feat(search): fuzzy/typo tolerance (pg_trgm) (#16)
CI / backend (push) Successful in 48s
CI / frontend (push) Successful in 13s
Deploy Staging / deploy (push) Successful in 25s
Security / secrets (push) Successful in 3s
Security / dependencies (push) Successful in 55s
CI / backend (pull_request) Successful in 49s
CI / frontend (pull_request) Successful in 12s
Security / secrets (pull_request) Successful in 3s
Security / dependencies (pull_request) Successful in 53s

This commit was merged in pull request #16.
This commit is contained in:
2026-07-01 22:45:10 +02:00
parent 9bbab5d32a
commit c0c1777d3f
6 changed files with 813 additions and 21 deletions
+7 -4
View File
@@ -2,6 +2,13 @@ using InboxIntel.Domain.Enums;
namespace InboxIntel.Application.DTOs;
/// <summary>
/// Row summary for search/list results. <c>MatchHighlight</c> is a "why this matched" body
/// fragment (ts_headline) with matched terms wrapped in U+E000/U+E001 sentinels — NOT HTML;
/// the client renders them as escaped &lt;mark&gt; elements, so untrusted email content can
/// never inject markup. Null unless the search had a free-text query; optional/last so other
/// DTO constructors are unaffected.
/// </summary>
public record EmailSummaryDto(
Guid Id,
string GmailMessageId,
@@ -17,10 +24,6 @@ public record EmailSummaryDto(
EmailCategory Category,
bool HasListUnsubscribe,
bool SupportsOneClick,
// "Why this matched": a ts_headline fragment of the body with matched terms wrapped in
// U+E000/U+E001 sentinels (NOT HTML — the client renders them as escaped <mark> elements,
// so untrusted email content can never inject markup). Null unless the search had a
// free-text query. Optional/last so other DTO constructors are unaffected.
string? MatchHighlight = null);
/// <summary>Full single-email view, including body text, for the detail pane.</summary>
@@ -0,0 +1,743 @@
// <auto-generated />
using System;
using InboxIntel.Infrastructure.Persistence;
using Microsoft.EntityFrameworkCore;
using Microsoft.EntityFrameworkCore.Infrastructure;
using Microsoft.EntityFrameworkCore.Migrations;
using Microsoft.EntityFrameworkCore.Storage.ValueConversion;
using Npgsql.EntityFrameworkCore.PostgreSQL.Metadata;
using NpgsqlTypes;
#nullable disable
namespace InboxIntel.Infrastructure.Migrations
{
[DbContext(typeof(AppDbContext))]
[Migration("20260701203001_EnablePgTrgm")]
partial class EnablePgTrgm
{
/// <inheritdoc />
protected override void BuildTargetModel(ModelBuilder modelBuilder)
{
#pragma warning disable 612, 618
modelBuilder
.HasAnnotation("ProductVersion", "8.0.4")
.HasAnnotation("Relational:MaxIdentifierLength", 63);
NpgsqlModelBuilderExtensions.HasPostgresExtension(modelBuilder, "pg_trgm");
NpgsqlModelBuilderExtensions.UseIdentityByDefaultColumns(modelBuilder);
modelBuilder.Entity("InboxIntel.Domain.Entities.AnalyticsAggregate", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<DateOnly>("Day")
.HasColumnType("date");
b.Property<string>("HourHistogramJson")
.HasColumnType("text");
b.Property<int>("NewsletterCount")
.HasColumnType("integer");
b.Property<int>("TotalReceived")
.HasColumnType("integer");
b.Property<long>("TotalSizeBytes")
.HasColumnType("bigint");
b.Property<int>("TotalUnread")
.HasColumnType("integer");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.Property<int>("WithAttachments")
.HasColumnType("integer");
b.HasKey("Id");
b.HasIndex("UserId", "Day")
.IsUnique();
b.ToTable("analytics_aggregates", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Attachment", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("EmailId")
.HasColumnType("uuid");
b.Property<string>("FileName")
.IsRequired()
.HasMaxLength(512)
.HasColumnType("character varying(512)");
b.Property<string>("GmailAttachmentId")
.HasColumnType("text");
b.Property<string>("MimeType")
.HasMaxLength(255)
.HasColumnType("character varying(255)");
b.Property<long>("SizeBytes")
.HasColumnType("bigint");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("EmailId");
b.HasIndex("UserId", "MimeType");
b.ToTable("attachments", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Email", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<string>("BodyText")
.HasColumnType("text");
b.Property<int>("Category")
.HasColumnType("integer");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<string>("GmailMessageId")
.IsRequired()
.HasMaxLength(64)
.HasColumnType("character varying(64)");
b.Property<bool>("HasAttachments")
.HasColumnType("boolean");
b.Property<bool>("HasListUnsubscribe")
.HasColumnType("boolean");
b.Property<bool>("IsImportant")
.HasColumnType("boolean");
b.Property<bool>("IsInInbox")
.HasColumnType("boolean");
b.Property<bool>("IsStarred")
.HasColumnType("boolean");
b.Property<bool>("IsTrashed")
.HasColumnType("boolean");
b.Property<bool>("IsUnread")
.HasColumnType("boolean");
b.Property<string>("ListUnsubscribeRaw")
.HasMaxLength(2048)
.HasColumnType("character varying(2048)");
b.Property<DateTimeOffset?>("ReceivedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<NpgsqlTsVector>("SearchVector")
.ValueGeneratedOnAddOrUpdate()
.HasColumnType("tsvector")
.HasComputedColumnSql("setweight(to_tsvector('english', coalesce(\"Subject\",'')), 'A') || setweight(to_tsvector('english', coalesce(\"BodyText\",'')), 'B')", true);
b.Property<Guid>("SenderId")
.HasColumnType("uuid");
b.Property<DateTimeOffset>("SentAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<long>("SizeEstimateBytes")
.HasColumnType("bigint");
b.Property<string>("Snippet")
.HasMaxLength(2048)
.HasColumnType("character varying(2048)");
b.Property<string>("Subject")
.HasMaxLength(1024)
.HasColumnType("character varying(1024)");
b.Property<bool>("SupportsOneClickUnsubscribe")
.HasColumnType("boolean");
b.Property<Guid>("ThreadId")
.HasColumnType("uuid");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("SearchVector");
NpgsqlIndexBuilderExtensions.HasMethod(b.HasIndex("SearchVector"), "GIN");
b.HasIndex("SenderId");
b.HasIndex("ThreadId");
b.HasIndex("UserId", "Category");
b.HasIndex("UserId", "GmailMessageId")
.IsUnique();
b.HasIndex("UserId", "IsInInbox");
b.HasIndex("UserId", "IsUnread");
b.HasIndex("UserId", "SenderId");
b.HasIndex("UserId", "SentAtUtc");
b.ToTable("emails", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.EmailLabel", b =>
{
b.Property<Guid>("EmailId")
.HasColumnType("uuid");
b.Property<Guid>("LabelId")
.HasColumnType("uuid");
b.HasKey("EmailId", "LabelId");
b.HasIndex("LabelId");
b.ToTable("email_labels", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Label", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<string>("ColorHex")
.HasColumnType("text");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<string>("GmailLabelId")
.IsRequired()
.HasMaxLength(64)
.HasColumnType("character varying(64)");
b.Property<string>("Name")
.IsRequired()
.HasMaxLength(255)
.HasColumnType("character varying(255)");
b.Property<string>("Type")
.IsRequired()
.HasColumnType("text");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("UserId", "GmailLabelId")
.IsUnique();
b.ToTable("labels", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.MailDomain", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("EmailCount")
.HasColumnType("integer");
b.Property<bool>("IsBulkSender")
.HasColumnType("boolean");
b.Property<string>("Name")
.IsRequired()
.HasMaxLength(255)
.HasColumnType("character varying(255)");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("UserId", "Name")
.IsUnique();
b.ToTable("domains", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.MailThread", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<DateTimeOffset?>("FirstMessageUtc")
.HasColumnType("timestamp with time zone");
b.Property<string>("GmailThreadId")
.IsRequired()
.HasMaxLength(64)
.HasColumnType("character varying(64)");
b.Property<DateTimeOffset?>("LastMessageUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("MessageCount")
.HasColumnType("integer");
b.Property<string>("Snippet")
.HasColumnType("text");
b.Property<string>("Subject")
.HasMaxLength(1024)
.HasColumnType("character varying(1024)");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("UserId", "GmailThreadId")
.IsUnique();
b.ToTable("threads", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Sender", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<string>("Address")
.IsRequired()
.HasMaxLength(320)
.HasColumnType("character varying(320)");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<string>("DisplayName")
.HasMaxLength(255)
.HasColumnType("character varying(255)");
b.Property<Guid>("DomainId")
.HasColumnType("uuid");
b.Property<int>("EmailCount")
.HasColumnType("integer");
b.Property<bool>("HasUnsubscribe")
.HasColumnType("boolean");
b.Property<DateTimeOffset?>("LastReceivedUtc")
.HasColumnType("timestamp with time zone");
b.Property<long>("TotalSizeBytes")
.HasColumnType("bigint");
b.Property<int>("UnreadCount")
.HasColumnType("integer");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("DomainId");
b.HasIndex("UserId", "Address")
.IsUnique();
b.HasIndex("UserId", "EmailCount");
b.ToTable("senders", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.SyncState", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset?>("CompletedUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("ConsecutiveFailures")
.HasColumnType("integer");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<string>("LastError")
.HasMaxLength(4000)
.HasColumnType("character varying(4000)");
b.Property<string>("LastHistoryId")
.HasColumnType("text");
b.Property<DateTimeOffset?>("LastSuccessfulSyncUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("LastSyncType")
.HasColumnType("integer");
b.Property<int>("MessagesProcessed")
.HasColumnType("integer");
b.Property<string>("ResumePageToken")
.HasColumnType("text");
b.Property<DateTimeOffset?>("StartedUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("Status")
.HasColumnType("integer");
b.Property<int>("TotalMessagesEstimate")
.HasColumnType("integer");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("UserId")
.IsUnique();
b.ToTable("sync_states", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.UnsubscribeItem", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<double>("Confidence")
.HasColumnType("double precision");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("EmailCount")
.HasColumnType("integer");
b.Property<DateTimeOffset?>("LastAttemptUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("Method")
.HasColumnType("integer");
b.Property<string>("ResultMessage")
.HasColumnType("text");
b.Property<Guid>("SenderId")
.HasColumnType("uuid");
b.Property<int>("Status")
.HasColumnType("integer");
b.Property<string>("UnsubscribeTarget")
.HasMaxLength(2048)
.HasColumnType("character varying(2048)");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.HasKey("Id");
b.HasIndex("SenderId");
b.HasIndex("UserId", "SenderId")
.IsUnique();
b.ToTable("unsubscribe_items", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.User", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset?>("AccessTokenExpiresAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<bool>("DigestEnabled")
.HasColumnType("boolean");
b.Property<string>("DisplayName")
.HasColumnType("text");
b.Property<string>("Email")
.IsRequired()
.HasMaxLength(320)
.HasColumnType("character varying(320)");
b.Property<byte[]>("EncryptedRefreshToken")
.HasColumnType("bytea");
b.Property<string>("GoogleSubjectId")
.IsRequired()
.HasMaxLength(64)
.HasColumnType("character varying(64)");
b.Property<DateTimeOffset?>("LastDigestSentUtc")
.HasColumnType("timestamp with time zone");
b.Property<DateTimeOffset?>("LastLoginUtc")
.HasColumnType("timestamp with time zone");
b.Property<string>("PictureUrl")
.HasColumnType("text");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.HasKey("Id");
b.HasIndex("Email")
.IsUnique();
b.HasIndex("GoogleSubjectId")
.IsUnique();
b.ToTable("users", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.WidgetLayout", b =>
{
b.Property<Guid>("Id")
.ValueGeneratedOnAdd()
.HasColumnType("uuid");
b.Property<DateTimeOffset>("CreatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<int>("H")
.HasColumnType("integer");
b.Property<string>("SettingsJson")
.HasColumnType("text");
b.Property<int>("SortOrder")
.HasColumnType("integer");
b.Property<DateTimeOffset?>("UpdatedAtUtc")
.HasColumnType("timestamp with time zone");
b.Property<Guid>("UserId")
.HasColumnType("uuid");
b.Property<bool>("Visible")
.HasColumnType("boolean");
