Skip to content

Commit 8e4bbd4

Browse files
martykulmamartykulma
committed
Add tests for unsupported syntax
1 parent 603554c commit 8e4bbd4

File tree

4 files changed

+390
-0
lines changed

4 files changed

+390
-0
lines changed

‎test/testdrive-old-kafka-src-syntax/load-generator-key-value.td‎

Lines changed: 298 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,298 @@
1+
# Copyright Materialize, Inc. and contributors. All rights reserved.
2+
#
3+
# Use of this software is governed by the Business Source License
4+
# included in the LICENSE file at the root of this repository.
5+
#
6+
# As of the Change Date specified in that file, in accordance with
7+
# the Business Source License, use of this software will be governed
8+
# by the Apache License, Version 2.0.
9+
10+
# Tests `LOAD GENERATOR KEY VALUE`
11+
12+
$ set-arg-default default-replica-size=1
13+
14+
$ postgres-execute connection=postgres://mz_system:materialize@${testdrive.materialize-internal-sql-addr}
15+
ALTER SYSTEM SET storage_statistics_collection_interval = 1000
16+
ALTER SYSTEM SET storage_statistics_interval = 2000
17+
ALTER SYSTEM SET enable_load_generator_key_value = true
18+
19+
> CREATE CLUSTER lg_cluster SIZE '${arg.default-replica-size}'
20+
21+
$ skip-if
22+
SELECT mz_version_num() < 13900;
23+
24+
# Error if trying to create with subsources
25+
! CREATE SOURCE g FROM LOAD GENERATOR KEY VALUE(
26+
KEYS 8,
27+
PARTITIONS 1,
28+
SNAPSHOT ROUNDS 1,
29+
VALUE SIZE 1,
30+
SEED 42,
31+
BATCH SIZE 4
32+
) FOR ALL TABLES;
33+
contains:FOR ALL TABLES
34+
35+
! CREATE SOURCE g FROM LOAD GENERATOR KEY VALUE(
36+
KEYS 8,
37+
PARTITIONS 1,
38+
SNAPSHOT ROUNDS 1,
39+
VALUE SIZE 1,
40+
SEED 42,
41+
BATCH SIZE 4
42+
) FOR TABLES ("foo");
43+
contains:FOR TABLES
44+
45+
! CREATE SOURCE g FROM LOAD GENERATOR KEY VALUE(
46+
KEYS 8,
47+
PARTITIONS 1,
48+
SNAPSHOT ROUNDS 1,
49+
VALUE SIZE 1,
50+
SEED 42,
51+
BATCH SIZE 4
52+
) FOR SCHEMAS ("foo");
53+
contains:FOR SCHEMAS
54+
55+
$ skip-end
56+
57+
# A loadgen that only snapshots.
58+
> CREATE SOURCE up_no_update
59+
IN CLUSTER lg_cluster
60+
FROM LOAD GENERATOR KEY VALUE (
61+
KEYS 16,
62+
PARTITIONS 4,
63+
SNAPSHOT ROUNDS 3,
64+
SEED 123,
65+
VALUE SIZE 10,
66+
BATCH SIZE 2
67+
)
68+
ENVELOPE UPSERT
69+
70+
> CREATE SOURCE up_quick
71+
IN CLUSTER lg_cluster
72+
FROM LOAD GENERATOR KEY VALUE (
73+
KEYS 16,
74+
PARTITIONS 4,
75+
SNAPSHOT ROUNDS 3,
76+
TRANSACTIONAL SNAPSHOT false,
77+
SEED 123,
78+
VALUE SIZE 10,
79+
BATCH SIZE 2
80+
)
81+
INCLUDE KEY AS whatever
82+
ENVELOPE UPSERT
83+
84+
# Ensure data is spread as expected.
85+
> SELECT partition, count(*) FROM up_no_update GROUP BY partition
86+
0 4
87+
1 4
88+
2 4
89+
3 4
90+
91+
92+
> SELECT MAX(key) FROM up_no_update;
93+
15
94+
95+
> SELECT partition, count(*) FROM up_quick GROUP BY partition
96+
0 4
97+
1 4
98+
2 4
99+
3 4
100+
101+
> SELECT MAX(whatever) FROM up_quick;
102+
15
103+
104+
# 48 values produced (3 snapshot rounds with 16 keys).
105+
# For the `TRANSACTIONAL SNAPSHOT = false` source, we produce 48 updates from the 3 rounds.
106+
# We expect and 6 quick round offsets (based on the batch size)
107+
> SELECT
108+
s.name,
109+
u.offset_known,
110+
u.offset_committed,
111+
u.snapshot_records_known,
112+
u.snapshot_records_staged,
113+
u.messages_received,
114+
u.records_indexed
115+
FROM mz_sources s
116+
JOIN mz_internal.mz_source_statistics u ON s.id = u.id
117+
WHERE s.name IN ('up_no_update', 'up_quick')
118+
up_no_update 3 3 48 48 48 16
119+
up_quick 6 6 0 0 48 16
120+
121+
$ set-from-sql var=pre-rehydration
122+
SELECT
123+
encode(value, 'base64')
124+
FROM up_no_update
125+
WHERE
126+
key = 14
127+
128+
$ set-from-sql var=pre-rehydration-quick
129+
SELECT
130+
encode(value, 'base64')
131+
FROM up_quick
132+
WHERE
133+
whatever = 14
134+
135+
> ALTER CLUSTER lg_cluster SET (REPLICATION FACTOR 0);
136+
137+
> ALTER CLUSTER lg_cluster SET (REPLICATION FACTOR 1);
