Introduction

Entity

A unified representation of a data record and its attributes.

In codenew ArxEntity("account")

Attributes

A dictionary-driven way of defining and accessing entity data.

In codeaccount["name"] = "Northwind Traders"

Conditions

A structured approach to defining data filters and query criteria.

In codequery.Criteria.AddCondition("employee_count", ArxConditionOperator.GreaterEqual, 40)

Service

A consistent interface for performing data operations.

In codeCreateAsync, RetrieveAsync, RetrieveMultipleAsync, UpdateAsync, DeleteAsync, ExecuteAsync

Provider Abstraction

Provider-specific implementations that translate the common programming model into operations supported by each database.

In codeARX.SDK.PostgreSql → ArxPgDbService

An entity, its attributes, a condition and the service, using the PostgreSQL provider.

using ARX.SDK.PostgreSql.Services;
using ARX.SDK.Query;
using ARX.SDK.Services;

// Declare each table's id column once, at startup.
var schema = ArxSchemaCatalog.Create()
    .Table("account", ArxIdColumn.Generated<long>("id"))
    .Build();

// The service: here, the PostgreSQL provider's.
var service = new ArxPgDbService(connectionString, schema);

// An entity and its attributes.
var account = new ArxEntity("account");
account["name"] = "Northwind Traders";
account["account_number"] = "HOME-0001";
account["employee_count"] = 240;

ArxRecordId id = await service.CreateAsync(account);

// A query with conditions.
var query = new ArxQueryExpression("account");
query.ColumnSet.AddColumns("name", "employee_count");
query.Criteria.AddCondition("employee_count", ArxConditionOperator.GreaterEqual, 40);
query.AddOrder("name", ArxOrderType.Ascending);

ArxEntityCollection result = await service.RetrieveMultipleAsync(query);

Taken from two samples that were built and run against PostgreSQL: Create one record and Retrieve many records with a query. PostgreSQL is the most complete provider today; see Supported Providers for the others.

ARX.SDK fits code that works with records whose tables are known by name at run time, or that must run on more than one database:

• a product sold to customers who each bring their own database (PostgreSQL, SQL Server, MySQL, MongoDB or DynamoDB);
• metadata-driven applications: configurable forms, import tools, admin screens, where tables and columns are data;
• teams coming from the Dynamics 365 / XRM SDK who want the same entity and query model on their own databases.

For an application on one database with a fixed schema, EF Core (typed models, LINQ, migrations) or Dapper (your own SQL) are usually the better tools: the EF Core and Dapper guide compares them honestly.

Working with an AI coding assistant? Give it llms.txt: a knowledge base per provider, generated from verified samples, and the arx-sdk skill for coding agents.