P7: thread backfill reads conversation containers and asks with If-None-Match

The FetchReplies job reads a thread's FEP-171b contextHistory before its FEP-7888 context: the posts each Add (or
Forte's plain Create) brought in, read from their own servers. A thread collection read whole keeps its ETag when the
document itself changes with every post (it counts them, or holds them all with no further page); the next read sends
it as If-None-Match and a 304 ends the job. The ETag is kept as sent, since NodeBB's has no quotes and the typed header
drops it. A context naming a post we hold (Forte's first post) is not fetched.

Checked in the pasture: Mastodon 4.7.3 answers the second read 304; NodeBB 4.16's unquoted ETag is kept (23/23 in its
scenario); a Forte thread completes through its replies.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_012CzABvBkbcFqoHdmi8b9WB

PrivaPub/Federation/Actors/RemoteActorService.cs

@@ -17,6 +17,8 @@ namespace PrivaPub.Federation.Actors
 	public interface IRemoteActorService
 	{
 		Task<FetchedJson> FetchObject(string uri, CancellationToken token);
+		// the same, or NotModified while the version `etag` names is still current
+		Task<FetchedJson> FetchObject(string uri, string etag, CancellationToken token) => FetchObject(uri, token);
 		Task<ForeignAvatar> GetActor(string actorUri, bool refresh, CancellationToken token);
 		Task<ForeignAvatar> GetActorByKeyId(string keyId, bool refresh, CancellationToken token);
 		bool KeyTemporarilyUnavailable(string keyId) => false;
@@ -50,13 +52,17 @@ namespace PrivaPub.Federation.Actors
 			_options = options;
 		}
 
-		public async Task<FetchedJson> FetchObject(string uri, CancellationToken token)
+		public Task<FetchedJson> FetchObject(string uri, CancellationToken token) => FetchObject(uri, default, token);
+
+		public async Task<FetchedJson> FetchObject(string uri, string etag, CancellationToken token)
 		{
 			using var scope = HttpScope.Default("object");
 			var signer = await _localActors.GetInstanceActor(token);
-			var fetched = await Get(uri, signer, token);
-			if (fetched == default)
-				return default;
+			var fetched = string.IsNullOrEmpty(etag)
+				? await Get(uri, signer, token)
+				: await _http.GetJsonIfChanged(uri, Accept, etag, request => HttpSignatures.Sign(request, signer, body: null), token);
+			if (fetched == default || fetched.NotModified)
+				return fetched;
 
 			var id = Text(fetched.Root, "id");
 			if (Origin.IsDocumentAt(id, fetched.FinalUri))

PrivaPub/Federation/Inbox/RemoteReplies.cs

@@ -20,9 +20,11 @@ namespace PrivaPub.Federation.Inbox
 	public sealed record RepliesPayload(string PostId, int Depth);
 
 	// What was said under a remote post on other servers, read when a persona opens its thread, so the thread shows more
-	// than what happened to reach us. The thread's own collection comes first (FEP-7888's `context`, which Mastodon 4.5+
-	// serves with every reply at any depth); otherwise the post's `replies` (PeerTube's `comments`), and the replies'
-	// own, two levels down. Read with the instance actor's signature, never a persona's, a few pages and at most
+	// than what happened to reach us. The thread's own collections come first: its conversation container (FEP-171b's
+	// `contextHistory`, every activity its owner added to the thread), then its posts (FEP-7888's `context`, which
+	// Mastodon 4.5+, Mitra and NodeBB serve with every reply at any depth), sent the ETag of the last whole read so an
+	// unchanged thread costs one 304; otherwise the post's `replies` (PeerTube's `comments`, Forte's whole thread), and the
+	// replies' own, two levels down. Read with the instance actor's signature, never a persona's, a few pages and at most
 	// MaxItems posts a job; each one is fetched from its own server and stored like any post fetched for a thread
 	// (RemotePosts.StoreContext: public and unlisted only, its author checked against its origin).
 	public class RepliesJobHandler : IJobHandler
@@ -74,14 +76,35 @@ namespace PrivaPub.Federation.Inbox
 			if (await Document(post, token) is not { } document)
 				return JobOutcome.Done;
 
-			if (await Collection(document["context"], token) is { } thread)
+			foreach (var name in new[] { "contextHistory", "context" })
 			{
-				await StoreAll(thread, token);
+				var uri = Id(document[name]);
+				// Forte's `context` is the thread's first post
+				if (uri != default && await _dbEntities.Posts.Match(p => p.ObjectURI == uri).ExecuteAnyAsync(token))
+					continue;
+				var known = uri == default ? default : await DB.Default.Find<ThreadCollection>().Match(c => c.URI == uri).ExecuteFirstAsync(token);
+				var (thread, etag, unchanged) = await Collection(document[name], known?.ETag, token);
+				if (unchanged)
+				{
+					await DB.Default.Update<ThreadCollection>().MatchID(known.ID).Modify(c => c.ReadAt, DateTime.UtcNow).ExecuteAsync(token);
+					return JobOutcome.Done;
+				}
+				if (thread == default)
+					continue;
+				var (_, read) = await StoreAll(thread, name == "contextHistory", token);
+				var whole = read && Counted(thread);
+				if (uri != default && (known != default || whole && etag != default))
+					await DB.Default.Update<ThreadCollection>()
+						.Match(c => c.URI == uri)
+						.Modify(c => c.ETag, whole ? etag : default)
+						.Modify(c => c.ReadAt, DateTime.UtcNow)
+						.Option(o => o.IsUpsert = true)
+						.ExecuteAsync(token);
 				return JobOutcome.Done;
 			}
-			if (await Collection(document["replies"] ?? document["comments"], token) is not { } replies)
+			if ((await Collection(document["replies"] ?? document["comments"], default, token)).Collection is not { } replies)
 				return JobOutcome.Done;
-			var answers = await StoreAll(replies, token);
+			var (answers, _) = await StoreAll(replies, false, token);
 			if (payload.Depth < MaxDepth)
 				await _queue.EnqueueMany(answers.Where(a => a.IsFederatedCopy && a.AnsweringToPostId == post.ID).Take(MaxBranches)
 					.Select(a => For(a, payload.Depth + 1)), token);
@@ -100,35 +123,41 @@ namespace PrivaPub.Federation.Inbox
 			return fetched == default ? default : JsonNode.Parse(fetched.Root.GetRawText()) as JsonObject;
 		}
 
