Data Model - EmployeePropsClick to view data model used in all examples
Tag with priority for categorization.
/// <summary>
/// Tag with priority for categorization.
/// </summary>
public class Tag
{
public string Name { get; set; } = string.Empty;
public int Priority { get; set; }
public string? Description { get; set; }
}
/// <summary>
/// Project metrics for analytics.
/// </summary>
[RedbScheme("ProjectMetrics")]
public class ProjectMetricsProps
{
public long ProjectId { get; set; }
public long? TasksCompleted { get; set; }
public long? TasksTotal { get; set; }
public long? BugsFixed { get; set; }
public double? Budget { get; set; }
public long? TeamSize { get; set; }
public Tag[]? Tags { get; set; }
public string[]? Technologies { get; set; }
}
/// <summary>
/// Address with city, street and building details.
/// </summary>
public class Address
{
public string City { get; set; } = string.Empty;
public string Street { get; set; } = string.Empty;
public BuildingInfo? Building { get; set; }
}
/// <summary>
/// Building information with floor and amenities.
/// </summary>
public class BuildingInfo
{
public int Floor { get; set; }
public string Name { get; set; } = string.Empty;
public string[]? Amenities { get; set; }
public int[]? AccessCodes { get; set; }
public string[]? ParkingSpots { get; set; }
public int[]? ElevatorFloors { get; set; }
}
/// <summary>
/// Contact detail key-value pair.
/// </summary>
public class ContactDetail
{
public string Label { get; set; } = string.Empty;
public string Value { get; set; } = string.Empty;
}
/// <summary>
/// Contact information (email, phone, etc).
/// </summary>
public class Contact
{
public string Type { get; set; } = string.Empty;
public string Value { get; set; } = string.Empty;
public bool IsVerified { get; set; }
public int[]? NotificationHours { get; set; }
public ContactDetail[]? Metadata { get; set; }
}
/// <summary>
/// Department info with tags, metrics and nested budget data.
/// </summary>
public class Department
{
public string Name { get; set; } = string.Empty;
public int HeadCount { get; set; }
public string[] Projects { get; set; } = [];
public Tag[] Leaders { get; set; } = [];
public Dictionary<string, int>? BudgetByYear { get; set; }
}
/// <summary>
/// Employee props - main model for examples.
/// Demonstrates all supported types:
/// - Simple types (int, string, DateTime, long, decimal)
/// - Arrays (string[], int[])
/// - Business classes (Address with nested BuildingInfo)
/// - Array of business classes (Contact[])
/// - RedbObject references (CurrentProject, PastProjects[])
/// - Dictionary types (various key and value types)
/// </summary>
[RedbScheme("Employee")]
public class EmployeeProps
{
// Basic info
public string FirstName { get; set; } = string.Empty;
public string LastName { get; set; } = string.Empty;
public int Age { get; set; }
public DateTime HireDate { get; set; }
public string Position { get; set; } = string.Empty;
public decimal Salary { get; set; }
public string Department { get; set; } = string.Empty;
public string? EmployeeCode { get; set; }
// Skills and certifications (arrays)
public string[]? Skills { get; set; }
public int[]? SkillLevels { get; set; }
public string[]? Certifications { get; set; }
public int[]? CertificationYears { get; set; }
// Addresses (business classes)
public Address? HomeAddress { get; set; }
public Address? WorkAddress { get; set; }
public Address? EmergencyAddress { get; set; }
// Contacts (array of business classes)
public Contact[]? Contacts { get; set; }
// Project references (RedbObject)
public RedbObject<ProjectMetricsProps>? CurrentProject { get; set; }
public RedbObject<ProjectMetricsProps>[]? PastProjects { get; set; }
// Phone directory: extension -> phone number
public Dictionary<string, string>? PhoneDirectory { get; set; }
// Office locations by city
public Dictionary<string, Address>? OfficeLocations { get; set; }
// Bonus history by year
public Dictionary<int, decimal>? BonusByYear { get; set; }
// Department details with complex nested data (Pro)
public Dictionary<string, Department>? DepartmentHistory { get; set; }
// Composite key: (year, quarter) -> performance score (Pro)
public Dictionary<(int Year, string Quarter), string>? PerformanceReviews { get; set; }
// Project metrics by code (Pro)
public Dictionary<string, RedbObject<ProjectMetricsProps>>? ProjectMetrics { get; set; }
}
/// <summary>
/// Department props for tree examples (corporate hierarchy).
/// Used with TreeRedbObject for organizational structure.
/// Similar to CategoryTestProps in ConsoleTest.
/// </summary>
[RedbScheme("Department")]
public class DepartmentProps
{
/// <summary>Department name (e.g. "IT Department Moscow").</summary>
public string Name { get; set; } = string.Empty;
/// <summary>Description (e.g. "Software development team").</summary>
public string Description { get; set; } = string.Empty;
/// <summary>Is department active?</summary>
public bool IsActive { get; set; } = true;
/// <summary>Budget in USD.</summary>
public decimal Budget { get; set; }
/// <summary>Department code (e.g. "IT-MSK-DEV").</summary>
public string Code { get; set; } = string.Empty;
}
/// <summary>
/// City props for list examples (linked objects).
/// Used as target object for RedbListItem.Object reference.
/// </summary>
[RedbScheme("City")]
public class CityProps
{
/// <summary>City name.</summary>
public string Name { get; set; } = string.Empty;
/// <summary>Population count.</summary>
public int Population { get; set; }
/// <summary>Region or district.</summary>
public string Region { get; set; } = string.Empty;
/// <summary>Is capital city?</summary>
public bool IsCapital { get; set; }
/// <summary>GPS coordinates [lat, lon].</summary>
public double[] Coordinates { get; set; } = [];
}
/// <summary>
/// Person props demonstrating ListItem fields.
/// Shows single ListItem and List of ListItems usage.
/// </summary>
[RedbScheme("Person")]
public class PersonProps
{
/// <summary>Person name.</summary>
public string Name { get; set; } = string.Empty;
/// <summary>Age in years.</summary>
public int Age { get; set; }
/// <summary>Email address.</summary>
public string Email { get; set; } = string.Empty;
/// <summary>Single ListItem field (e.g. status from dictionary).</summary>
public RedbListItem? Status { get; set; }
/// <summary>Array of ListItems (e.g. roles from dictionary).</summary>
public List<RedbListItem>? Roles { get; set; }
}
/// <summary>
/// Product with a unique business key (V4 [RedbUnique]): the SKU is unique per scheme,
/// enforced by the database over the hash of the canonical value. Used by E003.
/// </summary>
public class ProductProps
{
/// <summary>Unique business key. NULL never participates in uniqueness.</summary>
[RedbUnique]
public string? Sku { get; set; }
/// <summary>Display name (not unique).</summary>
public string Title { get; set; } = string.Empty;
/// <summary>Price in USD.</summary>
public decimal Price { get; set; }
}
WhereRedb - System Fields and Composition
WhereRedb: filtering on SYSTEM fields of _objects (id, parent_id, key, value_string, date_modify and friends) as opposed to Where, which filters on your Props. The two compose freely in one query - the system predicate and the Props predicate both run server-side. Patterns shown here mirror production usage: a lookup by the service value_string key, an id-set prefilter with Contains, and the composite WhereRedb().Where() shape (discussion #12, item on WhereRedb docs).
var sw = Stopwatch.StartNew();
await redb.SyncSchemeAsync<EmployeeProps>();
// Seed: three employees carrying SYSTEM fields alongside Props. The service fields
// (key, value_string, value_long...) live on the _objects row itself - no scheme
// change needed - and are ideal for lookup keys next to your business data.
var seeded = new List<long>();
for (var i = 0; i < 3; i++)
{
var employee = new RedbObject<EmployeeProps>
{
name = $"e205-employee-{i}",
key = 7000 + i, // numeric service key
value_string = i == 0 ? "e205:head-of-desk" : null, // string lookup key
Props = new EmployeeProps
{
FirstName = $"E205-{i}",
LastName = "Sample",
Department = i == 2 ? "Sales" : "IT",
Salary = 50000m + i * 10000m,
},
};
seeded.Add(await redb.SaveAsync(employee));
}
// 1. Lookup by the service string key - the settings/singleton pattern:
// one indexed system column instead of a dedicated Props field.
var head = await redb.Query<EmployeeProps>()
.WhereRedb(o => o.ValueString == "e205:head-of-desk")
.FirstOrDefaultAsync();
// 2. Id-set prefilter with Contains - "load these exact objects, filtered further".
var byIds = await redb.Query<EmployeeProps>()
.WhereRedb(o => seeded.Contains(o.Id))
.ToListAsync();
// 3. The composite shape: a SYSTEM predicate and a Props predicate in one query.
// Both are translated to SQL - the numeric key range narrows on _objects, the
// department condition narrows on Props values.
var itByKey = await redb.Query<EmployeeProps>()
.WhereRedb(o => o.Key >= 7000 && o.Key <= 7002)
.Where(e => e.Department == "IT")
.ToListAsync();
sw.Stop();
// Cleanup: the example leaves no residue behind.
await redb.DeleteAsync(seeded);