This is an automated email from the ASF dual-hosted git repository.
mchades pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/gravitino.git
The following commit(s) were added to refs/heads/main by this push:
new c696fed0dd [#9515] fix(catalog-starrocks): fallback to NONE when no
DISTRIBUTED BY (#11239)
c696fed0dd is described below
commit c696fed0ddc11d4e32a71b9a8229d887d9f126ac
Author: mchades <[email protected]>
AuthorDate: Wed May 27 17:29:39 2026 +0800
[#9515] fix(catalog-starrocks): fallback to NONE when no DISTRIBUTED BY
(#11239)
### What changes were proposed in this pull request?
This PR fixes StarRocks distribution parsing when loading tables whose
`SHOW CREATE TABLE` SQL does not include a valid `DISTRIBUTED BY` clause
(for example, external/Paimon-like table definitions).
- In `StarRocksUtils.extractDistributionInfoFromSql`, return
`Distributions.NONE` when a top-level distribution clause is absent.
- Keep strict behavior when a distribution clause exists but cannot be
parsed (still throw runtime exception).
- Tighten distribution-clause presence detection to top-level clause
boundaries and strategy shape (`RANDOM` or `strategy(`), to avoid false
positives from column comments containing `DISTRIBUTED BY`.
- Extend existing `TestStarRocksUtils.testDistributedInfoPattern` with:
- a no-distribution SQL case expecting `Distributions.NONE`
- a column-comment case containing `DISTRIBUTED BY` that still expects
`Distributions.NONE`
- an invalid distribution clause case expecting exception
### Why are the changes needed?
Issue #9515 reports table load failures with:
`Failed to extract distribution info in sql: CREATE TABLE ... PROPERTIES
(...)`
The root cause is that current StarRocks loading logic always parses
distribution info, while some table SQL does not contain a valid
top-level `DISTRIBUTED BY` clause. This patch makes that scenario
loadable while preserving fail-fast behavior for malformed distribution
syntax.
Fix: #9515
### Does this PR introduce _any_ user-facing change?
- StarRocks table loading no longer fails for table SQL that does not
define a valid top-level `DISTRIBUTED BY` clause.
- False positives caused by `DISTRIBUTED BY` appearing in column
comments are avoided.
- No API changes.
- No property key additions/removals.
### How was this patch tested?
- Formatting:
- `./gradlew :catalogs:catalog-jdbc-starrocks:spotlessApply`
- Unit test:
- `./gradlew :catalogs:catalog-jdbc-starrocks:test --tests
org.apache.gravitino.catalog.starrocks.utils.TestStarRocksUtils
-PskipITs`
- Result: BUILD SUCCESSFUL.
---------
Co-authored-by: Copilot Autofix powered by AI
<[email protected]>
---
.../catalog/starrocks/utils/StarRocksUtils.java | 13 +++++++++++++
.../catalog/starrocks/utils/TestStarRocksUtils.java | 20 ++++++++++++++++++++
2 files changed, 33 insertions(+)
diff --git
a/catalogs/catalog-jdbc-starrocks/src/main/java/org/apache/gravitino/catalog/starrocks/utils/StarRocksUtils.java
b/catalogs/catalog-jdbc-starrocks/src/main/java/org/apache/gravitino/catalog/starrocks/utils/StarRocksUtils.java
index 75edc0971f..f97e2e316f 100644
---
a/catalogs/catalog-jdbc-starrocks/src/main/java/org/apache/gravitino/catalog/starrocks/utils/StarRocksUtils.java
+++
b/catalogs/catalog-jdbc-starrocks/src/main/java/org/apache/gravitino/catalog/starrocks/utils/StarRocksUtils.java
@@ -57,6 +57,15 @@ public class StarRocksUtils {
Pattern.compile(
"DISTRIBUTED
BY\\s+(HASH|RANDOM)\\s*(\\(([^)]+)\\))?\\s*(BUCKETS\\s+(\\d+))?");
+ // Match DISTRIBUTED BY clause presence in a formatting-independent way.
+ // Accept any whitespace before DISTRIBUTED so single-line SQL, CRLF line
endings,
+ // and clauses following other table options (for example ENGINE=...) are
detected.
