You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After #728, it's possible users will have items that were previously valid / decodeable that have become invalid / un-decodeable because items now decode the datetime / start_datetime + end_datetime fields. If those fields are all missing, or if an item has them present but they can't be decoded in any of the time formats we recognize (e.g. if the middle T is missing), Franklin will throw an exception when serializing results. That's not great! We should provide a command that users can run that identifies the items that can't be decoded and prints their ids and their current datetime, start_datetime, and end_datetime values (focused on time since that's where the breaking change occurred).
Notes + Context
I discovered this issue with some test data missing the middle T. I think RFC3339 says that the middle T is required (though it can be a t), but I get sleepy every time I start reading that spec so not certain about that 😴
The text was updated successfully, but these errors were encountered:
Improvement
After #728, it's possible users will have items that were previously valid / decodeable that have become invalid / un-decodeable because items now decode the
datetime
/start_datetime
+end_datetime
fields. If those fields are all missing, or if an item has them present but they can't be decoded in any of the time formats we recognize (e.g. if the middleT
is missing), Franklin will throw an exception when serializing results. That's not great! We should provide a command that users can run that identifies the items that can't be decoded and prints their ids and their currentdatetime
,start_datetime
, andend_datetime
values (focused on time since that's where the breaking change occurred).Notes + Context
I discovered this issue with some test data missing the middle
T
. I think RFC3339 says that the middleT
is required (though it can be at
), but I get sleepy every time I start reading that spec so not certain about that 😴The text was updated successfully, but these errors were encountered: