Endor.TinyServices.OData 1.2.0

Endor.TinyServices.OData

Small Odata service protocol for dotnet.

Configuration

Install Endor.TinyServices.OData and Endor.TinyServices.OData.Sql on the main project

On program.cs declare:

builder.Services.UseODataInterpreter().UseSql("Default");

The UseSql parameter is the current connection string.

Declare OData endpoint after the build:

app.AddODataEndpoint();

assemblies names are necessary to find the entities to be expose over the protocol (declared on appsettings)

"ODataServiceConfig": {
    "Assemblies": [ "BLLEntities" ]
  }

The service finds al classes inside the "entities" folder name by default. It can be changed using "EntitiesFolder" property from ODataServiceConfig over appsettings file.

How to use

This service exposes over the endpoint: "https://.../OData/"

The command list included into the service:

• $metadata
• $filter
• $expand → Only one level and without additional parameters except $select
• $select
• $orderby
• $count
• $skip
• $take

Additional configuration

OData Conventions:

• [ODataTableName("TableName")] attribute can be used to associate a table with the current class. In otheway each class is associated with the name by default.

• The entity requires an "Id" field. Is possible to declare a entity key using [ODataKey] attribute.

• The Odata conventions define the relationships between entities with the parameter "{Entity}Id" format. To redefine for one property use [ODataReference(nameof("Class"))] attribute

• IDialect interface provides a way to extend the service to other kind of databases.

Limitations

The relationships over the entity must be relationship one to one o many to one but one to many are not allowed yet.

Performance test

Environment: AdventureWorks2019 and ASPNet 6.0 Minimal API using Dapper and EF to compare the performance between OData service protocol.

Performance average from 100 requests to each endpoint on each scenario.

Case #1

Query simple over Work Order table.

Table: WorkOrder

Performance Select (n) |Service (ms) |Dapper (ms) |EF (ms) |------|:------:|:-----:|:------:| |1 |7.02 |18.16 |23.74 |5 |4.49 |18.68 |23.4 |10 |5.45 |20.5 |23.59 |15 |4.53 |18.42 |21.59 |20 |4.38 |20.35 |21.28 |30 |7.34 |20.07 |23.55 |50 |6.05 |18.7 |21.12 |100 |6.49 |19.5 |21.93

Case #2

Query with one join statement.

Table: WorkOrder Join: Product

Performance Select (n) |Service (ms) |Dapper (ms) |EF (ms) |------|:------:|:-----:|:------:| |1 |6.73 |20.42 |23.9 |5 |6.62 |18.97 |21.88 |10 |6.28 |20.33 |21.58 |15 |7.62 |19.3 |22.45 |20 |8.36 |18.05 |23.71 |30 |8.94 |19.75 |21.71 |50 |13.11 |19.86 |23.48 |100|20.01 |20.73 |25.98

Case #3

Query with join and filter.

Table: WorkOrder Join: Product Filter: [Product].[WeightUnitMeasureCode] = 'G' & [WorkOrder].[StockedQty] >= 1

Performance Select (n) |Service (ms) |Dapper (ms) |EF (ms) |------|:------:|:-----:|:------:| |1 |6.75 |18.82 |22.23 |5 |5.07 |19.42 |21.61 |10 |5.93 |20.82 |23.56 |15 |7.13 |19.03 |23.38 |20 |7.4 |19.02 |22.01 |30 |7.99 |19.91 |21.43 |50 |9.96 |18.81 |21.67 |100|15.42 |20.89 |25.08

Case #4

Query with join, filter and order by statement.

Table: WorkOrder Join: Product Filter: [Product].[WeightUnitMeasureCode] = 'G' & [WorkOrder].[StockedQty] >= 1 OrderBy: [Product].[ProductNumber]

Performance Select (n) |Service (ms) |Dapper (ms) |EF (ms) |------|:------:|:-----:|:------:| |1 |5.3 |16.33 |21.05 |5 |5.67 |17.98 |21.51 |10 |5.57 |16.61 |20.165 |15 |6.27 |19.44 |22.04 |20 |6.42 |18.92 |21.21 |30 |6.85 |19.58 |21.83 |50 |10.66 |19.53 |21.48 |100 |15.46 |19.52 |22.15