Hi Chris,

Thanks a lot for the clarification!

As long as it is considered OK to give a boolean (whatever it becomes under
the hood), then I'm happy.
Indeed, adding a test case and official doc for using booleans makes total
sense and would avoid unexpected change in that area.

Best regards,

Jonathan



On Wed, Apr 28, 2021 at 9:46 PM Chris Marchbanks <[email protected]>
wrote:

> Hi Jonathan,
>
> As it stands I would say that is an implementation detail that could
> change, for example we might treat integers differently from floats in the
> future. That is just the status quo, and if the community would like to see
> this as the official behavior we could add a test case for setting values
> to booleans. OpenMetrics even specifies that booleans must be represented
> as 0/1 so I think there is a reasonable case to make it official.
>
> Chris
>
> On Wed, Apr 28, 2021 at 4:16 AM Jonathan Martin <[email protected]>
> wrote:
>
>> Hi there,
>>
>> I have a question relative to Gauge and booleans, if you don't mind.
>>
>> I'm considering to use a gauge to reflect the evolution of a boolean
>> value.
>>
>> Knowing that the set method of a Gauge cast the given value as a float, I
>> could call directly my_gauge.set(my_bool_value), and have it set to 0.0 or
>> 1.0.
>>
>> So the question here is: Is the casting as float part of the API design,
>> or simply an implementation detail which may change?
>>
>> If we have no guarantee that this cast will remain, then the call to set
>> needs to be protected by my_gauge.set(float(my_bool_value)), but it is a
>> pity to call it twice...
>>
>> It's a nice built in capability, so if it is going to last, I may argue
>> that it would be worth mentioning it explicitly somewhere, either in the
>> doc of the *set* function or the documentation of the Gauge (
>> https://github.com/prometheus/client_python#gauge).
>>
>> Thanks in advance!
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "Prometheus Developers" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/prometheus-developers/ffcee28e-d367-4897-8b30-ff5a1e28daean%40googlegroups.com
>> <https://groups.google.com/d/msgid/prometheus-developers/ffcee28e-d367-4897-8b30-ff5a1e28daean%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-developers/CAFiVW2E0ayndmk%3DUmRit76dbb-DMnVQUmoWz-F%2B_gpKzBwJ17Q%40mail.gmail.com.

Reply via email to