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
I was working on a personal project 5 months ago which had dbt, and this package, everything was working well.
Now as I was trying to run the project again, it is not going as well.
test_dbt__bi__yellow_m2m and dbt__bi__yellow_m2m_expected being seeds that have same configuration and are in the same directory and are also loaded to the same bigquery dataset.
the 2 seeds originate from 2 csv files with the same names.
Expected results
A big query job combining the 2 seeds and testing the results of apply a model to one of the input and comparing it to the output.
Actual results
the tool uses only 1 of the two table, the expected_output table, and attempts to create the other table.
/* {"app": "dbt", "dbt_version": "1.5.2", "profile_name": "default", "target_name": "dev", "node_id": "test.nytaxi.dbt_datamocktool_unit_test_dbt__bi__yellow_m2m_ref_seed_zones___ref_dbt__bi__yellow_m2m_expected___ref_test_dbt__bi__yellow_m2m_.ac26a1c36a"} */
with a as (
select
`vendors`
,
`year`
,
`month`
,
`monthly_fare_amount`
,
`average_passenger_count`
,
`average_trip_distance`
,
`monthly_extra_amount`
,
`monthly_mta_tax`
,
`monthly_tip_amount`
,
`monthly_tolls_amount`
,
`monthly_improvement_surcharge`
,
`monthly_total_amount`
,
`monthly_congestion_surcharge`
,
`pickup_borough`
,
`pickup_zone`
,
`dropoff_borough`
,
`dropoff_zone`
,
`avg_duration_minutes`
,
`trip_count`
,
`total_duration_hours`
,
`busiest_hour_of_day`
,
`airport_trip`
,
`payment_type`
from `stellarismusv5`.`models_ut`.`dbt__bi__yellow_m2m_expected`
),
b as (
select
`vendors`
,
`year`
,
`month`
,
`monthly_fare_amount`
,
`average_passenger_count`
,
`average_trip_distance`
,
`monthly_extra_amount`
,
`monthly_mta_tax`
,
`monthly_tip_amount`
,
`monthly_tolls_amount`
,
`monthly_improvement_surcharge`
,
`monthly_total_amount`
,
`monthly_congestion_surcharge`
,
`pickup_borough`
,
`pickup_zone`
,
`dropoff_borough`
,
`dropoff_zone`
,
`avg_duration_minutes`
,
`trip_count`
,
`total_duration_hours`
,
`busiest_hour_of_day`
,
`airport_trip`
,
`payment_type`
from -- depends_on: `stellarismusv5`.`models_seeds`.`seed_zones`
`stellarismusv5`.`models_historical`.`dbt__bi__yellow_m2m_dmt_49567973`
),
a_intersect_b as (
select * from a
intersect distinct
select * from b
),
a_except_b as (
select * from a
except distinct
select * from b
),
b_except_a as (
select * from b
except distinct
select * from a
),
all_records as (
select
*,
true as in_a,
true as in_b
from a_intersect_b
union all
select
*,
true as in_a,
false as in_b
from a_except_b
union all
select
*,
false as in_a,
true as in_b
from b_except_a
),
final as (
select * from all_records
where not (in_a and in_b)
order by in_a desc, in_b desc
)
select * from final
dbt__bi__yellow_m2m_dmt_49567973 is not a table that I have created. It is basically an identical copy of the table that is being tested, and it is even created in the same dataset as the original table.
while stellarismusv5.models_ut.dbt__bi__yellow_m2m_expected is the correct seed for expected output.
I checked issues first, and it does not look like this is a common issue, so I'd wager that this is something incorrect on my end.
I have spent the better part of the day fiddling around, trying things out, changing directories, names, nothing works. I cannot seem to
pinpoint the cause, so any hints would be appreciated.
The text was updated successfully, but these errors were encountered:
Describe the bug
I was working on a personal project 5 months ago which had dbt, and this package, everything was working well.
Now as I was trying to run the project again, it is not going as well.
Steps to reproduce
test_dbt__bi__yellow_m2m
anddbt__bi__yellow_m2m_expected
being seeds that have same configuration and are in the same directory and are also loaded to the same bigquery dataset.the 2 seeds originate from 2 csv files with the same names.
Expected results
A big query job combining the 2 seeds and testing the results of apply a model to one of the input and comparing it to the output.
Actual results
the tool uses only 1 of the two table, the
expected_output
table, and attempts to create the other table.dbt__bi__yellow_m2m_dmt_49567973
is not a table that I have created. It is basically an identical copy of the table that is being tested, and it is even created in the same dataset as the original table.while
stellarismusv5
.models_ut
.dbt__bi__yellow_m2m_expected
is the correct seed for expected output.Screenshots and log output
System information
Which database are you using dbt with?
The output of
dbt --version
:Additional context
I checked issues first, and it does not look like this is a common issue, so I'd wager that this is something incorrect on my end.
I have spent the better part of the day fiddling around, trying things out, changing directories, names, nothing works. I cannot seem to
pinpoint the cause, so any hints would be appreciated.
The text was updated successfully, but these errors were encountered: