You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@paimon.apache.org by "steFaiz (via GitHub)" <gi...@apache.org> on 2026/03/30 04:25:41 UTC
[PR] [flink] support vector search procedure for flink [paimon]
steFaiz opened a new pull request, #7550:
URL: https://github.com/apache/paimon/pull/7550
### Purpose
Flink supports vector search procedure. This is a flink version of https://github.com/apache/paimon/pull/6950
Usage:
``` sql
CALL sys.vector_search(
`table` => 'tableId',
vector_column => 'v',
query_vector => '1.0,2.0,3.0',
top_k => 5
)
```
Note that flink 2.0 has already supported native vector search, see [flink vector search](https://nightlies.apache.org/flink/flink-docs-release-2.2/docs/dev/table/sql/queries/vector-search/). It can cover this procedure through passing a constant vector, like:
``` sql
-- Searching with contant value
SELECT *
FROM TABLE(VECTOR_SEARCH(
TABLE vector_table,
ARRAY[10, 20],
DESCRIPTOR(index_column),
10,
));
```
We could support it in the future.
### Tests
see `org.apache.paimon.flink.procedure.VectorSearchProcedureITCase`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "steFaiz (via GitHub)" <gi...@apache.org>.
steFaiz commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008883614
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
Review Comment:
Thanks! I've changed it.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "JingsongLi (via GitHub)" <gi...@apache.org>.
JingsongLi commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008584073
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
Review Comment:
ProcedureBase already provides a table(String tableId) method that does the same thing
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "steFaiz (via GitHub)" <gi...@apache.org>.
steFaiz closed pull request #7550: [flink] support vector search procedure for flink
URL: https://github.com/apache/paimon/pull/7550
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "JingsongLi (via GitHub)" <gi...@apache.org>.
JingsongLi commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008573523
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
+ String col = columns[i].trim();
+ int idx = fieldNames.indexOf(col);
+ if (idx < 0) {
+ throw new IllegalArgumentException(
+ "Column '" + col + "' not found in table schema: " + fieldNames);
+ }
+ indices[i] = idx;
+ }
+ return indices;
+ }
+
+ private static RowType buildProjectedRowType(RowType rowType, int[] projection) {
+ List<DataField> fields = rowType.getFields();
Review Comment:
rowType.project
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "JingsongLi (via GitHub)" <gi...@apache.org>.
JingsongLi commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008587423
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
+ String col = columns[i].trim();
+ int idx = fieldNames.indexOf(col);
+ if (idx < 0) {
+ throw new IllegalArgumentException(
+ "Column '" + col + "' not found in table schema: " + fieldNames);
+ }
+ indices[i] = idx;
+ }
+ return indices;
+ }
+
+ private static RowType buildProjectedRowType(RowType rowType, int[] projection) {
+ List<DataField> fields = rowType.getFields();
+ List<DataField> projected = new ArrayList<>(projection.length);
+ for (int idx : projection) {
+ projected.add(fields.get(idx));
+ }
+ return new RowType(projected);
+ }
+
+ // --- Row to JSON conversion (reuses the same pattern as JsonFormatWriter) ---
+
+ private static String convertRowToJsonString(InternalRow row, RowType rowType)
+ throws Exception {
+ Map<String, Object> result = convertRowToMap(row, rowType);
+ return JsonSerdeUtil.writeValueAsString(result);
+ }
+
+ private static Map<String, Object> convertRowToMap(InternalRow row, RowType rowType) {
+ List<DataField> fields = rowType.getFields();
+ Map<String, Object> result = new LinkedHashMap<>(fields.size());
+ for (int i = 0; i < fields.size(); i++) {
+ DataField field = fields.get(i);
+ Object value = InternalRowUtils.get(row, i, field.type());
+ result.put(field.name(), convertValue(value, field.type()));
+ }
+ return result;
+ }
+
+ private static Object convertValue(Object value, DataType dataType) {
+ if (value == null) {
+ return null;
+ }
+ switch (dataType.getTypeRoot()) {
+ case INTEGER:
+ case BIGINT:
+ case FLOAT:
+ case DOUBLE:
+ case BOOLEAN:
+ case TINYINT:
+ case SMALLINT:
+ case CHAR:
+ case VARCHAR:
+ return value.toString();
Review Comment:
Can you use jackson to do correct json serialization?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "JingsongLi (via GitHub)" <gi...@apache.org>.