b.Property<int>("W")
.HasColumnType("integer");
b.Property<string>("WidgetKey")
.IsRequired()
.HasMaxLength(64)
.HasColumnType("character varying(64)");
b.Property<int>("X")
.HasColumnType("integer");
b.Property<int>("Y")
.HasColumnType("integer");
b.HasKey("Id");
b.HasIndex("UserId", "WidgetKey")
.IsUnique();
b.ToTable("widget_layouts", (string)null);
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Attachment", b =>
{
b.HasOne("InboxIntel.Domain.Entities.Email", "Email")
.WithMany("Attachments")
.HasForeignKey("EmailId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
b.Navigation("Email");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Email", b =>
{
b.HasOne("InboxIntel.Domain.Entities.Sender", "Sender")
.WithMany("Emails")
.HasForeignKey("SenderId")
.OnDelete(DeleteBehavior.Restrict)
.IsRequired();
b.HasOne("InboxIntel.Domain.Entities.MailThread", "Thread")
.WithMany("Emails")
.HasForeignKey("ThreadId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
b.HasOne("InboxIntel.Domain.Entities.User", null)
.WithMany("Emails")
.HasForeignKey("UserId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
b.Navigation("Sender");
b.Navigation("Thread");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.EmailLabel", b =>
{
b.HasOne("InboxIntel.Domain.Entities.Email", "Email")
.WithMany("EmailLabels")
.HasForeignKey("EmailId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
b.HasOne("InboxIntel.Domain.Entities.Label", "Label")
.WithMany("EmailLabels")
.HasForeignKey("LabelId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
b.Navigation("Email");
b.Navigation("Label");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Sender", b =>
{
b.HasOne("InboxIntel.Domain.Entities.MailDomain", "Domain")
.WithMany("Senders")
.HasForeignKey("DomainId")
.OnDelete(DeleteBehavior.Restrict)
.IsRequired();
b.Navigation("Domain");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.UnsubscribeItem", b =>
{
b.HasOne("InboxIntel.Domain.Entities.Sender", "Sender")
.WithMany()
.HasForeignKey("SenderId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
b.Navigation("Sender");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.WidgetLayout", b =>
{
b.HasOne("InboxIntel.Domain.Entities.User", null)
.WithMany("WidgetLayouts")
.HasForeignKey("UserId")
.OnDelete(DeleteBehavior.Cascade)
.IsRequired();
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Email", b =>
{
b.Navigation("Attachments");
b.Navigation("EmailLabels");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Label", b =>
{
b.Navigation("EmailLabels");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.MailDomain", b =>
{
b.Navigation("Senders");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.MailThread", b =>
{
b.Navigation("Emails");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.Sender", b =>
{
b.Navigation("Emails");
});
modelBuilder.Entity("InboxIntel.Domain.Entities.User", b =>
{
b.Navigation("Emails");
b.Navigation("WidgetLayouts");
});
#pragma warning restore 612, 618
}
}
}
@@ -0,0 +1,24 @@
using Microsoft.EntityFrameworkCore.Migrations;
#nullable disable
namespace InboxIntel.Infrastructure.Migrations
{
/// <inheritdoc />
public partial class EnablePgTrgm : Migration
{
/// <inheritdoc />
protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.AlterDatabase()
.Annotation("Npgsql:PostgresExtension:pg_trgm", ",,");
}
/// <inheritdoc />
protected override void Down(MigrationBuilder migrationBuilder)
{
migrationBuilder.AlterDatabase()
.OldAnnotation("Npgsql:PostgresExtension:pg_trgm", ",,");
}
}
}
@@ -21,6 +21,7 @@ namespace InboxIntel.Infrastructure.Migrations
.HasAnnotation("ProductVersion", "8.0.4")
.HasAnnotation("Relational:MaxIdentifierLength", 63);
NpgsqlModelBuilderExtensions.HasPostgresExtension(modelBuilder, "pg_trgm");
NpgsqlModelBuilderExtensions.UseIdentityByDefaultColumns(modelBuilder);
modelBuilder.Entity("InboxIntel.Domain.Entities.AnalyticsAggregate", b =>
@@ -65,6 +65,9 @@ public class AppDbContext : DbContext, IAppDbContext
// it otherwise (e.g. the InMemory provider used by tests).
if (Database.IsRelational())
{
// pg_trgm powers the fuzzy/typo fallback in SearchService (word_similarity).
