[PATCH 2/7] ldb:ldbdel tool - use LDB result constants for consistency

[Matthias Dieter Wallnöfer]

Simo,

you mean on each return/exit "-1" case LDB_ERR_OPERATIONS_ERROR? Do I 
understand it correctly?

Cheers,
Matthias

simo wrote:
> On Fri, 2011-02-04 at 08:54 +0100, Matthias Dieter Wallnöfer wrote:
>    
>> Simo,
>>
>> can you live with my decision?
>>      
> Why -1, and not simply LDB_ERR_OPERATIONS_ERROR ?
>
> Simo.
>
>    
>> Cheers,
>> Matthias
>>
>> Matthias Dieter Wallnöfer wrote:
>>      
>>> Okay Simo,
>>>
>>> I've updated the patchset in my private repo (branch ldb_tools):
>>> - I'm now returning "-1" on generic errors - where a LDB error does
>>> not really fit. As an exception I've left the "ldbedit" tool since the
>>> conversion to LDB errors there is very complicated
>>> - Otherwise the LDB error code is returned
>>>
>>> Cheers,
>>> Matthias
>>>
>>> simo wrote:
>>>        
>>>> On Tue, 2011-02-01 at 17:37 +0100, Matthias Dieter Wallnöfer wrote:
>>>>          
>>>>> Yeah simo,
>>>>>
>>>>> I agree. But which method do you prefer? "-1" or LDB errors?
>>>>>
>>>>> Then, after this has been cleared I'm willing to rework my patches to
>>>>> comply with the decision.
>>>>>            
>>>> Given it has been inconsistent for more than a year now, I have no
>>>> preference, unless someone is depending on the return value in some
>>>> scripts, like automatic testing.
>>>>
>>>> If nobody is depending on it then I would be ok with returning LDB error
>>>> codes
>>>>
>>>> Simo.
>>>>
>>>>          
>>>
>>>        
>
>