-		// a collection as its server serves it now: what a post carries is read again by its id, since an embedded first
-		// page is as old as the post; a value that is no web address (Pleroma's `context` is a tag: URI), or no collection,
-		// gives nothing
-		async Task<JsonObject> Collection(JsonNode node, CancellationToken token)
+		// a collection as its server serves it now, and its ETag: what a post carries is read again by its id, since an
+		// embedded first page is as old as the post; a value that is no web address (Pleroma's `context` is a tag: URI), or no
+		// collection, gives nothing, and one whose server still has the version `etag` names is unchanged
+		async Task<(JsonObject Collection, string ETag, bool Unchanged)> Collection(JsonNode node, string etag, CancellationToken token)
 		{
 			var uri = Id(node);
+			string served = default;
 			if (uri != default && Uri.TryCreate(uri, UriKind.Absolute, out var parsed) && (parsed.Scheme == Uri.UriSchemeHttps || parsed.Scheme == Uri.UriSchemeHttp))
 			{
 				using var scope = HttpScope.For("replies");
-				using var fetched = await _remoteActors.FetchObject(uri, token);
+				using var fetched = await _remoteActors.FetchObject(uri, etag, token);
+				if (fetched?.NotModified == true)
+					return (default, etag, true);
 				node = fetched == default ? default : JsonNode.Parse(fetched.Root.GetRawText());
+				served = fetched?.ETag;
 			}
-			return node is JsonObject collection && Value(collection, "type") is { } type && CollectionTypes.Contains(type) ? collection : default;
+			return (node is JsonObject collection && Value(collection, "type") is { } type && CollectionTypes.Contains(type) ? collection : default, served, false);
 		}
 
-		// every post the collection lists, stored when we lack it: its pages in order, at most MaxPages and MaxItems
-		async Task<List<PostEntity>> StoreAll(JsonObject collection, CancellationToken token)
+		// every post the collection lists, stored when we lack it: its pages in order, at most MaxPages and MaxItems; whether
+		// it was read to its end. A conversation container's items are its owner's Adds of the thread's activities.
+		async Task<(List<PostEntity> Posts, bool Whole)> StoreAll(JsonObject collection, bool activities, CancellationToken token)
 		{
 			var posts = new List<PostEntity>();
 			var seen = 0;
 			var page = collection["orderedItems"] is JsonArray || collection["items"] is JsonArray ? collection : await Page(collection["first"], token);
-			for (var pages = 0; page != default && pages < MaxPages && seen < MaxItems; pages++)
+			for (var pages = 0; page != default; pages++)
 			{
+				if (pages == MaxPages)
+					return (posts, false);
 				foreach (var item in Items(page))
 				{
 					if (++seen > MaxItems)
-						break;
-					// a context may list the activities that made the thread (FEP-f228) rather than its posts
-					var id = Value(item, "type") is "Create" or "Update" ? Id(item["object"]) : Id(item);
+						return (posts, false);
+					var id = activities ? await Added(item, token) : PostId(item);
 					if (id == default)
 						continue;
 					var post = await _dbEntities.Posts.Match(p => p.ObjectURI == id && !p.DeletedAt.HasValue).ExecuteFirstAsync(token)
@@ -138,7 +167,32 @@ namespace PrivaPub.Federation.Inbox
 				}
 				page = page["next"] is { } next ? await Page(next, token) : default;
 			}
-			return posts;
+			return (posts, true);
+		}
+
+		// whether the document itself changes with every post added to the thread, so its ETag is the thread's: it counts
+		// them (NodeBB), or holds every one with no further page (a short Mastodon thread); a long Mastodon thread's first
+		// page stays as it was while its last one grows
+		static bool Counted(JsonObject collection)
+		{
+			if (collection["totalItems"] is JsonValue)
+				return true;
+			var page = collection["orderedItems"] is JsonArray || collection["items"] is JsonArray ? collection : collection["first"] as JsonObject;
+			return page != default && (page["orderedItems"] is JsonArray || page["items"] is JsonArray) && page["next"] == default;
+		}
+
+		// a context may list the activities that made the thread (FEP-f228) rather than its posts
+		static string PostId(JsonNode item) => Value(item, "type") is "Create" or "Update" ? Id(item["object"]) : Id(item);
+
+		// the post an activity in a conversation container brought into the thread, added by its owner or (Forte's) not: a
+		// Create's or an Update's object; a like, a removal or anything else adds no post. Only its id is taken, and the post
+		// is read from its own server.
+		async Task<string> Added(JsonNode item, CancellationToken token)
+		{
+			var activity = await Page(item, token);
+			if (activity != default && Value(activity, "type") == "Add")
+				activity = await Page(activity["object"], token);
+			return activity != default && Value(activity, "type") is "Create" or "Update" ? Id(activity["object"]) : default;
 		}
 
 		async Task<JsonObject> Page(JsonNode node, CancellationToken token)