modelBuilder.HasPostgresExtension("pg_trgm");
modelBuilder.Entity<Email>().Property(e => e.SearchVector)
.HasColumnType("tsvector")
.HasComputedColumnSql(
@@ -1,6 +1,7 @@
using InboxIntel.Application.Abstractions;
using InboxIntel.Application.Common;
using InboxIntel.Application.DTOs;
using InboxIntel.Domain.Entities;
using InboxIntel.Domain.Enums;
using InboxIntel.Infrastructure.Persistence;
using Microsoft.EntityFrameworkCore;
@@ -64,30 +65,47 @@ public class SearchService : ISearchService
// OR, and -exclusions — the syntax users already expect from web search boxes.
var hasFreeTextQuery = !string.IsNullOrWhiteSpace(r.Query);
var term = r.Query?.Trim() ?? string.Empty;
if (hasFreeTextQuery)
q = q.Where(e => e.SearchVector!.Matches(EF.Functions.WebSearchToTsQuery("english", term)));
var total = await q.CountAsync(ct);
var structured = q; // structured filters only — reused by the fuzzy fallback below.
var matched = hasFreeTextQuery
? structured.Where(e => e.SearchVector!.Matches(EF.Functions.WebSearchToTsQuery("english", term)))
: structured;
// Relevance-ranked when there's a free-text query (ts_rank_cd via RankCoverDensity,
// recency as a tiebreaker); date-only otherwise — matches the existing browse
// behaviour when the user isn't searching for anything in particular.
var ranked = hasFreeTextQuery
? q.OrderByDescending(e => e.SearchVector!.RankCoverDensity(EF.Functions.WebSearchToTsQuery("english", term)))
.ThenByDescending(e => e.SentAtUtc)
: q.OrderByDescending(e => e.SentAtUtc);
var total = await matched.CountAsync(ct);
// Fuzzy/typo fallback: ONLY when a free-text search found nothing exact. word_similarity
// with an explicit 0.3 threshold — pg_trgm's default 0.6 misses real typos
// ("recieved" -> "received" scores ~0.39). Rare path, so the (non-indexed) scan over the
// user's own structured subset is acceptable. Needs the pg_trgm extension (see migration).
const double fuzzyThreshold = 0.3;
var fuzzy = hasFreeTextQuery && total == 0;
if (fuzzy)
{
matched = structured.Where(e =>
e.Subject != null && EF.Functions.TrigramsWordSimilarity(term, e.Subject) >= fuzzyThreshold);
total = await matched.CountAsync(ct);
}
// Ordering: fuzzy → by word similarity; exact free-text → ts_rank_cd; browse → date.
IQueryable<Email> ranked;
if (fuzzy)
ranked = matched.OrderByDescending(e => EF.Functions.TrigramsWordSimilarity(term, e.Subject!))
.ThenByDescending(e => e.SentAtUtc);
else if (hasFreeTextQuery)
ranked = matched.OrderByDescending(e => e.SearchVector!.RankCoverDensity(EF.Functions.WebSearchToTsQuery("english", term)))
.ThenByDescending(e => e.SentAtUtc);
else
ranked = matched.OrderByDescending(e => e.SentAtUtc);
var paged = ranked.Skip((r.Page - 1) * r.PageSize).Take(r.PageSize);
// Two unconditional projections (no DB function inside a C# ternary → no doubt about
// EF translation). The browse path never touches ts_headline, so it's byte-for-byte
// unchanged AND safe under the InMemory test provider.
// "Why this matched" ts_headline only for EXACT free-text hits (fuzzy/browse get no
// highlight — a fuzzy hit has no literal match to headline). Unconditional projections
// (no DB function in a ternary) keep EF translation unambiguous.
List<EmailSummaryDto> items;
if (hasFreeTextQuery)
if (hasFreeTextQuery && !fuzzy)
{
// "Why this matched": ts_headline body fragment with matched terms wrapped in
// U+E000/U+E001 sentinels (safe, non-HTML — the client renders them as escaped
// <mark> spans; see EmailSummaryDto).
// Matched terms wrapped in U+E000/U+E001 sentinels (safe, non-HTML; see EmailSummaryDto).
var headlineOpts =
$"StartSel={(char)0xE000},StopSel={(char)0xE001},MaxWords=16,MinWords=5,ShortWord=2,HighlightAll=false";
items = await paged