feat: add imt sync cleanup command to remove duplicates

Scans mods and tools collections for entries with the same name+author,
keeps the most recently updated entry, and deletes older duplicates.
Supports --dry-run to preview deletions.

Co-Authored-By: Claude Opus 4.5 <[email protected]>

Author: dyoung522

CHANGELOG.md

@@ -4,8 +4,11 @@ All notable changes to this project will be documented in this file.
 
 ## History (reverse chronological order)
 
-### v2.5.3 - 2026-03-04
+### v2.6.0 - 2026-03-04
 
+- Add `imt sync cleanup` command to remove duplicate entries from mods and tools collections
+  - Groups entries by name+author and keeps the most recently updated
+  - Supports `--dry-run` to preview what would be deleted
 - Fix `find_info` to match by both name AND author, not just name
   - Previously, mods/tools with the same name but different authors could cause incorrect deletion logic
   - Now consistent with `info_array` deduplication and `find_by_type` lookups

Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    Icarus-Mod-Tools (2.5.2)
+    Icarus-Mod-Tools (2.5.3)
       google-cloud-firestore (~> 2.7)
       octokit (~> 6.0)
       paint (~> 2.3)

lib/icarus/mod/cli/sync.rb

@@ -52,11 +52,63 @@ def tools
           sync_list(:tools)
         end
 
+        desc "cleanup", "Remove duplicate entries from mods and tools collections"
+        def cleanup
+          cleanup_duplicates(:mods)
+          cleanup_duplicates(:tools)
+        end
+
         no_commands do
           def firestore
             $firestore ||= Firestore.new
           end
 
+          def cleanup_duplicates(type)
+            singular_type = type.to_s.chomp("s").to_sym
+            collection = firestore.send(type)
+
+            puts "Scanning #{type} for duplicates..." if verbose?
+
+            # Group by [name, author]
+            grouped = collection.group_by { |item| [item.name, item.author] }
+            duplicates = grouped.select { |_, items| items.length > 1 }
+
+            if duplicates.empty?
+              puts "No duplicate #{type} found." if verbose?
+              return
+            end
+
+            puts "Found #{duplicates.length} duplicate #{singular_type}(s) to clean up:" if verbose?
+
+            duplicates.each do |key, items|
+              name, author = key
+              # Sort by updated_at descending, keep the most recent
+              sorted = items.sort_by { |item| item.updated_at || Time.at(0) }.reverse
+              keeper = sorted.first
+              to_delete = sorted[1..]
+
+              puts "  #{author}/#{name}: #{items.length} entries" if verbose?
+              puts "    Keeping: #{keeper.id} (updated: #{keeper.updated_at})" if verbose?
+
+              to_delete.each do |item|
+                if options[:dry_run]
+                  puts Paint["    Would delete: #{item.id} (updated: #{item.updated_at})", :yellow] if verbose?
+                else
+                  puts "    Deleting: #{item.id} (updated: #{item.updated_at})" if verbose?
+                  response = firestore.delete(singular_type, item)
+                  puts "      #{success_or_failure(response)}" if verbose > 1
+                end
+              end
+            end
+
+            deleted_count = duplicates.values.sum { |items| items.length - 1 }
+            if options[:dry_run]
+              puts Paint["Dry run; no changes made. Would have deleted #{deleted_count} duplicate #{type}.", :yellow] if verbose?
+            else
+              puts "Deleted #{deleted_count} duplicate #{type}." if verbose?
+            end
+          end
+
           def success_or_failure(status)
             format("%<status>10s", status: status ? Paint["Success", :green] : Paint["Failure", :red])
           end

lib/icarus/mod/version.rb

@@ -2,6 +2,6 @@
 
 module Icarus
   module Mod
-    VERSION = "2.5.3"
+    VERSION = "2.6.0"
   end
 end

spec/icarus/mod/cli/sync_spec.rb

@@ -233,4 +233,66 @@
       expect { sync_command.all }.to output(String).to_stdout
     end
   end
+
+  describe "#cleanup" do
+    let(:options) { { verbose: [true], dry_run: false } }
+    let(:mod1) { Icarus::Mod::Tools::Modinfo.new({ name: "DupeMod", author: "Author1", description: "Test" }, id: "mod1", updated: Time.now - 3600) }
+    let(:mod2) { Icarus::Mod::Tools::Modinfo.new({ name: "DupeMod", author: "Author1", description: "Test" }, id: "mod2", updated: Time.now) }
+    let(:mod3) { Icarus::Mod::Tools::Modinfo.new({ name: "UniqueMod", author: "Author2", description: "Test" }, id: "mod3", updated: Time.now) }
+    let(:tool1) { Icarus::Mod::Tools::Toolinfo.new({ name: "DupeTool", author: "Author1", description: "Test" }, id: "tool1", updated: Time.now - 3600) }
+    let(:tool2) { Icarus::Mod::Tools::Toolinfo.new({ name: "DupeTool", author: "Author1", description: "Test" }, id: "tool2", updated: Time.now) }
+
+    before do
+      allow(firestore_double).to receive(:mods).and_return([mod1, mod2, mod3])
+      allow(firestore_double).to receive(:tools).and_return([tool1, tool2])
+      allow(firestore_double).to receive(:delete).and_return(true)
+    end
+
+    it "identifies duplicate mods by name and author" do
+      expect { sync_command.cleanup }.to output(/Found 1 duplicate mod/).to_stdout
+    end
+
+    it "identifies duplicate tools by name and author" do
+      expect { sync_command.cleanup }.to output(/Found 1 duplicate tool/).to_stdout
+    end
+
+    it "keeps the most recently updated entry" do
+      expect { sync_command.cleanup }.to output(/Keeping.*mod2/).to_stdout
+    end
+
+    it "deletes older duplicate entries" do
+      sync_command.cleanup
+      expect(firestore_double).to have_received(:delete).with(:mod, mod1)
+      expect(firestore_double).to have_received(:delete).with(:tool, tool1)
+    end
+
+    it "does not delete unique entries" do
+      sync_command.cleanup
+      expect(firestore_double).not_to have_received(:delete).with(:mod, mod3)
+    end
+
+    context "when no duplicates exist" do
+      before do
+        allow(firestore_double).to receive(:mods).and_return([mod3])
+        allow(firestore_double).to receive(:tools).and_return([])
+      end
+
+      it "reports no duplicates found" do
+        expect { sync_command.cleanup }.to output(/No duplicate mods found/).to_stdout
+      end
+    end
+
+    context "with dry_run enabled" do
+      let(:options) { { verbose: [true], dry_run: true } }
+
+      it "does not delete any entries" do
+        expect { sync_command.cleanup }.to output(/Dry run/).to_stdout
+        expect(firestore_double).not_to have_received(:delete)
+      end
+
+      it "shows what would be deleted" do
+        expect { sync_command.cleanup }.to output(/Would delete.*mod1/).to_stdout
+      end
+    end
+  end
 end