+ // Keep the strategy-shape guard so RANDOM has no column list while others
are
+ // still expected to start with "strategy(".
+ private static final Pattern DISTRIBUTED_BY_CLAUSE_PATTERN =
+ Pattern.compile(
+ "(?:^|\\s|\\))DISTRIBUTED\\s+BY\\s+(?:RANDOM\\b|\\w+\\s*\\()",
Pattern.CASE_INSENSITIVE);
+
private static final Pattern TABLE_COMMENT_PATTERN =
Pattern.compile("COMMENT\\s*\"([^\\(]+?)\\s*\\(From Gravitino,.*\\)\"");
@@ -177,6 +186,10 @@ public class StarRocksUtils {
.build();
}
+ if (!DISTRIBUTED_BY_CLAUSE_PATTERN.matcher(createTableSql).find()) {
+ return Distributions.NONE;
+ }
+
throw new RuntimeException("Failed to extract distribution info in sql:" +
createTableSql);
}
diff --git
a/catalogs/catalog-jdbc-starrocks/src/test/java/org/apache/gravitino/catalog/starrocks/utils/TestStarRocksUtils.java
b/catalogs/catalog-jdbc-starrocks/src/test/java/org/apache/gravitino/catalog/starrocks/utils/TestStarRocksUtils.java
index f1e86abfb9..9f77b51748 100644
---
a/catalogs/catalog-jdbc-starrocks/src/test/java/org/apache/gravitino/catalog/starrocks/utils/TestStarRocksUtils.java
+++
b/catalogs/catalog-jdbc-starrocks/src/test/java/org/apache/gravitino/catalog/starrocks/utils/TestStarRocksUtils.java
@@ -27,6 +27,7 @@ import java.util.HashMap;
import java.util.Map;
import java.util.Optional;
import org.apache.gravitino.rel.expressions.distributions.Distribution;
+import org.apache.gravitino.rel.expressions.distributions.Distributions;
import org.apache.gravitino.rel.expressions.literals.Literal;
import org.apache.gravitino.rel.expressions.literals.Literals;
import org.apache.gravitino.rel.expressions.transforms.Transform;
@@ -34,6 +35,7 @@ import
org.apache.gravitino.rel.expressions.transforms.Transforms;
import org.apache.gravitino.rel.partitions.Partition;
import org.apache.gravitino.rel.partitions.Partitions;
import org.apache.gravitino.rel.types.Types;
+import org.junit.jupiter.api.Assertions;
import org.junit.jupiter.api.Test;
public class TestStarRocksUtils {
@@ -205,5 +207,23 @@ public class TestStarRocksUtils {
Distribution distribution2 =
StarRocksUtils.extractDistributionInfoFromSql(createTableSqlWithAuto);
assertEquals(distribution2.number(), -1);
+
+ String createTableSqlWithoutDistribution =
+ "CREATE TABLE `testTable` (\n`col1` date NOT NULL\n) PROPERTIES
(\"location\" = \"hdfs://path/table\")";
+ Distribution distribution3 =
+
StarRocksUtils.extractDistributionInfoFromSql(createTableSqlWithoutDistribution);
+ assertEquals(Distributions.NONE.strategy(), distribution3.strategy());
+
+ String createTableSqlWithKeywordInColumnComment =
+ "CREATE TABLE `testTable` (\n`col1` date NOT NULL COMMENT \"contains
DISTRIBUTED BY in comment\"\n) PROPERTIES (\"location\" =
\"hdfs://path/table\")";
+ Distribution distribution4 =
+
StarRocksUtils.extractDistributionInfoFromSql(createTableSqlWithKeywordInColumnComment);
+ assertEquals(Distributions.NONE.strategy(), distribution4.strategy());
+
+ String createTableSqlWithInvalidDistribution =
+ "CREATE TABLE `testTable` (\n`col1` date NOT NULL\n) DISTRIBUTED BY
INVALID(`col1`) BUCKETS 2";
+ Assertions.assertThrows(
+ RuntimeException.class,
+ () ->
StarRocksUtils.extractDistributionInfoFromSql(createTableSqlWithInvalidDistribution));
}
}
