* fix the relationship between entities TheadPost and `LatestReplie…

…r` should be many-to-one instead of one-to-one @ `CrawlerDbContext.OnModelCreating()`

* change param `existingPostPredicate` with type of expression to `postQueryTransformer` with callback @ `PostSaver.Save()`
* rename field `UserId` to `Uid` @ ReplySignature.cs
* move `ValueTuple<,>` to record `UniqueAuthorRevision` @ AuthorRevisionSaver.cs
@ c#/crawler

c#/crawler/src/Db/CrawlerDbContext.cs

@@ -71,11 +71,10 @@ protected override void OnModelCreating(ModelBuilder b)
         b.Entity<User>().ToTable("tbmc_user");
         b.Entity<LatestReplier>().ToTable("tbmc_latestReplier");
         b.Entity<LatestReplier>().Property(e => e.DisplayName).HasConversion<byte[]>();
-        b.Entity<LatestReplier>().HasOne<ThreadPost>().WithOne(e => e.LatestReplier)
-            .HasForeignKey<ThreadPost>(e => e.LatestReplierId);
         b.Entity<LatestReplierRevision>().ToTable("tbmcr_latestReplier").HasKey(e => new {e.TakenAt, e.Uid});
         b.Entity<LatestReplierRevision>().Property(e => e.DisplayName).HasConversion<byte[]>();
-        b.Entity<ThreadPost>().ToTable($"tbmc_f{Fid}_thread");
+        b.Entity<ThreadPost>().ToTable($"tbmc_f{Fid}_thread")
+            .HasOne<LatestReplier>(e => e.LatestReplier).WithMany().HasForeignKey(e => e.LatestReplierId);
         b.Entity<ThreadMissingFirstReply>().ToTable("tbmc_thread_missingFirstReply");
         b.Entity<ReplyPost>().ToTable($"tbmc_f{Fid}_reply");
         b.Entity<ReplyContent>().ToTable($"tbmc_f{Fid}_reply_content");

c#/crawler/src/Db/Post/ReplySignature.cs

@@ -3,7 +3,7 @@ namespace tbm.Crawler.Db.Post;
 
 public class ReplySignature : RowVersionedEntity
 {
-    public long UserId { get; set; }
+    public long Uid { get; set; }
     public uint SignatureId { get; set; }
     public required byte[] XxHash3 { get; set; }
     public required byte[] ProtoBufBytes { get; set; }

c#/crawler/src/Tieba/Crawl/Saver/Post/PostSaver.cs

@@ -27,11 +27,10 @@ protected SaverChangeSet<TPost> Save<TRevision>(
         CrawlerDbContext db,
         Func<TPost, PostId> postIdSelector,
         Func<TPost, TRevision> revisionFactory,
-        ExpressionStarter<TPost> existingPostPredicate)
+        Func<IQueryable<TPost>, IQueryable<TPost>> postQueryTransformer)
         where TRevision : TBaseRevision
     {
-        var existingPostsKeyById = db.Set<TPost>().AsTracking()
-            .Where(existingPostPredicate).ToDictionary(postIdSelector);
+        var existingPostsKeyById = postQueryTransformer(db.Set<TPost>().AsTracking()).ToDictionary(postIdSelector);
 
         // clone before entities get mutated by SaverWithRevision.SaveEntitiesWithRevision()
         var existingPostsBeforeMerge = existingPostsKeyById.Select(pair => (TPost)pair.Value.Clone()).ToList();

c#/crawler/src/Tieba/Crawl/Saver/Post/ReplySaver.cs

@@ -15,7 +15,7 @@ public override SaverChangeSet<ReplyPost> Save(CrawlerDbContext db)
     {
         var changeSet = Save(db, r => r.Pid,
             r => new ReplyRevision {TakenAt = r.UpdatedAt ?? r.CreatedAt, Pid = r.Pid},
-            LinqKit.PredicateBuilder.New<ReplyPost>(r => Posts.Keys.Contains(r.Pid)));
+            posts => posts.Where(r => Posts.Keys.Contains(r.Pid)));
 
         db.ReplyContents.AddRange(changeSet.NewlyAdded // https://github.com/dotnet/efcore/issues/33945
             .Select(r => new ReplyContent {Pid = r.Pid, ProtoBufBytes = r.Content}));

c#/crawler/src/Tieba/Crawl/Saver/Post/SubReplySaver.cs

@@ -13,7 +13,7 @@ public override SaverChangeSet<SubReplyPost> Save(CrawlerDbContext db)
     {
         var changeSet = Save(db, sr => sr.Spid,
             sr => new SubReplyRevision {TakenAt = sr.UpdatedAt ?? sr.CreatedAt, Spid = sr.Spid},
-            LinqKit.PredicateBuilder.New<SubReplyPost>(sr => Posts.Keys.Contains(sr.Spid)));
+            posts => posts.Where(sr => Posts.Keys.Contains(sr.Spid)));
 
