Redis OM . Net redis object mapping framework

Redis OM

Redis OM yes Redis The official object mapping framework , namely ：Object Mapping. Make it easier for developers 、 Convenient operation Redis data .Redis The stored data is abstracted as object mapping , Support object-based Redis Data persistence 、 Streaming query operation .

Currently only supported 4 Development languages :

• Redis OM for Spring
• Redis OM for .NET
• Redis OM for Node.js
• Redis OM for Python

Redis OM .NET

Redis OM .NET yes .Net Platform Redis OM, rely on StackExchange.Redis Realization . With the help of Redis OM .NET Can realize the way of object operation Redis data , Out of the key/value How to operate . Query supports most .Neter favorite LINQ.

Quick start

Install the corresponding package

dotnet add package Redis.OM

Redis Environmental preparation

Use it directly Docker Mode of installation Redis Environmental Science .

docker run -p 6379:6379 redislabs/redismod:preview

The standard official image cannot support Redis OM, need Redis Modules Support ,Redis OM Core index creation 、 Query data dependency RediSearch This Module Realization . Rely on the Module Yes :RediSearch、RedisJSON. RedisJSON Dependency is not necessary , But there will be a lack of corresponding functions , Such as : Model nesting 、 Complex queries （ Only support key Inquire about ）

Coding

Add abstract object definition

[Document]public class Customer{    [RedisIdField]    public string Id { get; set; }    [Indexed(Sortable = true, Aggregatable = true)]    public string FirstName { get; set; }    [Indexed(Sortable = true, Aggregatable = true)]    public string LastName { get; set; }    [Indexed]    public string Email { get; set; }    [Indexed(Sortable = true)]    public int Age { get; set; }}

Document、Indexed、Searchable Introduction to other characteristics , Introduces the reference Github -> document-attribute

Get the operation collection of abstract objects 、 Create index

var provider = new RedisConnectionProvider("redis://localhost:6377");var connection = provider.Connection;var customers = provider.RedisCollection<Customer>();connection.CreateIndex(typeof(Customer));

Query data 、 Aggregation operations need to be based on the index , So be sure to call connection.CreateIndex Create index , Corresponding RediSearch Of FT.CREATE command .

connection.CreateIndex(typeof(Customer)) An exception is thrown when the index is created repeatedly Index already exists. Although it can pass connection.Execute("FT.INFO", $"customer-idx") Get index information , But the first time the index does not exist, it will throw Unknown Index name. So in actual use, you may need a try-catch encase CreateIndex Methods to avoid exceptions .

insert data

var id = customers.Insert(new Customer { FirstName = "Steve", Email = "[email protected]", Age = 1 });var id2 = customers.Insert(new Customer { FirstName = "FirstName", LastName = "LastName", Email = "[email protected]" });

id,id2 For inserting data key, Is not specified Document Of Prefixes and IdGenerationStrategy, The default is ULID The format is {DocumentName}:{Ulid}, Such as :`Cust

When inserting data, you should pay attention to , If the value of the field is not explicitly specified , Such as LastName Ambiguous assignment ：customers.Insert(new Customer { FirstName = "Steve", Email = "[email protected]", Age = 1 });, see Redis The data stored in can find the current key Stored json The data does not correspond to an unspecified field key（ here Query or Aggregations There will be some strange mistakes ）. According to their own needs , The displayed field is assigned a zero value or used when defining an entity public string LastName { get; set; } = string.Empty; The way .

Query data

var customer = customers.Where(x => x.Age == 0).OrderBy(a => a.FirstName).FirstOrDefault();var customerById = customers.FindById(id);// according to Id lookup var emails = customers.Select(x => x.Email);// Query only the specified fields var takes = customers.Where(x => x.Age > 0).Take(10);// Get the specified number var adults = customers.Where(x => x.Age >= 0).Skip(5);// Query offset 

Judgment of null value ,x.FirstName == ""【 Grammar mistakes 】 or string.IsNullOrEmpty(x.FirstName)【 I won't support it 】.Redis Cannot have empty string in hash , Therefore, similar queries should be operated through aggregation Exists Method realization

foreach (var agg in customerAggregations.Apply(x => ApplyFunctions.Exists(x.RecordShell.LastName), "LastNameExists")){    Console.WriteLine($"{agg["LastNameExists"]}");}

Aggregation operation

Assembly line （Pipelining） Send multiple requests at the same time , So as to reduce the delay . The query and transformation of the results are Redis The complete .

RecordShell It's the far end Index The structure of the type ,RecordShell Should only be used inside the aggregation pipeline , There is no real value at runtime .

Put together FirstName and LastName, return FullName

var customerAggregations = provider.AggregationSet<Customer>();var age = customerAggregations.Average(x => x.RecordShell.Age);var sets = customerAggregations.Where(a => a.RecordShell.FirstName == "Steve").Apply(x => string.Format("{0} {1}", x.RecordShell.FirstName, x.RecordShell.LastName), "FullName");foreach (var item in sets){    Console.WriteLine(item["FullName"].ToString());}

Group together

adopt GroupBy Method , Group and aggregate according to different attributes （ Support single field grouping and multi field grouping ）.

var res = customerAggregations               .GroupBy(x => x.RecordShell.FirstName)               .GroupBy(x => x.RecordShell.LastName)               .ToArray();var res1 = customerAggregations.GroupBy(x => x.RecordShell.FirstName).CloseGroup().ToArray();

CloseGroup You can turn off grouping , Convert to normal aggregation operation , namely GroupedAggregationSet To RedisAggregationSet A transformation of .

public static RedisAggregationSet<T> CloseGroup<T>(this GroupedAggregationSet<T> source){    return new RedisAggregationSet<T>(source, source.Expression);}

ending

This article is just for Redis OM .NET Simple sorting of usage and usability verification . More usage and usage update reference Github

We are acting , New framework 、 New ecology

Our goal is The freedom of the 、 Easy-to-use 、 Highly malleable 、 functional 、 Robust .

So we learn from Building blocks Design concept of , Working on a new framework MASA Framework, What are its characteristics ？

• Native support Dapr, And allow Dapr Replace with traditional means of communication
• Unlimited architecture , Single application 、SOA、 Micro services support
• Support .Net Native framework , Reduce the burden of learning , In addition to the concepts that must be introduced in a specific field , Insist on not making new wheels
• Rich ecological support , In addition to the framework, there are component libraries 、 Authority Center 、 Configuration center 、 Troubleshooting center 、 A series of products such as Alarm Center
• Unit test coverage of the core code base 90%+
• Open source 、 free 、 Community driven
• What is the ？ We are waiting for you , Come together and discuss

After several months of production project practice , Completed POC, At present, the previous accumulation is being refactored into new open source projects

At present, the source code has been synchronized to Github（ The document site is under planning , Will gradually improve ）：

MASA.BuildingBlocks

MASA.Contrib

MASA.Utils

MASA.EShop

BlazorComponent

MASA.Blazor

QQ Group ：7424099

Wechat group ： Plus technology operation wechat （MasaStackTechOps）, Remarks , Invite in

​ ------ END ------

Author's brief introduction

** Ma Yue ：**MASA Technical team members .