Query Resultsets

Description — Couchbase Lite JavaScript — Working with Query Results
Related Content — SQL++ for Mobile | Live Queries | Indexing

Overview

After executing a query, you work with the results through query resultsets. This page shows you how to iterate through results, access result values, and work with result data in different formats.

Executing Queries

Couchbase Lite JavaScript provides two ways to execute queries:

Execute and Return Array

The execute() method without arguments returns an array of all results:

Example 1. Execute and return array

const query = database.createQuery('SELECT * FROM tasks WHERE completed = false');
const results = await query.execute();

console.log(`Found ${results.length} incomplete tasks`);

for (const row of results) {
    console.log(`Task: ${(row.tasks! as JSONObject).title}`);
}

Execute with Callback

The execute(callback) method calls your callback function for each result row, which is more memory-efficient for large result sets:

Example 2. Execute with callback

const query = database.createQuery('SELECT * FROM tasks WHERE completed = false');

let count = 0;
await query.execute(row => {
    console.log(`Task ${++count}: ${(row.tasks! as JSONObject).title}`);
    // Process each row without storing all in memory
});

console.log(`Processed ${count} tasks total`);

Use the callback pattern when processing large result sets to avoid loading all results into memory at once.

Result Structure

Each result row is a JavaScript object. The structure depends on your SELECT clause.

Getting Column Names

Use the columnNames property to get the names of result columns:

Example 3. Access column names

const query = database.createQuery(`
    SELECT title, completed, createdAt
    FROM tasks
`);

const columnNames = query.columnNames;
console.log('Columns:', columnNames);  // ['title', 'completed', 'createdAt']

SELECT * Results

When you use SELECT *, results are nested under the collection name:

Example 4. SELECT * structure

const starQuery = database.createQuery('SELECT * FROM tasks');
const starResults = await starQuery.execute();

for (const row of starResults) {
    // Results nested under collection name
    const task = row.tasks! as JSONObject;  // Access via collection name
    console.log(task.title, task.completed);
}

SELECT Properties Results

When you select specific properties, they appear as top-level properties in the result:

Example 5. SELECT properties structure

const propsQuery = database.createQuery(`
    SELECT title, completed
    FROM tasks
`);
const propsResults = await propsQuery.execute();

for (const row of propsResults) {
    // Properties at top level
    console.log(row.title, row.completed);
}

Accessing Result Values

Result values are accessed as regular JavaScript object properties:

Example 6. Accessing values

const query = database.createQuery('SELECT title, completed, priority FROM tasks');
const results = await query.execute();

for (const row of results) {
    const title = row.title;
    const completed = row.completed;
    const priority = row.priority;
    console.log(`${title}: ${completed ? 'Done' : 'Pending'} (Priority: ${priority})`);
}

Using Result Aliases

Use the AS keyword to create aliases for result columns:

Example 7. Result aliases

const aliasQuery = database.createQuery(`
    SELECT title     AS taskName,
           completed AS isDone,
           createdAt AS created
    FROM tasks
`);

const results = await aliasQuery.execute();

for (const row of results) {
    console.log(`${row.taskName} - Done: ${row.isDone}`);
}

Accessing Nested Properties

When querying nested objects, the result structure reflects the query:

Example 8. Nested properties

const nestedQuery = database.createQuery(`
    SELECT title,
           assignee.name,
           assignee.email,
           metadata.tags
    FROM tasks
`);

const results = await nestedQuery.execute();

for (const row of results) {
    const tags = row.tags! as JSONArray;
    console.log(`Task: ${row.title}`);
    console.log(`Assigned to: ${row.name} (${row.email})`);
    console.log(`Tags: ${tags.join(', ')}`);
}

Collecting Results

The execute() method without arguments returns all results as an array:

Example 9. Collect all results

const query = database.createQuery('SELECT * FROM tasks');
const allResults = await query.execute();

// Now you have an array of all results
console.log(`Total tasks: ${allResults.length}`);

// Can use array methods
const completedTasks = allResults.filter(row => {
    const task = row.tasks! as JSONObject;
    return task.completed;
});
console.log(`Completed: ${completedTasks.length}`);

For memory efficiency with large result sets, use the callback pattern instead:

Example 10. Process results with callback

const query = database.createQuery('SELECT * FROM tasks');

const completed = [];
await query.execute(row => {
    const task = row.tasks! as JSONObject;
    if (task.completed) {
        completed.push(task);
    }
    // Row is processed and can be garbage collected
});

console.log(`Found ${completed.length} completed tasks`);

Counting Results

To count results without collecting them all:

Example 11. Count results

const query = database.createQuery('SELECT COUNT(*) AS count FROM tasks WHERE completed = false');
const results = await query.execute();

const count = results[0].count;
console.log(`Incomplete tasks: ${count}`);

Working with Aggregate Results

When using aggregate functions, results contain the aggregated values:

Example 12. Aggregate results

const aggQuery = database.createQuery(`
    SELECT COUNT(*)            AS total,
           SUM(estimatedHours) AS totalHours,
           AVG(priority)       AS avgPriority,
           MIN(createdAt)      AS earliest,
           MAX(createdAt)      AS latest
    FROM tasks
`);

const results = await aggQuery.execute();
const stats = results[0];

console.log(`Total: ${stats.total}`);
console.log(`Total Hours: ${stats.totalHours}`);
console.log(`Avg Priority: ${stats.avgPriority}`);

GROUP BY Results

GROUP BY queries return one result per group:

Example 13. GROUP BY results

const groupQuery = database.createQuery(`
    SELECT status,
           COUNT(*) AS count,
    AVG(priority) AS avgPriority
    FROM tasks
    GROUP BY status
`);

const results = await groupQuery.execute();

for (const row of results) {
    console.log(`${row.status}: ${row.count} tasks (Avg Priority: ${row.avgPriority})`);
}

JOIN Results

JOIN queries combine data from multiple collections:

Example 14. JOIN results

const joinQuery = database.createQuery(`
    SELECT tasks.title,
           tasks.status,
           users.username,
           users.email
    FROM tasks
             LEFT OUTER JOIN users ON tasks.assignedTo = users._id
`);

const results = await joinQuery.execute();

for (const row of results) {
    console.log(`Task: ${row.title}`);
    if (row.username) {
        console.log(`Assigned to: ${row.username} (${row.email})`);
    } else {
        console.log('Unassigned');
    }
}

Handling NULL and MISSING Values

SQL++ distinguishes between NULL and MISSING values:

NULL - Explicitly set to null
MISSING - Property doesn’t exist

Example 15. NULL and MISSING handling

const query = database.createQuery(`
    SELECT title,
           assignedTo,
           dueDate,
           IFNULL(assignedTo, 'unassigned')  AS assigned,
           IFMISSING(dueDate, 'no due date') AS due
    FROM tasks
`);

const results = await query.execute();

for (const row of results) {
    // NULL values appear as null
    if (row.assignedTo === null) {
        console.log('Task not assigned');
    }

    // MISSING values don't exist in the object
    if (!('dueDate' in row)) {
        console.log('No due date set');
    }
}

Converting Results to JSON

You can convert result objects to JSON strings:

Example 16. JSON conversion

const query = database.createQuery('SELECT * FROM tasks LIMIT 10');
const results = await query.execute();

// Convert to JSON string
const jsonString = JSON.stringify(results, null, 2);
console.log(jsonString);

// Or save to file/localStorage
localStorage.setItem('taskResults', jsonString);

TypeScript Result Types

When using TypeScript, specify the result type using generics for type safety:

Example 17. Typed results

interface TaskResult {
    title: string;
    completed: boolean;
    priority: number;
    createdAt: string;
}

const query = database.createQuery(`
    SELECT title, completed, priority, createdAt
    FROM tasks
    WHERE completed = false
`);

const results = await query.execute<TaskResult>();

// TypeScript knows the result type
for (const row of results) {
    const title: string = row.title;  // Type-safe access
    const priority: number = row.priority;
}

The Query class is generic: Query<Schema> where Schema is inherited from its Database. You can also specify the result type when calling execute<T>():

Example 18. Specify result type

interface TaskSummary {
    taskName: string;
    isDone: boolean;
}

const query = database.createQuery(`
    SELECT title     AS taskName,
           completed AS isDone
    FROM tasks
`);

const results = await query.execute<TaskSummary>();

// Results are typed as TaskSummary[]
results.forEach(task => {
    console.log(`${task.taskName}: ${task.isDone ? 'Done' : 'Pending'}`);
});

TypeScript result types are not runtime-checked. It’s your responsibility to ensure the type matches the actual query results.

