Skip to content

Flink: fix out of order appends in upsert UT #4532

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
openinx merged 1 commit into from
Apr 11, 2022
Merged

Flink: fix out of order appends in upsert UT #4532

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
40 changes: 20 additions & 20 deletions flink/v1.13/flink/src/test/java/org/apache/iceberg/flink/TestFlinkUpsert.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -130,15 +130,15 @@ public void testUpsertAndQuery() {

try {
sql("INSERT INTO %s VALUES " +
"(1, 'a', TO_DATE('2022-03-01'))," +
"(2, 'b', TO_DATE('2022-03-01'))," +
"(1, 'b', TO_DATE('2022-03-01'))",
"(1, 'a', DATE '2022-03-01')," +
"(2, 'b', DATE '2022-03-01')," +
"(1, 'b', DATE '2022-03-01')",
tableName);

sql("INSERT INTO %s VALUES " +
"(4, 'a', TO_DATE('2022-03-02'))," +
"(5, 'b', TO_DATE('2022-03-02'))," +
"(1, 'b', TO_DATE('2022-03-02'))",
"(4, 'a', DATE '2022-03-02')," +
"(5, 'b', DATE '2022-03-02')," +
"(1, 'b', DATE '2022-03-02')",
tableName);

List<Row> rowsOn20220301 = Lists.newArrayList(Row.of(2, "b", dt20220301), Row.of(1, "a", dt20220301));
Expand Down Expand Up @@ -211,27 +211,27 @@ public void testPrimaryKeyFieldsAtBeginningOfSchema() {
tableName, toWithClause(tableUpsertProps));

sql("INSERT INTO %s VALUES " +
"('aaa', TO_DATE('2022-03-01'), 1)," +
"('aaa', TO_DATE('2022-03-01'), 2)," +
"('bbb', TO_DATE('2022-03-01'), 3)",
"('aaa', DATE '2022-03-01', 1)," +
"('aaa', DATE '2022-03-01', 2)," +
"('bbb', DATE '2022-03-01', 3)",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of("aaa", dt, 2), Row.of("bbb", dt, 3)));

sql("INSERT INTO %s VALUES " +
"('aaa', TO_DATE('2022-03-01'), 4)," +
"('bbb', TO_DATE('2022-03-01'), 5)",
"('aaa', DATE '2022-03-01', 4)," +
"('bbb', DATE '2022-03-01', 5)",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of("aaa", dt, 4), Row.of("bbb", dt, 5)));

sql("INSERT INTO %s VALUES " +
"('aaa', TO_DATE('2022-03-01'), 6)," +
"('bbb', TO_DATE('2022-03-01'), 7)",
"('aaa', DATE '2022-03-01', 6)," +
"('bbb', DATE '2022-03-01', 7)",
tableName);

TestHelpers.assertRows(
Expand All @@ -254,27 +254,27 @@ public void testPrimaryKeyFieldsAtEndOfTableSchema() {
tableName, toWithClause(tableUpsertProps));

sql("INSERT INTO %s VALUES " +
"(1, 'aaa', TO_DATE('2022-03-01'))," +
"(2, 'aaa', TO_DATE('2022-03-01'))," +
"(3, 'bbb', TO_DATE('2022-03-01'))",
"(1, 'aaa', DATE '2022-03-01')," +
"(2, 'aaa', DATE '2022-03-01')," +
"(3, 'bbb', DATE '2022-03-01')",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of(2, "aaa", dt), Row.of(3, "bbb", dt)));

sql("INSERT INTO %s VALUES " +
"(4, 'aaa', TO_DATE('2022-03-01'))," +
"(5, 'bbb', TO_DATE('2022-03-01'))",
"(4, 'aaa', DATE '2022-03-01')," +
"(5, 'bbb', DATE '2022-03-01')",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of(4, "aaa", dt), Row.of(5, "bbb", dt)));

sql("INSERT INTO %s VALUES " +
"(6, 'aaa', TO_DATE('2022-03-01'))," +
"(7, 'bbb', TO_DATE('2022-03-01'))",
"(6, 'aaa', DATE '2022-03-01')," +
"(7, 'bbb', DATE '2022-03-01')",
tableName);

