EFCore.ParadeDB.PgSearch adds ParadeDB's pg_search extension support to Npgsql.EntityFrameworkCore.PostgreSQL. It exposes ParadeDB search functions through EF.Functions API for LINQ-based full-text search queries. BM25 index creation must be done via raw SQL. See the EF Core documentation for more details on adding raw SQL to migrations.
Install the Npgsql.EntityFrameworkCore.PostgreSQL NuGet package and configure your DbContext by calling UsePgSearch()
on the NpgsqlDbContextOptionsBuilder
to enable pg_search function mappings.
var builder = WebApplication.CreateBuilder(args);
builder.Services.AddDbContextPool<AppDbContext>(opt =>
{
opt.UseNpgsql(
builder.Configuration.GetConnectionString("AppDatabase"),
o => o.UsePgSearch()
);
});
Register the pg_search
extension in the DbContext.
public class AppDbContext : DbContext
{
public AppDbContext(DbContextOptions<AppDbContext> options)
: base(options) { }
protected override void OnModelCreating(ModelBuilder modelBuilder)
{
base.OnModelCreating(modelBuilder);
modelBuilder.HasPostgresExtension("pg_search");
}
}
A complete example is also available in the samples directory demonstrating PgSearch configuration and BM25 index creation.
The following ParadeDB operations are available through the EF.Functions
API:
ParadeDB Operation | LINQ Methods |
---|---|
Match | MatchDisjunction() , MatchConjunction() |
Phrase | Phrase() |
Term | Term() |
BM25 scoring | Score() |
Highlighting | Snippet() |
Proximity | Proximity() , ProximityRegex() |
var products = await dbContext
.Products.Where(p =>
EF.Functions.MatchDisjunction(
p.Description,
"with shoes and",
Fuzzy.With(1),
Boost.With(2.3f)
)
)
.Select(p => new
{
p.Id,
p.Description,
Score = EF.Functions.Score(p.Id),
})
.ToListAsync();
SELECT p.id AS "Id", p.description AS "Description", COALESCE(paradedb.score(p.id), 0) AS "Score"
FROM products AS p
WHERE p.description ||| 'with shoes and'::fuzzy(1)::boost(2.3)