[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[bug#62490] [PATCH] services: nginx: Make logging level configurable.
From: |
Maxim Cournoyer |
Subject: |
[bug#62490] [PATCH] services: nginx: Make logging level configurable. |
Date: |
Tue, 28 Mar 2023 11:03:45 -0400 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/28.2 (gnu/linux) |
OK,
Bruno Victal <mirai@makinata.eu> writes:
> Hi Maxim,
>
> On 2023-03-28 15:12, Maxim Cournoyer wrote:
>> Bruno Victal <mirai@makinata.eu> writes:
>>>
>>> +(define-compile-time-procedure (assert-valid-log-level (level symbol?))
>>> + "Ensure @var{level} is one of @code{'debug}, @code{'info},
>>> @code{'notice},
>>> +@code{'warn}, @code{'error}, @code{'crit}, @code{'alert}, or
>>> @code{'emerg}."
>>> + (unless (memq level '(debug info notice warn error crit alert emerg))
>>> + (raise
>>> + (make-compound-condition
>>> + (formatted-message (G_ "unknown log level '~a'") level)
>>> + (condition (&error-location
>>> + (location
>>> + (source-properties->location
>>> procedure-call-location)))))))
>>> + level)
>>
>> It's the first time I've seen define-compile-time-procedure in actual
>> use. Is it really necessary? What happens if you omit wrapping
>> assert-valid-log-level with it?
>
> It will still work, provided the declaration is adjusted accordingly.
> As for the reasons and benefits of using define-compile-time-procedure,
> it's best explained at <https://logs.guix.gnu.org/guix/2023-03-20.log#131047>.
I guess it's not actually useful here, since none of the fields thunked?
As another note, the nginx-configuration record looks simple enough that
perhaps it'd best be migrated to use the define-configuration method,
which could be made as a prior commit to this change.
As a benefit it'd validate the other field types as well.
--
Thanks,
Maxim