Re: [PATCH 2/2] Support publishing evaluation events

[Christopher Baines]

Ludovic Courtès <address@hidden> writes:

> Christopher Baines <address@hidden> skribis:
>
>> ---
>>  src/cuirass/base.scm     | 5 +++++
>>  src/cuirass/database.scm | 6 +++++-
>>  src/cuirass/http.scm     | 9 +++++++++
>>  3 files changed, 19 insertions(+), 1 deletion(-)
>>
>> diff --git a/src/cuirass/base.scm b/src/cuirass/base.scm
>> index fd10e35..943a4f2 100644
>> --- a/src/cuirass/base.scm
>> +++ b/src/cuirass/base.scm
>> @@ -724,6 +724,11 @@ started)."
>>               (checkouts (fetch-inputs spec))
>>               (eval-id (db-add-evaluation name checkouts)))
>>          (when eval-id
>> +          (db-add-event 'evaluation
>> +                        (time-second (current-time time-utc))
>> +                        `((#:evalutaion    . ,eval-id)
>                                       ^^^
> Typo.

Hmm, I remember testing this, but I guess it just sent out a JSON object
with that typo in it.

>> +                          (#:specification . ,name)
>> +                          (#:in_progress   . #t)))
>                                   ^
> Should it be a hyphen?

This'll end up being part of the JSON object sent out. It looks like the
existing /api/evaluations reponse uses in-progress, so this should
probably also be a hyphen for consistency.

> Otherwise LGTM.
>
> I think we should eventually use records everywhere instead of alists,
> as that would catch such typos.  We could use an approach similar to
> that of ‘define-json-mapping’ to factorize
> serialization/deserialization.

That sounds like a good idea :)

Thanks,

Chris

signature.asc
Description: PGP signature