[PATCH] samba-tools: add computer subcommand

[joeg at catalyst.net.nz]

Hi,

This patch was ignored, could any one review and merge it, please?


On 16/03/18 12:18, joeg--- via samba-technical wrote:
> Sorry, there was a small mistake in that copy.
>
> Please ignore that v2 one, and use this v3 one.
>
>
> On 16/03/18 12:08, joeg at catalyst.net.nz wrote:
>> Hi Andrew,
>>
>> Thank you for pointing out the comments issue in the
>> `join_add_dns_records()` function.
>>
>> I tried to reuse the code in join.py,  but the context of the usage is
>> quite different.
>> It needs some extra effort to modify join.py to support this new case.
>> Also, that may break existing code.
>> So I chose to copy that function and modify the new copy.
>>
>> You are right, the comments are not relevant to this case.
>> I've trim them out, made some more improvement, and rename the function.
>>
>> The new patch is attached.
>>
>> Review appreciated!
>>
>>
>> On 15/03/18 17:09, Andrew Bartlett wrote:
>>> On Thu, 2018-03-15 at 11:56 +1300, joeg at catalyst.net.nz wrote:
>>>> Hi Bjoern,
>>>>
>>>> Thank you for making the samba-tools computer patch.
>>>>
>>>> According to Andrew and Garming's advice, I extract some of the
>>>> functions in my previous patch, and created a new patch based on yours:
>>>>
>>>> 1. Add `--ip-address` option for the create subcommand, to allow user
>>>> set DNS A or AAAA records while creating the computer.
>>>> 2. Delete above DNS records while deleting the computer.
>>>> 3. Add `--service-principal-name` option for the create command, to
>>>> allow user set `servicePrincipalName` while creating the computer.
>>>> 4. Tests.
>>>>
>>>> Could you or any one else review this patch? Any feedback are welcome
>>>> and appreciated!
>>>>
>>> Thanks Joe!
>>>
>>> I've focussed on your changes in particular, I haven't looked at
>>> Bjoern's changes yet.
>>>
>>> For your changes, can you either share the join_add_dns_records() with
>>> the code in join.py or trim out the comments to be relevents to the add
>>> computer case?  
>>>
>>> It would be great if the preparatory work to change the owner could be
>>> moved inside that function also, to keep the logic in the library
>>> functions and not in the main command.  
>>>
>>> Particularly if you can share it with join.py the actual business logic
>>> should be outside the netcmd stuff, and as a re-usable function.  (See
>>> how samdb.newcomputer() isn't inline). 
>>>
>>> Thanks!
>>>
>>> Andrew Bartlett
>>>

-- 
Joe Guo
joeg at catalyst.net.nz
Catalyst IT