ValueError: could not assign tuple of length 4 to structure with 3 fields.

[nalcabio-tom]

Adding p_timestamp as a filter in the superset dashboard causes the error below. Using ctx#timestamp instead doesn't cause an error, although that might be because no data is returned since the start and end times restrict the query to the last 10 minutes. (Last time data was added to parseable was yesterday, fyi.)

025-02-25 17:38:07,714:WARNING:superset.connectors.sqla.models:Query SELECT "ctx#timestamp" AS "ctx#timestamp", "ctx#event" AS "ctx#event", "data#device-state" AS "data#device-state", p_timestamp AS p_timestamp
FROM nalca1
WHERE p_timestamp >= '2025-02-24 00:00:00.000000' AND p_timestamp < '2025-03-03 00:00:00.000000'
 LIMIT 100 on schema default failed
Traceback (most recent call last):
  File "/app/superset/connectors/sqla/models.py", line 1761, in query
    df = self.database.get_df(
  File "/app/superset/models/core.py", line 684, in get_df
    result_set = SupersetResultSet(
  File "/app/superset/result_set.py", line 135, in __init__
    array = np.array(data, dtype=numpy_dtype)
ValueError: could not assign tuple of length 4 to structure with 3 fields.

See also: #5

[AdheipSingh]

Hi @nalcabio-tom , thanks for bringing up this issue.
Is it possible for you to sure the exact log ? if you are running superset you should see logs such that original query in superset vs modified query for parseable. It will help me understand better.

Though i get your issue, basically query is returning 4 columns and structure expected to receive data is 3 columns.
When you execute a query and get results back, the current column structure is being created based on the first row's keys here.

I'll try to re-create this issue and get back to you.

[nalcabio-tom]

Hi @AdheipSingh . Sorry, I don't understand your question. The query above does work in the parseable dashboard.

[nalcabio-tom]

@AdheipSingh , is there anything I can do to help with this? Thanks!

[AdheipSingh]

Hi @nalcabio-tom , can you try out with the latest updated package ie 0.1.5. I tried with a similar query in my dataset.

[nalcabio-tom]

Thanks @AdheipSingh ! This appears to work!