         db.SubReplyContents.AddRange(changeSet.NewlyAdded.Select(sr => // https://github.com/dotnet/efcore/issues/33945
             new SubReplyContent {Spid = sr.Spid, ProtoBufBytes = sr.Content}));

c#/crawler/src/Tieba/Crawl/Saver/Post/ThreadSaver.cs

@@ -17,7 +17,7 @@ public override SaverChangeSet<ThreadPost> Save(CrawlerDbContext db)
         var changeSet = Save(db,
             th => th.Tid,
             th => new ThreadRevision {TakenAt = th.UpdatedAt ?? th.CreatedAt, Tid = th.Tid},
-            PredicateBuilder.New<ThreadPost>(th => Posts.Keys.Contains(th.Tid)));
+            posts => posts.Where(th => Posts.Keys.Contains(th.Tid)));
 
         PostSaveHandlers += threadLatestReplierSaver.SaveFromThread(db, changeSet.AllAfter);
 

c#/crawler/src/Tieba/Crawl/Saver/Related/AuthorRevisionSaver.cs

@@ -1,14 +1,12 @@
 namespace tbm.Crawler.Tieba.Crawl.Saver.Related;
 
-// locks only using AuthorRevision.Fid and Uid, ignoring TriggeredBy
-// this prevents inserting multiple entities with similar time and other fields with the same values
 public class AuthorRevisionSaver(
     ILogger<AuthorRevisionSaver> logger,
-    SaverLocks<(Fid Fid, Uid Uid)>.New saverLocksFactory,
+    SaverLocks<AuthorRevisionSaver.UniqueAuthorRevision>.New saverLocksFactory,
     PostType triggeredByPostType)
 {
-    private static readonly HashSet<(Fid Fid, Uid Uid)> GlobalLockedAuthorExpGradeKeys = [];
-    private readonly Lazy<SaverLocks<(Fid Fid, Uid Uid)>> _authorExpGradeLocksSaverLocks =
+    private static readonly HashSet<UniqueAuthorRevision> GlobalLockedAuthorExpGradeKeys = [];
+    private readonly Lazy<SaverLocks<UniqueAuthorRevision>> _authorExpGradeLocksSaverLocks =
         new(() => saverLocksFactory(GlobalLockedAuthorExpGradeKeys));
 
     public delegate AuthorRevisionSaver New(PostType triggeredByPostType);
@@ -42,7 +40,7 @@ public Action SaveAuthorExpGradeRevisions<TPostWithAuthorExpGrade>
     private void Save<TPost, TRevision, TValue>(
         CrawlerDbContext db,
         IReadOnlyCollection<TPost> posts,
-        SaverLocks<(Fid Fid, Uid Uid)> locks,
+        SaverLocks<UniqueAuthorRevision> locks,
         IQueryable<TRevision> dbSet,
         Func<TPost, TValue?> postRevisioningFieldSelector,
         Func<TValue?, TValue?, bool> isValueChangedPredicate,
@@ -85,12 +83,16 @@ private void Save<TPost, TRevision, TValue>(
         var newRevisions = newRevisionOfNewUsers
             .Concat(newRevisionOfExistingUsers)
             .Select(revisionFactory)
-            .ToDictionary(revision => (revision.Fid, revision.Uid), revision => revision);
+            .ToDictionary(revision => new UniqueAuthorRevision(revision.Fid, revision.Uid), revision => revision);
         db.Set<TRevision>().AddRange(newRevisions
             .IntersectByKey(locks.Acquire(newRevisions.Keys))
             .Values());
     }
 
+    // locking key only using AuthorRevision.Fid and Uid, ignoring TriggeredBy
+    // this prevents inserting multiple entities with similar time and other fields with the same values
+    public record UniqueAuthorRevision(Fid Fid, Uid Uid);
+
     private sealed class LatestAuthorRevisionProjection<TValue>
     {
         public Time DiscoveredAt { get; init; }

c#/crawler/src/Tieba/Crawl/Saver/Related/ReplySignatureSaver.cs

@@ -21,7 +21,7 @@ public Action Save(CrawlerDbContext db, IEnumerable<ReplyPost> replies)
             .DistinctBy(r => r.SignatureId)
             .Select(r => new ReplySignature
             {
-                UserId = r.AuthorUid,
+                Uid = r.AuthorUid,
                 SignatureId = (uint)r.SignatureId!,
                 XxHash3 = XxHash3.Hash(r.Signature!),
                 ProtoBufBytes = r.Signature!,