JingsongLi commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008592314
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
+ String col = columns[i].trim();
+ int idx = fieldNames.indexOf(col);
+ if (idx < 0) {
+ throw new IllegalArgumentException(
+ "Column '" + col + "' not found in table schema: " + fieldNames);
+ }
+ indices[i] = idx;
+ }
+ return indices;
+ }
+
+ private static RowType buildProjectedRowType(RowType rowType, int[] projection) {
+ List<DataField> fields = rowType.getFields();
+ List<DataField> projected = new ArrayList<>(projection.length);
+ for (int idx : projection) {
+ projected.add(fields.get(idx));
+ }
+ return new RowType(projected);
+ }
+
+ // --- Row to JSON conversion (reuses the same pattern as JsonFormatWriter) ---
+
+ private static String convertRowToJsonString(InternalRow row, RowType rowType)
+ throws Exception {
+ Map<String, Object> result = convertRowToMap(row, rowType);
+ return JsonSerdeUtil.writeValueAsString(result);
+ }
+
+ private static Map<String, Object> convertRowToMap(InternalRow row, RowType rowType) {
+ List<DataField> fields = rowType.getFields();
+ Map<String, Object> result = new LinkedHashMap<>(fields.size());
+ for (int i = 0; i < fields.size(); i++) {
+ DataField field = fields.get(i);
+ Object value = InternalRowUtils.get(row, i, field.type());
+ result.put(field.name(), convertValue(value, field.type()));
+ }
+ return result;
+ }
+
+ private static Object convertValue(Object value, DataType dataType) {
+ if (value == null) {
+ return null;
+ }
+ switch (dataType.getTypeRoot()) {
+ case INTEGER:
+ case BIGINT:
+ case FLOAT:
+ case DOUBLE:
+ case BOOLEAN:
+ case TINYINT:
+ case SMALLINT:
+ case CHAR:
+ case VARCHAR:
+ return value.toString();
Review Comment:
Maybe you can use JsonFormatWriter
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
+ String col = columns[i].trim();
+ int idx = fieldNames.indexOf(col);
+ if (idx < 0) {
+ throw new IllegalArgumentException(
+ "Column '" + col + "' not found in table schema: " + fieldNames);
+ }
+ indices[i] = idx;
+ }
+ return indices;
+ }
+
+ private static RowType buildProjectedRowType(RowType rowType, int[] projection) {
+ List<DataField> fields = rowType.getFields();
+ List<DataField> projected = new ArrayList<>(projection.length);
+ for (int idx : projection) {
+ projected.add(fields.get(idx));
+ }
+ return new RowType(projected);
+ }
+
+ // --- Row to JSON conversion (reuses the same pattern as JsonFormatWriter) ---
+
+ private static String convertRowToJsonString(InternalRow row, RowType rowType)
+ throws Exception {
+ Map<String, Object> result = convertRowToMap(row, rowType);
+ return JsonSerdeUtil.writeValueAsString(result);
+ }
+
+ private static Map<String, Object> convertRowToMap(InternalRow row, RowType rowType) {
+ List<DataField> fields = rowType.getFields();
+ Map<String, Object> result = new LinkedHashMap<>(fields.size());
+ for (int i = 0; i < fields.size(); i++) {
+ DataField field = fields.get(i);
+ Object value = InternalRowUtils.get(row, i, field.type());
+ result.put(field.name(), convertValue(value, field.type()));
+ }
+ return result;
+ }
+
+ private static Object convertValue(Object value, DataType dataType) {
+ if (value == null) {
+ return null;
+ }
+ switch (dataType.getTypeRoot()) {
+ case INTEGER:
+ case BIGINT:
+ case FLOAT:
+ case DOUBLE:
+ case BOOLEAN:
+ case TINYINT:
+ case SMALLINT:
+ case CHAR:
+ case VARCHAR:
+ return value.toString();
Review Comment:
Maybe you can use JsonFormatWriter
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "steFaiz (via GitHub)" <gi...@apache.org>.
steFaiz commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008881648
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
+ String col = columns[i].trim();
+ int idx = fieldNames.indexOf(col);
+ if (idx < 0) {
+ throw new IllegalArgumentException(
+ "Column '" + col + "' not found in table schema: " + fieldNames);
+ }
+ indices[i] = idx;
+ }
+ return indices;
+ }
+
+ private static RowType buildProjectedRowType(RowType rowType, int[] projection) {
+ List<DataField> fields = rowType.getFields();
Review Comment:
Thanks! modified!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "JingsongLi (via GitHub)" <gi...@apache.org>.
