Missing columns: 'a' while processing query with more than 2 tables in cross join

[UnamedRus]

How to reproduce
Clickhouse 21.5

WITH
    x AS
    (
        SELECT 1 AS a
    ),
    xx AS
    (
        SELECT *
        FROM x
        , x AS x1
    )
SELECT *
FROM xx
WHERE a = 1

Query id: f4fc159a-a136-4191-a0f0-c11c5ecdea74

┌─a─┬─x1.a─┐
│ 1 │    1 │
└───┴──────┘

WITH
    x AS
    (
        SELECT 1 AS a
    ),
    xx AS
    (
        SELECT *
        FROM x
        , x AS x1
        , x AS x2
    )
SELECT *
FROM xx
WHERE a = 1

Query id: bee7bab0-1f6a-4f29-bab5-f3b845e7095d


0 rows in set. Elapsed: 0.004 sec.

Received exception from server (version 21.5.1):
Code: 47. DB::Exception: Received from localhost:9000. DB::Exception: Missing columns: 'a' while processing query: 'WITH x AS (SELECT 1 AS a), xx AS (SELECT * FROM x , x AS x1 , x AS x2) SELECT `x.a`, `x1.a`, `x2.a` FROM xx WHERE a = 1', required columns: 'x.a' 'x1.a' 'x2.a' 'a' 'x.a' 'x1.a' 'x2.a' 'a'.

[adrian17]

Also reproducible with other join types than CROSS.

Example, reproducible on 21.3.4:

with example as (select toDate('2021-01-01') as date, 1 as node, 1 as user)
SELECT extra_data
FROM (
    SELECT join1.*
    FROM example
    LEFT JOIN ( SELECT toDate('2021-01-01') as date, 1 as extra_data ) AS join1
    ON example.date = join1.date
    LEFT JOIN ( SELECT toDate('2021-01-01') as date ) AS join2
    ON (example.date = join2.date)
)

Produces Missing columns: 'extra_data'. Removing the second JOIN appears to fix the issue.

[alexey-milovidov]

Also works perfectly with the analyzer! Will be easy to close after the analyzer is merged.