Result Processing Patterns

Streaming Processing

Process results as they arrive without storing them all in memory:

Example 19. Streaming pattern

const query = database.createQuery('SELECT * FROM tasks');

let processedCount = 0;
await query.execute(row => {
    // Process each row immediately
    processRow(row.tasks);
    processedCount++;

    // Row is not stored, can be garbage collected
});

console.log(`Streamed ${processedCount} results`);

Filtering Results

Filter results in your callback:

Example 20. Filtering pattern

const query = database.createQuery('SELECT * FROM tasks');

const urgentTasks = [];
await query.execute(row => {
    const task = row.tasks as JSONObject;

    // Only keep urgent tasks
    if (task.priority === 'high' && !task.completed) {
        urgentTasks.push(task);
    }
});

console.log(`Found ${urgentTasks.length} urgent tasks`);

Transforming Results

Transform result data during iteration:

Example 21. Transformation pattern

const query = database.createQuery(
    'SELECT * FROM tasks'
);

const taskSummaries = [];
await query.execute(row => {
    const task = row.tasks as JSONObject;

    // Transform to summary format
    taskSummaries.push({
        id: task._id,
        title: task.title,
        status: task.completed ? 'Done' : 'Pending',
        daysOld: calculateDaysOld(task.createdAt as string)
    });
});

Asynchronous Processing

The execute() method itself is asynchronous. If you use the callback pattern, your callback can also be asynchronous:

Example 22. Async callback processing

const query = database.createQuery(
    'SELECT * FROM tasks'
);

await query.execute(async (row) => {
    const task = row.tasks;

    // Can perform async operations
    await saveToCache(task);
    await updateUI(task);

    // Query waits for callback to complete
});

console.log('All results processed asynchronously');

With async callbacks, the execute() method waits for each callback to complete before processing the next result.

Error Handling

Handle errors during query execution:

Example 23. Error handling

try {
    const query = database.createQuery(
        'SELECT * FROM tasks WHERE priority > $minPriority'
    );
    query.parameters = { minPriority: 5 };
    const results = await query.execute();

    console.log(`Found ${results.length} high priority tasks`);
} catch (error) {
    if (error instanceof N1QLParseError) {
        console.error('Invalid query syntax:', error.message);
    } else if (error instanceof InterruptedQueryError) {
        console.error('Query was interrupted');
    } else {
        console.error('Query failed:', error);
    }
}

Performance Tips

Process Results Efficiently:

Use the callback pattern to process results one at a time
Avoid collecting all results if you only need to process them
Use LIMIT to restrict result count when appropriate

Memory Management:

Don’t store large result sets in memory unnecessarily
Process and release results as you go
Consider pagination for large datasets

Query Optimization:

Select only the properties you need
Use indexes to speed up queries (see Indexing)
Check query explanations to verify index usage

Controlling Result Iteration

When using the callback pattern, execute() returns a boolean indicating whether the query completed:

true - Query completed normally
false - Query was interrupted (see Interrupting Queries)

Example 24. Check if query completed

const query = database.createQuery(
    'SELECT * FROM tasks'
);

let count = 0;
const completed = await query.execute(row => {
    count++;
    // Process row
});

if (completed) {
    console.log(`Query completed successfully, processed ${count} rows`);
} else {
    console.log('Query was interrupted');
}

Interrupting Queries

You can stop a running query using the interrupt() method:

Example 25. Interrupt a query

const query = database.createQuery(
    'SELECT * FROM tasks'
);

// Start query execution
const executePromise = query.execute(row => {
    // Long-running processing
    processRow(row);
});

// Interrupt after timeout
setTimeout(() => {
    query.interrupt();
    console.log('Query interrupted');
}, 5000);

try {
    await executePromise;
} catch (error) {
    if (error instanceof InterruptedQueryError) {
        console.log('Query was interrupted as expected');
    }
}

When interrupted, execute() throws an InterruptedQueryError.

How to . . .

Learn more . . .

Dive Deeper . . .

Mobile Forum | Blog | Tutorials