Recursive planning with zero columns fail
- Dominant language
- C
- Stars
- 12.8k
- Forks
- 794
- Avg merge
- 2d 14h
- Merged PRs (30d)
- 31
Description
```SQL
CREATE TABLE test_a (id int, k int);
CREATE TABLE test_b (id int, k int);
SELECT create_distributed_table('test_a','id');
SELECT create_distributed_table('test_b','id');
SELECT *, random() FROM ( SELECT from test_a where k>1 LIMIT 1) as foo;;
ERROR: a column definition list is required for functions returning "record"
LINE 1: SELECT *, random() FROM ( SELECT from test_a where k>1 LIMI...
^
CREATE OR REPLACE VIEW v AS SELECT *, random() FROM ( SELECT from test_a where k>1 LIMIT 1) as foo;
table v;
ERROR: a column definition list is required for functions returning "record"
```
Contributor guide
Research direction
Start by reproducing the SQL statements involving create_distributed_table and the empty SELECT from test_a, then trace the recursive planning path that reports the record column-definition error. Clarify the expected behavior for a zero-column subquery before identifying the relevant planner code and adding a regression case; done means the intended query and view behavior is verified.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- postgresql, sql
- Domain
- databases, distributed-systems
- Issue type
- Bug
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100