138+
139+
# Ensure that we rehydrate and keep the same value as before.
140+
> SELECT
141+
encode(value, 'base64') = '${pre-rehydration}'
142+
FROM up_no_update
143+
WHERE
144+
key = 14
145+
true
146+
147+
> SELECT
148+
encode(value, 'base64') = '${pre-rehydration-quick}'
149+
FROM up_quick
150+
WHERE
151+
whatever = 14
152+
true
153+
154+
> SELECT
155+
s.name,
156+
u.offset_known,
157+
u.offset_committed,
158+
u.snapshot_records_known,
159+
u.snapshot_records_staged,
160+
u.messages_received,
161+
u.records_indexed
162+
FROM mz_sources s
163+
JOIN mz_internal.mz_source_statistics u ON s.id = u.id
164+
WHERE s.name IN ('up_no_update', 'up_quick')
165+
up_no_update 3 3 48 48 48 16
166+
up_quick 6 6 0 0 48 16
167+
168+
> DROP SOURCE up_no_update
169+
> DROP SOURCE up_quick
170+
171+
# Create a source with 1s updates after snapshotting.
172+
> CREATE SOURCE up_with_update
173+
IN CLUSTER lg_cluster
174+
FROM LOAD GENERATOR KEY VALUE (
175+
KEYS 16,
176+
PARTITIONS 4,
177+
SNAPSHOT ROUNDS 3,
178+
SEED 123,
179+
VALUE SIZE 10,
180+
BATCH SIZE 2,
181+
TICK INTERVAL '1s'
182+
)
183+
ENVELOPE UPSERT
184+
185+
# Ensure data is partitioned correctly.
186+
> SELECT partition, count(*) FROM up_with_update GROUP BY partition
187+
0 4
188+
1 4
189+
2 4
190+
3 4
191+
192+
# Doesn't work reliably under high load in CI
193+
# Higher offsets than before, as we produce more values.
194+
# > SELECT
195+
# s.name,
196+
# u.offset_known > 3,
197+
# u.offset_committed = u.offset_known,
198+
# u.snapshot_records_known,
199+
# u.snapshot_records_staged,
200+
# u.messages_received > 48,
201+
# u.records_indexed
202+
# FROM mz_sources s
203+
# JOIN mz_internal.mz_source_statistics u ON s.id = u.id
204+
# WHERE s.name IN ('up_with_update')
205+
# up_with_update true true 48 48 true 16
206+
207+
# Also, despite the same seed, values should be different than the snapshot-only source.
208+
> SELECT
209+
encode(value, 'base64') != '${pre-rehydration}'
210+
FROM up_with_update
211+
WHERE
212+
key = 14
213+
true
214+
215+
> ALTER CLUSTER lg_cluster SET (REPLICATION FACTOR 0);
216+
217+
> ALTER CLUSTER lg_cluster SET (REPLICATION FACTOR 1);
218+
219+
$ set-from-sql var=pre-rehydration-with-update
220+
SELECT
221+
encode(value, 'base64')
222+
FROM up_with_update
223+
WHERE
224+
key = 14
225+
226+
# After restarting, we should also still see new updates override values.
227+
> SELECT
228+
encode(value, 'base64') != '${pre-rehydration-with-update}'
229+
FROM up_with_update
230+
WHERE
231+
key = 14
232+
true
233+
234+
# Test NONE-envelope
235+
> CREATE SOURCE kv_none
236+
IN CLUSTER lg_cluster
237+
FROM LOAD GENERATOR KEY VALUE (
238+
KEYS 16,
239+
PARTITIONS 4,
240+
SNAPSHOT ROUNDS 3,
241+
TRANSACTIONAL SNAPSHOT false,
242+
SEED 123,
243+
VALUE SIZE 10,
244+
BATCH SIZE 2
245+
)
246+
ENVELOPE NONE
247+
248+
> SELECT partition, count(*) FROM kv_none GROUP BY partition
249+
0 12
250+
1 12
251+
2 12
252+
3 12
253+
254+
# Test INCLUDE OFFSET
255+
> CREATE SOURCE kv_offset
256+
IN CLUSTER lg_cluster
257+
FROM LOAD GENERATOR KEY VALUE (
258+
KEYS 16,
259+
PARTITIONS 4,
260+
SNAPSHOT ROUNDS 3,
261+
TRANSACTIONAL SNAPSHOT false,
262+
SEED 123,
263+
VALUE SIZE 10,
264+
BATCH SIZE 2
265+
)
266+
INCLUDE OFFSET
267+
ENVELOPE NONE
268+
269+
> SELECT partition, MAX("offset") FROM kv_offset GROUP BY partition
270+
0 5
271+
1 5
272+
2 5
273+
3 5
274+
275+
> CREATE SOURCE kv_offset2
276+
IN CLUSTER lg_cluster
277+
FROM LOAD GENERATOR KEY VALUE (
278+
KEYS 16,
279+
PARTITIONS 4,
280+
SNAPSHOT ROUNDS 3,
281+
TRANSACTIONAL SNAPSHOT false,
282+
SEED 123,
283+
VALUE SIZE 10,
284+
BATCH SIZE 2
285+
)
286+
INCLUDE
287+
OFFSET AS something_else,
288+
KEY AS whatever
289+
ENVELOPE NONE
290+
291+
> SELECT partition, MAX(something_else) FROM kv_offset2 GROUP BY partition
292+
0 5
293+
1 5
294+
2 5
295+
3 5
296+
297+
> SELECT MAX(whatever) FROM kv_offset2;
298+
15

