ClickHouse/tests/queries/0_stateless/01410_nullable_key_and_index.sql
Azat Khuzhin dee27fcbb9 Fix on-disk format breakage for secondary indices over Nullable column
[1] breaks on disk format (and the relevant change in the:

  [1]: https://github.com/ClickHouse/ClickHouse/pull/12455#discussion_r682830812

Too bad that I checked this patchset only for compatibility after
reverting this patch [2] (use case: I've applied it manually, then
revert it, and data skipping indexes over Nullable column had been
broken)

  [2]: https://github.com/ClickHouse/ClickHouse/pull/12455#issuecomment-823423772

But this patchset actually breaks compatibility with older versions of
clickhouse for Nullable data skipping indexes after simple upgrade:

Here is a simple reproducer:

    --
    -- run this with 21.6 or similar (i.e. w/o this patch)
    --

    CREATE TABLE data
    (
        `key` Int,
        `value` Nullable(Int),
        INDEX value_index value TYPE minmax GRANULARITY 1
    )
    ENGINE = MergeTree
    ORDER BY key;

    INSERT INTO data SELECT
        number,
        number
    FROM numbers(10000);

    SELECT * FROM data WHERE value = 20000 SETTINGS force_data_skipping_indices = 'value_index' SETTINGS force_data_skipping_indices = 'value_index', max_rows_to_read=1;

Now upgrade and run the query again:

    SELECT * FROM data WHERE value = 20000 SETTINGS force_data_skipping_indices = 'value_index' SETTINGS force_data_skipping_indices = 'value_index', max_rows_to_read=1;

And it will fail because of on disk format changes:

    $ ll --time-style=+ data/*/data/all_1_1_0/skp*.idx
    -rw-r----- 1 azat azat 36  data/with_nullable_patch/data/all_1_1_0/skp_idx_value_index.idx
    -rw-r----- 1 azat azat 37  data/without_nullable_patch/data/all_1_1_0/skp_idx_value_index.idx

    $ md5sum data/*/data/all_1_1_0/skp*.idx
    a19c95c4a14506c65665a1e30ab404bf  data/with_nullable_patch/data/all_1_1_0/skp_idx_value_index.idx
    e50e2fcfa873b232196623d56ab26105  data/without_nullable_patch/data/all_1_1_0/skp_idx_value_index.idx

Note, that there is no stable release with this patch included yet, so
no need to backport.

Also note that you may create data skipping indexes over Nullable
column even before [3].

  [3]: https://github.com/ClickHouse/ClickHouse/pull/12455

v2: break cases when granulas has Null in values due to backward
compatibility
2021-08-05 00:19:38 +03:00

66 lines
2.7 KiB
SQL

DROP TABLE IF EXISTS nullable_key;
DROP TABLE IF EXISTS nullable_key_without_final_mark;
DROP TABLE IF EXISTS nullable_minmax_index;
SET max_threads = 1;
CREATE TABLE nullable_key (k Nullable(int), v int) ENGINE MergeTree ORDER BY k SETTINGS allow_nullable_key = 1, index_granularity = 1;
INSERT INTO nullable_key SELECT number * 2, number * 3 FROM numbers(10);
INSERT INTO nullable_key SELECT NULL, -number FROM numbers(3);
SELECT * FROM nullable_key ORDER BY k;
SET force_primary_key = 1;
SET max_rows_to_read = 3;
SELECT * FROM nullable_key WHERE k IS NULL;
SET max_rows_to_read = 10;
SELECT * FROM nullable_key WHERE k IS NOT NULL;
SET max_rows_to_read = 5;
SELECT * FROM nullable_key WHERE k > 10;
SELECT * FROM nullable_key WHERE k < 10;
OPTIMIZE TABLE nullable_key FINAL;
SET max_rows_to_read = 4; -- one additional left mark needs to be read
SELECT * FROM nullable_key WHERE k IS NULL;
SET max_rows_to_read = 10;
SELECT * FROM nullable_key WHERE k IS NOT NULL;
-- Nullable in set and with transform_null_in = 1
SET max_rows_to_read = 3;
SELECT * FROM nullable_key WHERE k IN (10, 20) SETTINGS transform_null_in = 1;
SET max_rows_to_read = 5;
SELECT * FROM nullable_key WHERE k IN (3, NULL) SETTINGS transform_null_in = 1;
CREATE TABLE nullable_key_without_final_mark (s Nullable(String)) ENGINE MergeTree ORDER BY s SETTINGS allow_nullable_key = 1, write_final_mark = 0;
INSERT INTO nullable_key_without_final_mark VALUES ('123'), (NULL);
SET max_rows_to_read = 0;
SELECT * FROM nullable_key_without_final_mark WHERE s IS NULL;
SELECT * FROM nullable_key_without_final_mark WHERE s IS NOT NULL;
CREATE TABLE nullable_minmax_index (k int, v Nullable(int), INDEX v_minmax v TYPE minmax GRANULARITY 4) ENGINE MergeTree ORDER BY k SETTINGS index_granularity = 1;
INSERT INTO nullable_minmax_index VALUES (1, 3), (2, 7), (3, 4), (2, NULL); -- [3, +Inf]
INSERT INTO nullable_minmax_index VALUES (1, 1), (2, 2), (3, 2), (2, 1); -- [1, 2]
INSERT INTO nullable_minmax_index VALUES (2, NULL), (3, NULL); -- [+Inf, +Inf]
SET force_primary_key = 0;
SELECT * FROM nullable_minmax_index ORDER BY k;
SET max_rows_to_read = 6;
SELECT * FROM nullable_minmax_index WHERE v IS NULL;
-- NOTE: granuals with Null values cannot be filtred in data skipping indexes,
-- due to backward compatibility
SET max_rows_to_read = 0;
SELECT * FROM nullable_minmax_index WHERE v IS NOT NULL;
SET max_rows_to_read = 6;
SELECT * FROM nullable_minmax_index WHERE v > 2;
-- NOTE: granuals with Null values cannot be filtred in data skipping indexes,
-- due to backward compatibility
SET max_rows_to_read = 0;
SELECT * FROM nullable_minmax_index WHERE v <= 2;
DROP TABLE nullable_key;
DROP TABLE nullable_key_without_final_mark;
DROP TABLE nullable_minmax_index;