JingsongLi commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008572550
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
Review Comment:
rowType.projectIndexes
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "steFaiz (via GitHub)" <gi...@apache.org>.
steFaiz commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008880277
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
Review Comment:
Thanks! modified.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "steFaiz (via GitHub)" <gi...@apache.org>.
steFaiz commented on code in PR #7550:
URL: https://github.com/apache/paimon/pull/7550#discussion_r3008887798
##########
paimon-flink/paimon-flink-common/src/main/java/org/apache/paimon/flink/procedure/VectorSearchProcedure.java:
##########
@@ -0,0 +1,280 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.paimon.flink.procedure;
+
+import org.apache.paimon.casting.CastExecutor;
+import org.apache.paimon.casting.CastExecutors;
+import org.apache.paimon.catalog.Identifier;
+import org.apache.paimon.data.BinaryString;
+import org.apache.paimon.data.InternalArray;
+import org.apache.paimon.data.InternalMap;
+import org.apache.paimon.data.InternalRow;
+import org.apache.paimon.data.InternalVector;
+import org.apache.paimon.globalindex.GlobalIndexResult;
+import org.apache.paimon.reader.RecordReader;
+import org.apache.paimon.table.Table;
+import org.apache.paimon.table.source.ReadBuilder;
+import org.apache.paimon.table.source.TableScan;
+import org.apache.paimon.types.ArrayType;
+import org.apache.paimon.types.DataField;
+import org.apache.paimon.types.DataType;
+import org.apache.paimon.types.DataTypeRoot;
+import org.apache.paimon.types.DataTypes;
+import org.apache.paimon.types.MapType;
+import org.apache.paimon.types.RowType;
+import org.apache.paimon.types.VectorType;
+import org.apache.paimon.utils.InternalRowUtils;
+import org.apache.paimon.utils.JsonSerdeUtil;
+import org.apache.paimon.utils.StringUtils;
+
+import org.apache.flink.table.annotation.ArgumentHint;
+import org.apache.flink.table.annotation.DataTypeHint;
+import org.apache.flink.table.annotation.ProcedureHint;
+import org.apache.flink.table.procedure.ProcedureContext;
+
+import java.util.ArrayList;
+import java.util.Base64;
+import java.util.LinkedHashMap;
+import java.util.List;
+import java.util.Map;
+
+/**
+ * Vector search procedure. This procedure takes one vector and searches for topK nearest vectors.
+ * Usage:
+ *
+ * <pre><code>
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5
+ * )
+ *
+ * -- with projection and options
+ * CALL sys.vector_search(
+ * `table` => 'tableId',
+ * vector_column => 'v',
+ * query_vector => '1.0,2.0,3.0',
+ * top_k => 5,
+ * projection => 'id,name',
+ * options => 'k1=v1;k2=v2'
+ * )
+ * </code></pre>
+ */
+public class VectorSearchProcedure extends ProcedureBase {
+
+ public static final String IDENTIFIER = "vector_search";
+
+ private static final Base64.Encoder BASE64_ENCODER = Base64.getEncoder();
+
+ @ProcedureHint(
+ argument = {
+ @ArgumentHint(name = "table", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "vector_column", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "query_vector", type = @DataTypeHint("STRING")),
+ @ArgumentHint(name = "top_k", type = @DataTypeHint("INT")),
+ @ArgumentHint(
+ name = "projection",
+ type = @DataTypeHint("STRING"),
+ isOptional = true),
+ @ArgumentHint(name = "options", type = @DataTypeHint("STRING"), isOptional = true)
+ })
+ public String[] call(
+ ProcedureContext procedureContext,
+ String tableId,
+ String vectorColumn,
+ String queryVectorStr,
+ Integer topK,
+ String projection,
+ String options)
+ throws Exception {
+ Table table = catalog.getTable(Identifier.fromString(tableId));
+
+ Map<String, String> optionsMap = optionalConfigMap(options);
+ if (!optionsMap.isEmpty()) {
