[OE-core] [PATCH] sanity.bbclass: Improved error message

[Mark Hatle]

On 1/12/17 6:36 PM, Paul Eggleton wrote:
> On Thu, 12 Jan 2017 16:04:09 Mark Hatle wrote:
>> On 1/12/17 3:42 PM, Richard Purdie wrote:
>>> On Thu, 2017-01-12 at 11:24 -0800, Juro Bystricky wrote:
>>>> When a non-existing MACHINE is specified, sanity check issues
>>>>
>>>> the following message:
>>>>     Please set a valid MACHINE in your local.conf or environment
>>>>
>>>> However, MACHINE can also be set in multiconfig .conf file(s).
>>>> Hence we may have several different MACHINE settings within one
>>>> (multiconfig) build, so the present error message is fairly
>>>> ambiguous.
>>>>
>>>> This patch remedies this by explicitly naming the offending MACHINE
>>>> and
>>>> adding conf/multiconfig/*.conf to the list of places where this
>>>> erroneous
>>>> MACHINE definition could have originated.
>>>>
>>>>     Invalid MACHINE=xyz. Please set a valid MACHINE in your
>>>>
>>>> local.conf, conf/multiconfig/*.conf or environment
>>>
>>> Can we change this to:
>>>
>>> "MACHINE=xyz is invalid. Please set a valid MACHINE in your local.conf,
>>> environment or other configuration file."
>>
>> I commented elsewhere as well -- but is there any way to see where MACHINE
>> was actually set and display that -- or is it limited to 'bitbake -e'
>> usage?
> 
> You need to have history enabled in the datastore to be able to retrieve that, 
> and we don't during normal mode for performance over the entire run. It's 
> possible we could enable it reparse the configuration when errors like this 
> occur to get it though - it probably wouldn't take too long, but I don't know 
> if there are other practical barriers.

Ok.. I couldn't remember if history was needed or not.  I know one of the
implementations included a history reference for the last one (instead of a full
history...)

It's not worth making this more difficult to implement then.

--Mark

> Cheers,
> Paul
>