queries that run in CLI fail in UI with error "Error rendering results: Node.appendChild: Argument 1 is not an object."

[littlegreenbean33]

Hi,

I am using v1.2.1

I systematically encounter the following error for queries that are fine in CLI but fail in UI like the following example ( from home assistant database converted to duckdb)

not sure if there any additional info I can provide

I am using firefox on Linux

JC

[littlegreenbean33]

Not sure if it is the same issue but again I found a query where the UI and the CLI produce different results/formating.

while in CLI state_n is correctly formatted with same query

[jraymakers]

In general, differences in formatting between the UI and the CLI are expected. These are different interfaces that make different choices. That said, the formatting of FLOATs could be improved.

Regarding the "Node.appendChild" issue, that is certainly a bug. It's hard to diagnose from the example you gave, however, because it relies on data we don't have. Could you find a self-contained query that reproduces the issue? It's likely due to some specific data type or values in your data.

If you can't find a self-contained query, then at least providing the DESCRIBE output for the query might help.

[littlegreenbean33]

Hi,

This is the state of the UI. Not sure if it is very helpful.

describe select hour(s.last_updated_ts) as h,* from states s where s.last_updated_ts > '2025-01-01' and s.last_updated_ts < '2025-01-02' and s.metadata_id = 527 order by h asc

gives

Analize produces

`Physical Plan

┌───────────────────────────┐
│ PROJECTION │
│ ──────────────────── │
│ #0 │
│__internal_decompress_integ│
│ ral_bigint(#1, 527) │
│ #2 │
│__internal_decompress_integ│
│ ral_bigint(#3, 0) │
│__internal_decompress_integ│
│ ral_bigint(#4, 1) │
│__internal_decompress_integ│
│ ral_bigint(#5, 16834) │
│__internal_decompress_integ│
│ ral_bigint(#6, 16834) │
│ #7 │
│ #8 │
│ #9 │
│ #10 │
│ #11 │
│ │
│ ~38279 Rows │
└─────────────┬─────────────┘
┌─────────────┴─────────────┐
│ ORDER_BY │
│ ──────────────────── │
│ "hour"(s.last_updated_ts) │
│ ASC │
└─────────────┬─────────────┘
┌─────────────┴─────────────┐
│ PROJECTION │
│ ──────────────────── │
│ #0 │
│__internal_compress_integra│
│ l_utinyint(#1, 527) │
│ #2 │
│__internal_compress_integra│
│ l_utinyint(#3, 0) │
│__internal_compress_integra│
│ l_uinteger(#4, 1) │
│__internal_compress_integra│
│ l_uinteger(#5, 16834) │
│__internal_compress_integra│
│ l_uinteger(#6, 16834) │
│ #7 │
│ #8 │
│ #9 │
│ #10 │
│ #11 │
│ │
│ ~38279 Rows │
└─────────────┬─────────────┘
┌─────────────┴─────────────┐
│ PROJECTION │
│ ──────────────────── │
│ h │
│ metadata_id │
│ context_id_bin │
│ origin_idx │
│ attributes_id │
│ old_state_id │
│ state_id │
│ last_changed_ts │
│ last_updated_ts │
│ last_reported_ts │
│ state │
│ state_n │
│ │
│ ~38279 Rows │
└─────────────┬─────────────┘
┌─────────────┴─────────────┐
│ SEQ_SCAN │
│ ──────────────────── │
│ Table: states │
│ Type: Sequential Scan │
│ │
│ Projections: │
│ last_updated_ts │
│ metadata_id │
│ context_id_bin │
│ origin_idx │
│ attributes_id │
│ old_state_id │
│ state_id │
│ last_changed_ts │
│ last_reported_ts │
│ state │
│ state_n │
│ │
│ Filters: │
│ last_updated_ts>'2024-12 │
│ -31 23:00:00+00': │
│ :TIMESTAMP WITH TIME ZONE │
│ AND last_updated_ts<'2025│
│ -01-01 23:00:00+00': │
│ :TIMESTAMP WITH TIME ZONE │
│ metadata_id=527 │
│ │
│ ~38279 Rows │
└───────────────────────────┘
`

Happy to help if you need additional info let me know.

JC

[mxdvl]

I’ve got the same bug when using geometry types and could provide a reproducible repo using public datasets such as the UK Geography.

[jraymakers]

I’ve got the same bug when using geometry types and could provide a reproducible repo using public datasets such as the UK Geography.

Please, any reliable repro would help.

[mxdvl]

Unfortunately I’m unable to reproduce it now… but it would happen in the following case, only with the “pivot table” mode and switching to “flat table” would fix the issue:

CREATE TABLE IF NOT EXISTS squares (
    id INTEGER PRIMARY KEY,
    name VARCHAR,
    geometry GEOMETRY
);

INSERT INTO squares (id, name, geometry) VALUES
  (1, 'Square A', ST_GeomFromText('POLYGON ((0 0, 1 0, 1 1, 0 1, 0 0))')),
  (2, 'Square B', ST_GeomFromText('POLYGON ((2 2, 3 2, 3 3, 2 3, 2 2))')),
  (3, 'Square C', ST_GeomFromText('POLYGON ((-1 -1, -0.5 -1, -0.5 -0.5, -1 -0.5, -1 -1))')),
  (4, 'Square D', ST_GeomFromText('POLYGON ((5 5, 10 5, 10 10, 5 10, 5 5))')),
  (5, 'Square E', ST_GeomFromText('POLYGON ((-3 1, -2 1, -2 2, -3 2, -3 1))')),
  (6, 'Square M', ST_GeomFromText('POLYGON ((7 -2, 8 -2, 8 -1, 7 -1, 7 -2))')),
  (7, 'Square M', ST_GeomFromText('POLYGON ((0 -5, 0 -4, 1 -4, 1 -5, 0 -5))')),
  (8, 'Square M', ST_GeomFromText('POLYGON ((-5 -5, -4 -5, -4 -4, -5 -4, -5 -5))')),
  (9, 'Square M', ST_GeomFromText('POLYGON ((10 0, 11 0, 11 1, 10 1, 10 0))')),
  (10, 'Final Square', ST_GeomFromText('POLYGON ((3 -3, 4 -3, 4 -2, 3 -2, 3 -3))')),
;

SELECT
  case when id % 3 = 0 then 'Fizz' when id % 5 = 0 then 'Buzz' else name end as fizzbuzz,
  ST_Area(ST_Union_Agg(geometry)) as area,
  ST_AsText(ST_Union_Agg(geometry)) as geometry,
from squares
group by fizzbuzz

But as of now it works, even with pivot:

[jraymakers]

With recent updates we've made to the results view, this should be addressed.