List Behavior and Usage in Cache

Note

This feature is only available in NCache Enterprise Edition.

A list in NCache is an extension of the .NET IList, which also provides additional properties. A list is an unordered data type, where data can be added or removed from any point of the list.

For example, lists can be used to maintain a list of the products being added to cart for an e-commerce website. Let's suppose user1 adds the products Umbrella, Green Apples and Coffee to the cart. Before making the transaction, the product Green Apples is removed and a new product Pears is added. This is possible because a list can be updated from any point in the list.

NCache further enhances this list by providing NCache specific features such as expiration, dependency, groups and more to the list. In this scenario, the company wants the cart list to be maintained only as long as the session is active. Hence, expiration can be associated with each list created which is equal to the session timeout value.

Since NCache List is an extension of the .NET Framework IList, you can still use all methods exposed by IList along with NCache's methods which are cache-specific. This ensures familiarity and flexibility with the existing framework.

Behavior

A list can be of any primitive type or custom object.
A list of CacheItem and nested lists are not supported yet.
Lists can be directly accessed by index.
Lists are named; you need to provide a unique cache key for the list.
Null is not a supported value type.
Duplicate values are supported.

Pre-requisites

Include the following namespaces in your application:
- Alachisoft.NCache.Client.DataTypes
- Alachisoft.NCache.Client.DataTypes.Collections
- Alachisoft.NCache.Client
- Alachisoft.NCache.Runtime.Exceptions
The application must be connected to cache before performing the operation.
Cache must be running.
To ensure the operation is fail safe, it is recommended to handle any potential exceptions within your application, as explained in Handling Failures.

Create List and Add Data

The following code sample shows how a list of Product type can be created in cache using CreateList against the cache key "ProductList". Products are added to the list and then a new range of products is added to the list using Add and AddRange.

Tip

You can also configure searchable attributes such as groups/tags/named tags and invalidation attributes such as expiration/eviction/dependency while creating a data type.

try
{
    // Pre-condition: Cache must be connected

    // Specify unique cache key for list
    string key = "ProductList";

    // Create list of Product type
    IDistributedList<Product> list = cache.DataTypeManager.CreateList<Product>(key);

    // Check if the list exists in cache
    // using cache.Contains() 

    // Get products to add to list
    Product[] products = FetchProducts();

    foreach (var product in products)
    {
        // Add products to list
        list.Add(product);
    }

    // Get new products
    Product[] newProducts = FetchNewProducts();

    // Append list of new Products to existing list
    list.AddRange(newProducts);

    // You can check count of items in list using
    // list.Count
}
catch (OperationFailedException ex)
{
    // NCache specific exception
    if(ex.ErrorCode == NCacheErrorCodes.KEY_ALREADY_EXISTS)
    {
        // The specified key already exists in cache,
        // Either remove the existing object from cache
        // Or specify another key
    }   
    else if (ex.ErrorCode == NCacheErrorCodes.CACHEITEM_IN_DATA_STRUCTURES)
    {
        // Data structures cannot be of CacheItem type
        // CacheItems cannot be added in data structures
    }
    else
    {
        // Exception can occur due to:
        // Connection Failures
        // Operation Timeout
        // Operation performed during state transfer
    }
}
catch (Exception ex)
{
    // Any generic exception like ArgumentNullException or ArgumentException
}

Recommendation: To ensure the operation is fail safe, it is recommended to handle any potential exceptions within your application, as explained in Handling Failures.

Update Items in List

Since lists can be accessed via index, you can update lists and items in them using indexes. The following code sample updates a value in an existing list (created in previous example) using the index. It then gets a sale item and adds it to the first index of the list using InsertAtHead.

try
{
    // Pre-condition: Cache is already connected
    // "list" is created in previous example

    // Update value of index with updated product 
    Product updatedProduct = GetUpdatedProductByID(11);
    list[11] = updatedProduct;

    // Get product on sale to insert at head of List
    Product saleProduct = FetchSaleItem();
    list.InsertAtHead(saleProduct);
}
catch (OperationFailedException ex)
{
    // NCache specific exception
    // Exception can occur due to:
    // Connection Failures
    // Operation Timeout
    // Operation performed during state transfer
}
catch (Exception ex)
{
    // Any generic exception like ArgumentNullException or ArgumentException
    // Item being added on an index out of range of list. Make sure index is
    // lesser than the size of list, or
    // a non-negative integer
}

Fetch List from Cache

You can fetch a list from cache using GetList, which takes a cache key as parameter. This key is the name of the list which is specified during list creation.

Warning

If the item being fetched is not of List type, a Type mismatch exception is thrown.

 try
{
    // Pre-condition: Cache is already connected

    // List with this key already exists in cache
    string key = "ProductList";

    // Get list and show items of list
    IDistributedList<Product> retrievedList = 
                                cache.DataTypeManager.GetList<Product>(key);

    if (retrievedList != null)
    {
        foreach (var item in retrievedList)
        {
            // Perform operations
        }
    }
    else
    {
       // List does not exist
    }
}
catch (OperationFailedException ex)
{
    // NCache specific exception

    if (ex.ErrorCode == NCacheErrorCodes.NOT_A_LIST)
    {
        // Item being fetched is not of List type;
        // Cache key corresponds to an item of different data type
    }
    else
    {
        // NCache specific exception
        // Exception can occur due to:
        // Connection Failures
        // Operation Timeout
        // Operation performed during state transfer
    }
}
catch (Exception ex)
{
    // Any generic exception like ArgumentNullException or ArgumentException
}

Recommendation: To ensure the operation is fail safe, it is recommended to handle any potential exceptions within your application, as explained in Handling Failures.