+ table = table.copy(optionsMap);
+ }
+
+ float[] queryVector = parseVector(queryVectorStr);
+
+ GlobalIndexResult result =
+ table.newVectorSearchBuilder()
+ .withVector(queryVector)
+ .withVectorColumn(vectorColumn)
+ .withLimit(topK)
+ .executeLocal();
+
+ RowType tableRowType = table.rowType();
+ int[] projectionIndices = parseProjection(projection, tableRowType);
+
+ ReadBuilder readBuilder = table.newReadBuilder();
+ if (projectionIndices != null) {
+ readBuilder.withProjection(projectionIndices);
+ }
+
+ TableScan.Plan plan = readBuilder.newScan().withGlobalIndexResult(result).plan();
+
+ RowType readType =
+ projectionIndices != null
+ ? buildProjectedRowType(tableRowType, projectionIndices)
+ : tableRowType;
+
+ List<String> rows = new ArrayList<>();
+ try (RecordReader<InternalRow> reader = readBuilder.newRead().createReader(plan)) {
+ reader.forEachRemaining(
+ row -> {
+ try {
+ rows.add(convertRowToJsonString(row, readType));
+ } catch (Exception e) {
+ throw new RuntimeException("Failed to convert row to JSON string", e);
+ }
+ });
+ }
+
+ return rows.toArray(new String[0]);
+ }
+
+ private static float[] parseVector(String vectorStr) {
+ String[] parts = vectorStr.split(",");
+ float[] vector = new float[parts.length];
+ for (int i = 0; i < parts.length; i++) {
+ vector[i] = Float.parseFloat(parts[i].trim());
+ }
+ return vector;
+ }
+
+ private static int[] parseProjection(String projection, RowType rowType) {
+ if (StringUtils.isNullOrWhitespaceOnly(projection)) {
+ return null;
+ }
+ String[] columns = projection.split(",");
+ int[] indices = new int[columns.length];
+ List<String> fieldNames = rowType.getFieldNames();
+ for (int i = 0; i < columns.length; i++) {
+ String col = columns[i].trim();
+ int idx = fieldNames.indexOf(col);
+ if (idx < 0) {
+ throw new IllegalArgumentException(
+ "Column '" + col + "' not found in table schema: " + fieldNames);
+ }
+ indices[i] = idx;
+ }
+ return indices;
+ }
+
+ private static RowType buildProjectedRowType(RowType rowType, int[] projection) {
+ List<DataField> fields = rowType.getFields();
+ List<DataField> projected = new ArrayList<>(projection.length);
+ for (int idx : projection) {
+ projected.add(fields.get(idx));
+ }
+ return new RowType(projected);
+ }
+
+ // --- Row to JSON conversion (reuses the same pattern as JsonFormatWriter) ---
+
+ private static String convertRowToJsonString(InternalRow row, RowType rowType)
+ throws Exception {
+ Map<String, Object> result = convertRowToMap(row, rowType);
+ return JsonSerdeUtil.writeValueAsString(result);
+ }
+
+ private static Map<String, Object> convertRowToMap(InternalRow row, RowType rowType) {
+ List<DataField> fields = rowType.getFields();
+ Map<String, Object> result = new LinkedHashMap<>(fields.size());
+ for (int i = 0; i < fields.size(); i++) {
+ DataField field = fields.get(i);
+ Object value = InternalRowUtils.get(row, i, field.type());
+ result.put(field.name(), convertValue(value, field.type()));
+ }
+ return result;
+ }
+
+ private static Object convertValue(Object value, DataType dataType) {
+ if (value == null) {
+ return null;
+ }
+ switch (dataType.getTypeRoot()) {
+ case INTEGER:
+ case BIGINT:
+ case FLOAT:
+ case DOUBLE:
+ case BOOLEAN:
+ case TINYINT:
+ case SMALLINT:
+ case CHAR:
+ case VARCHAR:
+ return value.toString();
Review Comment:
I previously copied those code from JsonFormatWriter. Now I've introduced a JsonFormatWriter to write those rows to memory bytes.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
Re: [PR] [flink] support vector search procedure for flink [paimon]
Posted by "JingsongLi (via GitHub)" <gi...@apache.org>.
JingsongLi merged PR #7550:
URL: https://github.com/apache/paimon/pull/7550
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscribe@paimon.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org