ML-155: Compact collection summary and add stats preamble

backlog/completed/ml-155 - Compact-collection-summary-format-statistical-preamble-for-chat-context.md

@@ -0,0 +1,26 @@
+---
+id: ML-155
+title: Compact collection summary format + statistical preamble for chat context
+status: Done
+assignee: []
+created_date: '2026-05-01 21:43'
+updated_date: '2026-05-01 21:49'
+labels: []
+dependencies: []
+priority: high
+---
+
+## Description
+
+<!-- SECTION:DESCRIPTION:BEGIN -->
+Reduce the token count of collection_summary/0 by:
+1. Compacting format_group/1 (year-only dates, remove type field, reduce max genres 3→2)
+2. Adding statistical preamble (genre/formats/decade distribution) computed inline from fetched records
+3. Updating tests for the new format
+<!-- SECTION:DESCRIPTION:END -->
+
+## Final Summary
+
+<!-- SECTION:FINAL_SUMMARY:BEGIN -->
+Compacted collection_summary format_group/1: year-only dates, removed type field, reduced max genres 3→2. Added statistical preamble (genre/formats/eras distribution + artist count) computed in-memory from already-fetched records. ~27% token reduction (~26.4k → ~19.4k for 1200 records). All 886 tests pass.
+<!-- SECTION:FINAL_SUMMARY:END -->

lib/music_library/collection.ex

@@ -197,7 +197,8 @@ defmodule MusicLibrary.Collection do
     |> MapSet.new()
   end
 
-  @max_genres_per_record 3
+  @max_genres_per_record 2
+  @stats_limit 10
 
   @spec collection_summary() :: {String.t(), non_neg_integer()}
   def collection_summary do
@@ -215,15 +216,92 @@ defmodule MusicLibrary.Collection do
       |> Enum.map(fn {_id, group} -> format_group(group) end)
       |> Enum.sort()
 
-    summary = Enum.join(groups, "\n")
+    catalog = Enum.join(groups, "\n")
+    stats = build_stats(records, length(groups))
+
+    summary =
+      cond do
+        stats == "" and catalog == "" -> ""
+        stats == "" -> catalog
+        catalog == "" -> stats
+        true -> stats <> "\n\n" <> catalog
+      end
 
     {summary, length(groups)}
   end
 
+  defp build_stats([], _group_count), do: ""
+
+  defp build_stats(records, group_count) do
+    genre_counts = compute_genre_counts(records)
+    format_counts = compute_format_counts(records)
+    decade_counts = compute_decade_counts(records)
+
+    artist_count =
+      records
+      |> Enum.flat_map(& &1.artists)
+      |> Enum.uniq_by(& &1.musicbrainz_id)
+      |> length()
+
+    [
+      "# Stats: #{group_count} releases, #{artist_count} artists",
+      genres_line(genre_counts),
+      formats_line(format_counts),
+      eras_line(decade_counts)
+    ]
+    |> Enum.join("\n")
+  end
+
+  defp compute_genre_counts(records) do
+    records
+    |> Enum.flat_map(&(&1.genres || []))
+    |> Enum.reject(&(&1 in @excluded_genres))
+    |> Enum.frequencies()
+    |> Enum.sort_by(&elem(&1, 1), :desc)
+    |> Enum.take(@stats_limit)
+  end
+
+  defp compute_format_counts(records) do
+    records
+    |> Enum.map(& &1.format)
+    |> Enum.frequencies()
+    |> Enum.sort_by(&elem(&1, 1), :desc)
+  end
+
+  defp compute_decade_counts(records) do
+    records
+    |> Enum.map(&extract_decade(&1.release_date))
+    |> Enum.reject(&is_nil/1)
+    |> Enum.frequencies()
+    |> Enum.sort_by(&elem(&1, 1), :desc)
+  end
+
+  defp genres_line([]), do: ""
+
+  defp genres_line(counts) do
+    items = Enum.map(counts, fn {genre, count} -> "#{genre} #{count}" end)
+    "Genres: " <> Enum.join(items, ", ")
+  end
+
+  defp formats_line([]), do: ""
+
+  defp formats_line(counts) do
+    items = Enum.map(counts, fn {format, count} -> "#{format} #{count}" end)
+    "Formats: " <> Enum.join(items, ", ")
+  end
+
+  defp eras_line([]), do: ""
+
+  defp eras_line(counts) do
+    items = Enum.map(counts, fn {decade, count} -> "#{decade} #{count}" end)
+    "Eras: " <> Enum.join(items, ", ")
+  end
+
   defp format_group(records) do
     record = hd(records)
     artist_names = Record.artist_names(record)
     formats = records |> Enum.map(& &1.format) |> Enum.uniq() |> Enum.join("/")
+    year = extract_year(record.release_date)
 
     genres =
       records
@@ -231,8 +309,7 @@ defmodule MusicLibrary.Collection do
       |> Enum.uniq()
       |> Enum.take(@max_genres_per_record)
 