Remove Items from List

Items can be removed from a list against a given range of items. The following code sample removes the range of items for the expired products using RemoveRange.

Tip

To remove the whole list from cache, refer to the Remove Datatypes From Cache page.

Note

If key specified to be removed does not exist, nothing is returned. You can verify number of keys returned using the return type of RemoveRange.

try
{
    // Pre-condition: Cache is already connected

    // List with this key already exists in cache
    string key = "ProductList";

    // Get list to remove items
    IDistributedList<Product> retrievedList = 
                                cache.DataTypeManager.GetList<Product>(key);

    // Get range of expired products to be removed
    List<Product> itemsToRemove = FetchExpiredProducts();

    // Remove this range from retrievedList
    // Number of keys removed is returned
    int itemsRemoved = retrievedList.RemoveRange(itemsToRemove);
}
catch (OperationFailedException ex)
{
    // NCache specific exception

    if (ex.ErrorCode == NCacheErrorCodes.NOT_A_LIST)
    {
        // Item being fetched is not of List type;
        // Cache key corresponds to an item of different data type
    }
    else
    {
        // NCache specific exception
        // Exception can occur due to:
        // Connection Failures
        // Operation Timeout
        // Operation performed during state transfer
    }
}
catch (Exception ex)
{
    // Any generic exception like ArgumentNullException or ArgumentException
}

Recommendation: To ensure the operation is fail safe, it is recommended to handle any potential exceptions within your application, as explained in Handling Failures.

Event Notifications on Lists

You can register cache events as well as data type events on a data type such as list. For behavior, refer to the Feature Wise behavior topic.

The following code sample registers a cache event of ItemAdded and ItemUpdated as well as registers an event for ItemAdded and ItemUpdated on the list in cache.

Once a list is created in cache, an ItemAdded cache level event is fired. However, once an item is added to list, an ItemAdded data type event is fired, and an ItemUpdated cache level event is fired.

Register Event on List Created

try
{
    // Pre-condition: Cache is connected

    // Unique cache key for list
    string key = "ProductList";

    // Create list of Product type
    IDistributedList<Product> list = cache.DataTypeManager.CreateList<Product>(key);

    // Register ItemAdded, ItemUpdated, ItemRemoved events on list created
    // DataTypeNotificationCallback is callback method specified
    list.RegisterNotification(DataTypeDataNotificationCallback, EventType.ItemAdded | 
            EventType.ItemUpdated | EventType.ItemRemoved,
            DataTypeEventDataFilter.Data);

    // Perform operations
}
catch (OperationFailedException ex)
{
    // NCache specific exception
    // Exception can occur due to:
    // Connection Failures
    // Operation Timeout
    // Operation performed during state transfer
}   
catch (Exception ex)
{
    // Any generic exception like ArgumentNullException or ArgumentException
}

Specify Callback for Event Notification

private void DataTypeDataNotificationCallback(string collectionName, DataTypeEventArg collectionEventArgs)
{
    switch (collectionEventArgs.EventType)
    {
        case EventType.ItemAdded:
            // Item has been added to the collection
            break;

        case EventType.ItemUpdated:
            if (collectionEventArgs.CollectionItem != null)
            {
                // Item has been updated in the collection
                // Perform operations
            }
            break;

        case EventType.ItemRemoved:
            // Item has been removed from the collection
            break;
    }
}

Locking Lists

Lists can be explicitly locked and unlocked to ensure data consistency. The following code sample creates a list and locks it for a period of 10 seconds using Lock(), and then unlocks it using Unlock().

try
{
    // Pre-conditions: Cache is already connected 
    // List exists with key "ProductList"

    // Cache Key
    string key = "ProductList";

    // Get list
    IDistributedList<Product> list = cache.DataTypeManager.GetList<Product>(key);

    bool isLocked = list.Lock(TimeSpan.FromSeconds(10));

    if (isLocked)
    {
        // List is successfully locked for 10 seconds
        // Unless explicitly unlocked
    }
    else
    {
        // List is not locked because either:
        // List is not present in the cache
        // List is already locked
    }

    list.Unlock();
}
catch (OperationFailedException ex)
{
    // NCache specific exception
    // Exception can occur due to:
    // Connection Failures
    // Operation Timeout
    // Operation performed during state transfer
}
catch (Exception ex)
{
    // Any other generic exception like ArgumentNullException or ArgumentException
}

Additional Resources

NCache provides sample application for List datatype at:

GitHub
Shipped with NCache: %NCHOME%\samples\dotnet\DataTypes\DistributedList

List Behavior and Usage in Cache

Note

Behavior

Pre-requisites

Create List and Add Data

Tip

Update Items in List

Fetch List from Cache

Warning

Remove Items from List

Tip

Note

Event Notifications on Lists

Register Event on List Created

Specify Callback for Event Notification

Locking Lists

Additional Resources

See Also