queryIterator()

This operation returns a query iterator to iterate data.

java

public QueryIterator queryIterator(QueryIteratorReq request)

Request Syntax

java

queryIterator(QueryIteratorReq.builder()
    .databaseName(String databaseName)
    .collectionName(String collectionName)
    .partitionNames(List<String> partitionNames)
    .outputFields(List<String> outputFields)
    .expr(String expr)
    .consistencyLevel(ConsistencyLevel consistencyLevel)
    .offset(long offset)
    .limit(long limit)
    .ignoreGrowing(boolean ignoreGrowing)
    .timezone(String timezone)
    .batchSize(long batchSize)
    .reduceStopForBest(boolean reduceStopForBest)
    .cursor(QueryIteratorCursor cursor)
    .filterTemplateValues(Map<String, Object> filterTemplateValues)
    .build()
);

BUILDER METHODS:

• databaseName(String databaseName)

  The name of the database. Defaults to the current database if not specified.

• collectionName(String collectionName)

  The name of the target collection.

• partitionNames(List<String> partitionNames)

  A list of partition names to target.

• outputFields(List<String> outputFields)

  A list of field names to include in the output.

• expr(String expr)

  A boolean expression to filter results.

• consistencyLevel(ConsistencyLevel consistencyLevel)

  The consistency level for the operation.

• offset(long offset)

  The number of results to skip before returning.

• limit(long limit)

  The maximum number of results to return.

• ignoreGrowing(boolean ignoreGrowing)

  Whether to ignore growing segments during the operation.

• timezone(String timezone)

  The timezone string for time-related filters.

• batchSize(long batchSize)

  The batch size for iterator operations.

• reduceStopForBest(boolean reduceStopForBest)

  Whether to stop iteration when the best result is found.

• cursor(QueryIteratorCursor cursor)

  A previously captured cursor to resume pagination from. When set, the iterator continues from the cursor's session timestamp and position instead of starting over; offset() is ignored in that case.

• filterTemplateValues(Map<String, Object> filterTemplateValues)

  A map of template variable values for parameterized filters.

RETURN TYPE:

QueryIterator

RETURNS:

A QueryIterator object to iterate query results page by page, which offers the following methods:

PARAMETERS:

• next() (List<QueryResultsWrapper.RowRecord>) -

  Returns the next batch of rows matching the query expression.

• close() -

  Releases the iterator and the resources held by its cursor.

• getCursor() (QueryIteratorCursor) -

  Returns the cursor that holds the iteration progress.

EXCEPTIONS:

• MilvusClientException

  This exception will be raised when any error occurs during this operation.

Example

java

import io.milvus.orm.iterator.QueryIterator;
import io.milvus.response.QueryResultsWrapper;
import io.milvus.v2.service.vector.request.QueryIteratorReq;
import io.milvus.v2.common.ConsistencyLevel;

import java.util.Arrays;
import java.util.List;

// Create a query iterator to retrieve results in batches
QueryIterator queryIterator = client.queryIterator(QueryIteratorReq.builder()
        .collectionName("my_collection")
        .expr("userID < 3000")
        .outputFields(Arrays.asList("userID", "userAge"))
        .batchSize(100)
        .offset(0)
        .limit(10000)
        .consistencyLevel(ConsistencyLevel.BOUNDED)
        .build());

// Iterate through all results
int counter = 0;
while (true) {
    List<QueryResultsWrapper.RowRecord> res = queryIterator.next();
    if (res.isEmpty()) {
        queryIterator.close();
        break;
    }
    for (QueryResultsWrapper.RowRecord record : res) {
        System.out.println(record);
        counter++;
    }
}
System.out.printf("%d query results returned%n", counter);