Filters: words and posts a persona would rather not see

Mastodon's v2 filters replace the empty stubs: a filter's title, contexts,
action (warn, hide, blur) and expiry, its keywords (whole words or not,
taken as JSON objects, listed or numbered form fields, with id and _destroy
on update) and its statuses, each with their own endpoints; the v1 API is
the same filters seen keyword by keyword. Every status a persona reads
carries the filters it matches in `filtered` (a boost as what it boosts),
matched as Mastodon matches: warning, title, text, poll options and media
descriptions. Clients apply context and action. Filters never federate, and
go with a deleted persona.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LsXgEaXee4GCU1hwYgPJXw

PrivaPub/Domain/Social/Filters.cs

@@ -0,0 +1,115 @@
+using System.Text.RegularExpressions;
+
+using MongoDB.Entities;
+
+using PrivaPub.Api.Mastodon.Entities;
+using PrivaPub.Models.Social;
+
+using PostEntity = PrivaPub.Models.Post.Post;
+
+namespace PrivaPub.Domain.Social
+{
+	// A viewer's active filters, matched against posts as Mastodon matches them: a keyword anywhere in the post's words
+	// (its warning, title, text, poll options and media descriptions), case-insensitive, whole words when asked (a word
+	// boundary only where the keyword starts or ends with a word character); a filtered status by its id.
+	public sealed class Filters
+	{
+		static readonly AngleSharp.Html.Parser.HtmlParser Html = new();
+
+		readonly List<(PersonaFilter Filter, FilterEntity View, List<(string Keyword, Regex Pattern)> Keywords, HashSet<string> PostIds)> _active;
+
+		Filters(List<PersonaFilter> filters)
+		{
+			var now = DateTime.UtcNow;
+			_active = filters.Where(f => f.IsActive(now))
+				.Select(f => (f, View(f), f.Keywords.Select(k => (k.Keyword, Pattern(k))).ToList(), f.Statuses.Select(s => s.PostId).ToHashSet()))
+				.ToList();
+		}
+
+		public static readonly Filters None = new(new List<PersonaFilter>());
+
+		public static async Task<Filters> Of(string viewerId, CancellationToken token) =>
+			viewerId == default ? None : new Filters(await DB.Default.Find<PersonaFilter>().Match(f => f.AvatarId == viewerId).ExecuteAsync(token));
+
+		public bool IsEmpty => _active.Count == 0;
+
+		public static Regex Pattern(FilterKeyword keyword)
+		{
+			var escaped = Regex.Escape(keyword.Keyword ?? string.Empty);
+			if (keyword.WholeWord)
+			{
+				var start = Regex.IsMatch(keyword.Keyword ?? string.Empty, @"^\w") ? @"\b" : string.Empty;
+				var end = Regex.IsMatch(keyword.Keyword ?? string.Empty, @"\w$") ? @"\b" : string.Empty;
+				escaped = start + escaped + end;
+			}
+			return new Regex(escaped, RegexOptions.IgnoreCase | RegexOptions.CultureInvariant, TimeSpan.FromMilliseconds(50));
+		}
+
+		public List<FilterResultEntity> Of(PostEntity post)
+		{
+			if (_active.Count == 0 || post == default)
+				return new List<FilterResultEntity>();
+			string words = default;
+			var results = new List<FilterResultEntity>();
+			foreach (var (filter, view, keywords, postIds) in _active)
+			{
+				var keywordMatches = new List<string>();
+				if (keywords.Count > 0)
+				{
+					words ??= Words(post);
+					foreach (var (keyword, pattern) in keywords)
+						if (Matches(pattern, words))
+							keywordMatches.Add(keyword);
+				}
+				var statusMatches = postIds.Contains(post.ID) ? new List<string> { filter.Statuses.First(s => s.PostId == post.ID).Id } : new List<string>();
+				if (keywordMatches.Count > 0 || statusMatches.Count > 0)
+					results.Add(new FilterResultEntity
+					{
+						Filter = view,
+						KeywordMatches = keywordMatches.Count > 0 ? keywordMatches : default,
+						StatusMatches = statusMatches.Count > 0 ? statusMatches : default
+					});
+			}
+			return results;
+		}
+
+		static bool Matches(Regex pattern, string words)
+		{
+			try
+			{
+				return pattern.IsMatch(words);
+			}
+			catch (RegexMatchTimeoutException)
+			{
+				return false;
+			}
+		}
+
+		static string Words(PostEntity post)
+		{
+			var text = !string.IsNullOrEmpty(post.Text) && post.IsFederatedCopy == false
+				? post.Text
+				: string.IsNullOrEmpty(post.ContentHtml) ? post.Text : Html.ParseDocument($"<body>{post.ContentHtml}</body>").Body!.TextContent;
+			return string.Join("\n", new[] { post.SpoilerText, post.Title, text }
+				.Concat(post.Poll?.Options.Select(o => o.Title) ?? Enumerable.Empty<string>())
+				.Concat(post.Media.Select(m => m.Description))
+				.Where(s => !string.IsNullOrEmpty(s)));
+		}
+
+		public static FilterEntity View(PersonaFilter filter) => new()
+		{
+			Id = filter.ID,
+			Title = filter.Title,
+			Context = filter.Context,
+			ExpiresAt = filter.ExpiresAt.HasValue ? Api.Mastodon.Infrastructure.MastodonJson.Time(filter.ExpiresAt.Value) : default,
+			FilterAction = filter.Action,
+			Keywords = filter.Keywords.Select(Keyword).ToList(),
+			Statuses = filter.Statuses.Select(Status).ToList()
+		};
+
+		public static FilterKeywordEntity Keyword(FilterKeyword keyword) =>
+			new() { Id = keyword.Id, Keyword = keyword.Keyword, WholeWord = keyword.WholeWord };
+
+		public static FilterStatusEntity Status(FilterStatus status) => new() { Id = status.Id, StatusId = status.PostId };
+	}
+}