-    base =
+    base = "#{artist_names} - #{record.title} (#{year}, #{formats})"
-      "#{artist_names} - #{record.title} (#{record.release_date || "Unknown"}, #{formats}, #{record.type})"
 
     if genres == [] do
       base
@@ -241,6 +318,25 @@ defmodule MusicLibrary.Collection do
     end
   end
 
+  defp extract_year(nil), do: "Unknown"
+
+  defp extract_year(date_str) when is_binary(date_str) do
+    if String.length(date_str) >= 4 do
+      String.slice(date_str, 0, 4)
+    else
+      date_str
+    end
+  end
+
+  defp extract_decade(nil), do: nil
+
+  defp extract_decade(date_str) when is_binary(date_str) do
+    case Integer.parse(String.slice(date_str, 0, 4)) do
+      {year, _} -> "#{div(year, 10) * 10}s"
+      :error -> nil
+    end
+  end
+
   defp base_search do
     from r in SearchIndex,
       where: not is_nil(r.purchased_at)

test/music_library/collection_test.exs

@@ -372,7 +372,7 @@ defmodule MusicLibrary.CollectionTest do
       assert Collection.collection_summary() == {"", 0}
     end
 
-    test "returns one line per collected record with record count" do
+    test "returns stats preamble and one catalog line per collected record" do
       record_with_artist("Radiohead", %{
         title: "OK Computer",
         format: :cd,
@@ -392,20 +392,28 @@ defmodule MusicLibrary.CollectionTest do
       })
 
       {summary, record_count} = Collection.collection_summary()
-      lines = String.split(summary, "\n")
+      [stats, catalog] = String.split(summary, "\n\n")
+      lines = String.split(catalog, "\n")
 
       assert record_count == 2
       assert length(lines) == 2
 
+      # Stats section
+      assert stats =~ "# Stats: 2 releases"
+      assert stats =~ "Genres:"
+      assert stats =~ "Formats:"
+      assert stats =~ "Eras:"
+
+      # Catalog lines use year-only dates and exclude type field
       assert Enum.any?(lines, fn line ->
                line =~ "Radiohead" and line =~ "OK Computer" and
-                 line =~ "1997-06-16" and line =~ "cd" and line =~ "album" and
+                 line =~ "(1997, cd)" and
                  line =~ "alternative rock, art rock"
              end)
 
       assert Enum.any?(lines, fn line ->
                line =~ "Pink Floyd" and line =~ "The Dark Side of the Moon" and
-                 line =~ "1973-03-01" and line =~ "vinyl" and line =~ "album" and
+                 line =~ "(1973, vinyl)" and
                  line =~ "progressive rock"
              end)
     end
@@ -434,7 +442,8 @@ defmodule MusicLibrary.CollectionTest do
       })
 
       {summary, record_count} = Collection.collection_summary()
-      lines = String.split(summary, "\n")
+      catalog = String.split(summary, "\n\n") |> List.last()
+      lines = String.split(catalog, "\n")
 
       assert record_count == 1
       assert length(lines) == 1
@@ -444,7 +453,7 @@ defmodule MusicLibrary.CollectionTest do
       assert line =~ "vinyl"
     end
 
-    test "caps genres to 3 per record" do
+    test "caps genres to 2 per record" do
       record_with_artist("ABC", %{
         title: "The Lexicon of Love",
         format: :vinyl,
@@ -455,10 +464,11 @@ defmodule MusicLibrary.CollectionTest do
       })
 
       {summary, _count} = Collection.collection_summary()
-      [genres_str] = Regex.run(~r/\[(.+)\]/, summary, capture: :all_but_first)
+      catalog = String.split(summary, "\n\n") |> List.last()
+      [genres_str] = Regex.run(~r/\[(.+)\]/, catalog, capture: :all_but_first)
       genre_count = genres_str |> String.split(", ") |> length()
 
-      assert genre_count == 3
+      assert genre_count == 2
     end
 
     test "omits genre brackets when genres are empty" do
@@ -472,8 +482,9 @@ defmodule MusicLibrary.CollectionTest do
       })
 
       {summary, _count} = Collection.collection_summary()
-      refute summary =~ "["
+      catalog = String.split(summary, "\n\n") |> List.last()
-      refute summary =~ "]"
+      refute catalog =~ "["
+      refute catalog =~ "]"
     end
 
     test "excludes wishlist records" do
@@ -489,8 +500,11 @@ defmodule MusicLibrary.CollectionTest do
 
       {summary, record_count} = Collection.collection_summary()
       assert record_count == 1
-      assert summary =~ "OK Computer"
+
-      refute summary =~ "Wish You Were Here"
+      # Catalog portion should contain the purchased record but not the wishlist one
+      catalog = String.split(summary, "\n\n") |> List.last()
+      assert catalog =~ "OK Computer"
+      refute catalog =~ "Wish You Were Here"
     end
   end
 end