TestHelpers.assertRows(
Expand Down
40 changes: 20 additions & 20 deletions flink/v1.14/flink/src/test/java/org/apache/iceberg/flink/TestFlinkUpsert.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -130,15 +130,15 @@ public void testUpsertAndQuery() {

try {
sql("INSERT INTO %s VALUES " +
"(1, 'a', TO_DATE('2022-03-01'))," +
"(2, 'b', TO_DATE('2022-03-01'))," +
"(1, 'b', TO_DATE('2022-03-01'))",
"(1, 'a', DATE '2022-03-01')," +
"(2, 'b', DATE '2022-03-01')," +
"(1, 'b', DATE '2022-03-01')",
tableName);

sql("INSERT INTO %s VALUES " +
"(4, 'a', TO_DATE('2022-03-02'))," +
"(5, 'b', TO_DATE('2022-03-02'))," +
"(1, 'b', TO_DATE('2022-03-02'))",
"(4, 'a', DATE '2022-03-02')," +
"(5, 'b', DATE '2022-03-02')," +
"(1, 'b', DATE '2022-03-02')",
Comment on lines +139 to +141
Copy link
Contributor

@kbendick kbendick Apr 11, 2022 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can we addd a comment about the usage of DATE vs TO_DATE, given that we're relying on DATE to affect the generated job graph differently than using TO_DATE does, which affects ordering etc?

A short comment and then directing to this comment might be sufficent: #4515 (comment)

Copy link
Contributor Author

@yittg yittg Apr 11, 2022 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the commit message is a more proper place to explain it.
I'll split this PR into two parts as you mentioned, and describe the reason in the commit message of the first.
WDYT @kbendick ?

Copy link
Contributor

@kbendick kbendick Apr 11, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

+1. As well as a proper description in the PR comment (which can be the same). As those are the two most commonly searched places when reviewing older things.

Thanks again for the great work.

tableName);

List<Row> rowsOn20220301 = Lists.newArrayList(Row.of(2, "b", dt20220301), Row.of(1, "a", dt20220301));
Expand Down Expand Up @@ -211,27 +211,27 @@ public void testPrimaryKeyFieldsAtBeginningOfSchema() {
tableName, toWithClause(tableUpsertProps));

sql("INSERT INTO %s VALUES " +
"('aaa', TO_DATE('2022-03-01'), 1)," +
"('aaa', TO_DATE('2022-03-01'), 2)," +
"('bbb', TO_DATE('2022-03-01'), 3)",
"('aaa', DATE '2022-03-01', 1)," +
"('aaa', DATE '2022-03-01', 2)," +
"('bbb', DATE '2022-03-01', 3)",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of("aaa", dt, 2), Row.of("bbb", dt, 3)));

sql("INSERT INTO %s VALUES " +
"('aaa', TO_DATE('2022-03-01'), 4)," +
"('bbb', TO_DATE('2022-03-01'), 5)",
"('aaa', DATE '2022-03-01', 4)," +
"('bbb', DATE '2022-03-01', 5)",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of("aaa", dt, 4), Row.of("bbb", dt, 5)));

sql("INSERT INTO %s VALUES " +
"('aaa', TO_DATE('2022-03-01'), 6)," +
"('bbb', TO_DATE('2022-03-01'), 7)",
"('aaa', DATE '2022-03-01', 6)," +
"('bbb', DATE '2022-03-01', 7)",
tableName);

TestHelpers.assertRows(
Expand All @@ -254,27 +254,27 @@ public void testPrimaryKeyFieldsAtEndOfTableSchema() {
tableName, toWithClause(tableUpsertProps));

sql("INSERT INTO %s VALUES " +
"(1, 'aaa', TO_DATE('2022-03-01'))," +
"(2, 'aaa', TO_DATE('2022-03-01'))," +
"(3, 'bbb', TO_DATE('2022-03-01'))",
"(1, 'aaa', DATE '2022-03-01')," +
"(2, 'aaa', DATE '2022-03-01')," +
"(3, 'bbb', DATE '2022-03-01')",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of(2, "aaa", dt), Row.of(3, "bbb", dt)));

sql("INSERT INTO %s VALUES " +
"(4, 'aaa', TO_DATE('2022-03-01'))," +
"(5, 'bbb', TO_DATE('2022-03-01'))",
"(4, 'aaa', DATE '2022-03-01')," +
"(5, 'bbb', DATE '2022-03-01')",
tableName);

TestHelpers.assertRows(
sql("SELECT * FROM %s", tableName),
Lists.newArrayList(Row.of(4, "aaa", dt), Row.of(5, "bbb", dt)));

sql("INSERT INTO %s VALUES " +
"(6, 'aaa', TO_DATE('2022-03-01'))," +
"(7, 'bbb', TO_DATE('2022-03-01'))",
"(6, 'aaa', DATE '2022-03-01')," +
"(7, 'bbb', DATE '2022-03-01')",
tableName);

TestHelpers.assertRows(
Expand Down