Former patch 8 is now patches 5 and 6
>    -> Patch 5 handles just the cleanup path for virSecretObjListGetUUIDs
> 
>    -> Patch 6 changes the variable names
> 
>  - Former patch 9 is now patches 7 and 8
>    -> Patch 7 focuses only on changing the name of @filter to @aclfilter
>       (something that wasn't called out specifically in the initial review
>        but that I figured I'd do anyway)
> 
>    -> Patch 8 does the split and keeps the Callback functions together
>       so it doesn't appear that virSecretObjListNumOfSecrets moved. Again
>       I'll pick that battle another day. I did change the name of the
>       structures to include "virSecret" prefix though.
> 
>  - Former patch 10 is now patch 9
>    -> This was ACK'd but not "easy" to extract out, so nothing to do here.
> 
>  - Former patch 11 is now patch 10
>    -> Perhaps merge conflicts were changed here. The algorithm hasn't
>       changed though.
> 
>  - Former patch 12 is now patch 11
>    -> This was ACKd already - just forgot to extract it out for a push.
> 
>  - Former patch 13 is removed
> 
> So I've tried to go with the spirit of the review even though I disagree
> with the don't do the code motion part.  I'll deal with that later when
> things really converge and the functions are shorter.

ACK series if you fix the nits pointed out for patch 03 and 04 and if you
drop patch 10.

Pavel
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

On 04/26/2017 04:46 AM, Pavel Hrdina wrote:
> On Tue, Apr 25, 2017 at 05:15:24PM -0400, John Ferlan wrote:
>> v1: https://www.redhat.com/archives/libvir-list/2017-April/msg01051.html
>>
>> Changes since v1:
>>
>>  - Patches 1, 2, 4, and 14 were pushed since they were ACK and "separable"
>>
>>  - Former patch 3 is now patch 1
>>    -> Restore the comments for the functions.
>>
>>  - Former patch 5 is now split into patch 2 and 3
>>    -> Patch 2 addresses the "if (!obj)" check in virSecretObjListRemove
>>
>>    -> Patch 3 handles the naming of the variables
>>
>>  - Former patch 6 is now patch 4
>>    -> Alterations here are not creating a @def for the express purpose
>>       of only dereferencing obj->def if that's all that's used. If though
>>       there is an obj->def->X, then the @def is created
>>
>>  - Former patch 7 is removed (e.g. keep virSecretObjListGetUUIDs where
>>    it was). A battle for a different day perhaps.
>>
>>  - Former patch 8 is now patches 5 and 6
>>    -> Patch 5 handles just the cleanup path for virSecretObjListGetUUIDs
>>
>>    -> Patch 6 changes the variable names
>>
>>  - Former patch 9 is now patches 7 and 8
>>    -> Patch 7 focuses only on changing the name of @filter to @aclfilter
>>       (something that wasn't called out specifically in the initial review
>>        but that I figured I'd do anyway)
>>
>>    -> Patch 8 does the split and keeps the Callback functions together
>>       so it doesn't appear that virSecretObjListNumOfSecrets moved. Again
>>       I'll pick that battle another day. I did change the name of the
>>       structures to include "virSecret" prefix though.
>>
>>  - Former patch 10 is now patch 9
>>    -> This was ACK'd but not "easy" to extract out, so nothing to do here.
>>
>>  - Former patch 11 is now patch 10
>>    -> Perhaps merge conflicts were changed here. The algorithm hasn't
>>       changed though.
>>
>>  - Former patch 12 is now patch 11
>>    -> This was ACKd already - just forgot to extract it out for a push.
>>
>>  - Former patch 13 is removed
>>
>> So I've tried to go with the spirit of the review even though I disagree
>> with the don't do the code motion part.  I'll deal with that later when
>> things really converge and the functions are shorter.
> 
> ACK series if you fix the nits pointed out for patch 03 and 04 and if you
> drop patch 10.
> 

Removed extraneous newline adjustments from secret_driver in patch 3
(that's a type A type thing and I forgot to separate it...)...  Fixed
commit message in patch 4, dropped patch 10...

Thanks for the reviews for this and the nwfilter series - it also helps
me for "direction" for the other vir*obj and *drivers that will be
adjusted as well...


John


--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list