joellubi commented on code in PR #1456:
URL: https://github.com/apache/arrow-adbc/pull/1456#discussion_r1464167851
##########
go/adbc/driver/snowflake/record_reader.go:
##########
@@ -389,7 +386,15 @@ func jsonDataToArrow(ctx context.Context, bldr
*array.RecordBuilder, ld gosnowfl
return nil, err
}
- if tz != time.UTC {
+ snowflakeType, ok :=
bldr.Schema().Field(i).Metadata.GetValue(MetadataKeySnowflakeType)
+ if !ok {
+ return nil, errToAdbcErr(
+ adbc.StatusInvalidData,
+ fmt.Errorf("key %s not found in
metadata for field %s", MetadataKeySnowflakeType, bldr.Schema().Field(i).Name),
+ )
+ }
Review Comment:
Ok I refactored this section. I kept the `SNOWFLAKE_TYPE` key check in place
so that this can fail fast if someone changes the logic in the future. The part
I changed was order of the checks for the timestamp variants.
Previously it was `timestamp_ltz` -> `timestamp_ntz` -> fallthrough
`timestamp_tz`.
It turns out that `timestamp_ltz` and `timestamp_ntz` can be stored in the
exact same way. Using `time.Unix(sec, nsec).In(tz)` was only changing the
location metadata for the `time.Time` object. It doesn't have an effect on the
output of `arrow.TimestampFromTime()` so for the nanosecond case they are
equivalent.
Now we check if it's a `timestamp_tz` first. If so, we perform the offset
logic, etc.
Otherwise we just extract and use the epoch nanoseconds directly, which
applies to `timestamp_ltz`, `timestamp_ntz`, and the default case for unset
type metadata.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