‎test/testdrive-old-kafka-src-syntax/load-generator.td‎

Lines changed: 33 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -50,6 +50,39 @@ exact:COUNTER load generators do not support SCALE FACTOR values
5050
IN CLUSTER ${arg.single-replica-cluster}
5151
FROM LOAD GENERATOR AUCTION (AS OF 300, UP TO 301);
5252

53+
$ skip-if
54+
SELECT mz_version_num() < 13900;
55+
56+
# Error if trying to create with subsources
57+
! CREATE SOURCE g FROM LOAD GENERATOR COUNTER FOR ALL TABLES;
58+
contains:FOR ALL TABLES
59+
60+
! CREATE SOURCE g FROM LOAD GENERATOR CLOCK FOR ALL TABLES;
61+
contains:FOR ALL TABLES
62+
63+
! CREATE SOURCE g FROM LOAD GENERATOR DATUMS FOR ALL TABLES;
64+
contains:FOR ALL TABLES
65+
66+
! CREATE SOURCE g FROM LOAD GENERATOR COUNTER FOR TABLES ("foo");
67+
contains:FOR TABLES
68+
69+
! CREATE SOURCE g FROM LOAD GENERATOR CLOCK FOR TABLES ("foo");
70+
contains:FOR TABLES
71+
72+
! CREATE SOURCE g FROM LOAD GENERATOR DATUMS FOR TABLES ("foo");
73+
contains:FOR TABLES
74+
75+
! CREATE SOURCE g FROM LOAD GENERATOR COUNTER FOR SCHEMAS ("foo");
76+
contains:FOR SCHEMAS
77+
78+
! CREATE SOURCE g FROM LOAD GENERATOR CLOCK FOR SCHEMAS ("foo");
79+
contains:FOR SCHEMAS
80+
81+
! CREATE SOURCE g FROM LOAD GENERATOR DATUMS FOR SCHEMAS ("foo");
82+
contains:FOR SCHEMAS
83+
84+
# skip-end
85+
5386
> CREATE TABLE accounts FROM SOURCE auction_house (REFERENCE accounts);
5487
> CREATE TABLE auctions FROM SOURCE auction_house (REFERENCE auctions);
5588
> CREATE TABLE bids FROM SOURCE auction_house (REFERENCE bids);

‎test/testdrive/load-generator-key-value.td‎

Lines changed: 31 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -18,6 +18,37 @@ ALTER SYSTEM SET enable_load_generator_key_value = true
1818

1919
> CREATE CLUSTER lg_cluster SIZE '${arg.default-replica-size}'
2020

21+
# Error if trying to create with subsources
22+
! CREATE SOURCE g FROM LOAD GENERATOR KEY VALUE(
23+
KEYS 8,
24+
PARTITIONS 1,
25+
SNAPSHOT ROUNDS 1,
26+
VALUE SIZE 1,
27+
SEED 42,
28+
BATCH SIZE 4
29+
) FOR ALL TABLES;
30+
contains:FOR ALL TABLES
31+
32+
! CREATE SOURCE g FROM LOAD GENERATOR KEY VALUE(
33+
KEYS 8,
34+
PARTITIONS 1,
35+
SNAPSHOT ROUNDS 1,
36+
VALUE SIZE 1,
37+
SEED 42,
38+
BATCH SIZE 4
39+
) FOR TABLES ("foo");
40+
contains:FOR TABLES
41+
42+
! CREATE SOURCE g FROM LOAD GENERATOR KEY VALUE(
43+
KEYS 8,
44+
PARTITIONS 1,
45+
SNAPSHOT ROUNDS 1,
46+
VALUE SIZE 1,
47+
SEED 42,
48+
BATCH SIZE 4
49+
) FOR SCHEMAS ("foo");
50+
contains:FOR SCHEMAS
51+
2152
# A loadgen that only snapshots.
2253
> CREATE SOURCE up_no_update
2354
IN CLUSTER lg_cluster

0 commit comments

Comments
 (0)