moult 0.3.0 → 0.4.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 5da8cf8ae9bf6dae1b9a05494fdf1280d1b3c62742cd8700e24a60ca9fc66eeb
-  data.tar.gz: f487eadae0717dc3f221612234699ea175c005a0b3c1801af57f71b34aa9cfc1
+  metadata.gz: 5ad87ca8d76d09c7808a172aa24b42bf8c2031c4c5e05604f1c98e2a1120aacc
+  data.tar.gz: 683a307fa930966927516e37f6b1445b7b0e08c31da996d2c250d8ace63b2e67
 SHA512:
-  metadata.gz: 136f27d5540be7e6fb4d2174b198703ec44dbab7c3bff8735f8f6cfb40217af2df230fd48e70965a4680740c3fdf8f2325c9c3c31b589b88a56939d2bfcf5b06
-  data.tar.gz: 677b2b5fdbb133f49046baa84e4a4a286e587c9da8904e44397a01d6ef116f5d7e4480bba88b73cfbaae45a37318e36b02d1981dc8ec6d6c44be912bcea20329
+  metadata.gz: 04d0d5fb568b8135a854d36a89e4b1134a8666007e033465f7e5d56c7566e80dce839a383b333c4dad72cb02054b4b5e87b88954003f8a8d29555e545d5d276f
+  data.tar.gz: 82836397a09f0314f00ae108c339c2a119beb9c0a648323805bbd1a1d9c87b2bbf8191c94e6815837fa1627eb535d9e7c2511df21e8e12ea78ca570ec564dd61

data/CHANGELOG.md

@@ -11,6 +11,19 @@ follow [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 ## [Unreleased]
 
+### Added
+- `clone_group` on duplication findings and gate contributions — a
+  `"<kind>:<structural-hash>"` join key (kind `identical`|`similar`) shared by
+  every occurrence of a clone group, so consumers can link exact twins. Null on
+  non-duplication contributions. Additive/optional in both schemas, so
+  `schema_version` stays 1.
+
+### Changed
+- The gate now emits one duplication contribution per in-diff occurrence
+  (previously one per clone group, attributed to its first in-diff
+  occurrence), so every site of a clone is visible downstream. Verdicts are
+  unchanged, and rule reasons still count clone groups.
+
 ## [0.3.0] - 2026-07-02
 
 (v0.2.0 was tagged in git but never published as a gem — its changes are

data/lib/moult/duplication.rb

@@ -63,6 +63,7 @@ module Moult
         kind: set.kind,
         node_type: set.node_type,
         mass: set.mass,
+        clone_group: "#{set.kind}:#{set.structural_hash}",
         reasons: assessment.reasons,
         occurrences: occurrences
       )

data/lib/moult/duplication_report.rb

@@ -21,8 +21,10 @@ module Moult
     end
 
     # A confidence-graded clone group. Carries its reasons so no claim is made
-    # without a recorded justification.
-    Finding = Struct.new(:confidence, :kind, :node_type, :mass, :reasons, :occurrences) do
+    # without a recorded justification. +clone_group+ ("<kind>:<structural-hash>")
+    # is the group's join key, shared by every occurrence; stable within a report
+    # only (the hash comes from the detector backend).
+    Finding = Struct.new(:confidence, :kind, :node_type, :mass, :clone_group, :reasons, :occurrences) do
       def to_h
         {
           category: Duplication::Confidence::CATEGORY,
@@ -30,6 +32,7 @@ module Moult
           kind: kind.to_s,
           node_type: node_type,
           mass: mass,
+          clone_group: clone_group,
           reasons: reasons.map(&:to_h),
           occurrences: occurrences.map(&:to_h)
         }

data/lib/moult/gate/evaluation.rb

@@ -19,7 +19,7 @@ module Moult
       DeadCodeObs = Struct.new(:symbol_id, :path, :line, :confidence)
       BoundaryObs = Struct.new(:symbol_id, :path, :line, :severity, :violation_type)
       ComplexityObs = Struct.new(:symbol_id, :path, :line, :abc)
-      DuplicationObs = Struct.new(:symbol_id, :path, :line, :mass)
+      DuplicationObs = Struct.new(:symbol_id, :path, :line, :mass, :clone_group)
 
       # The full input. Each analysis's list is nil when that analysis was skipped
       # (errored, or not applicable — e.g. a non-packwerk repo for boundaries);
@@ -39,10 +39,11 @@ module Moult
 
       # One contributing finding behind a rule outcome. Stays confidence-graded:
       # +value+ is the observed signal (confidence/abc/mass/severity), never a
-      # claim of certainty.
-      Contribution = Struct.new(:category, :path, :symbol_id, :line, :value) do
+      # claim of certainty. +clone_group+ links occurrences of the same clone
+      # group ("<kind>:<structural-hash>"); null outside duplication.
+      Contribution = Struct.new(:category, :path, :symbol_id, :line, :value, :clone_group) do
         def to_h
-          {category: category, path: path, symbol_id: symbol_id, line: line, value: value}
+          {category: category, path: path, symbol_id: symbol_id, line: line, value: value, clone_group: clone_group}
         end
       end
 
@@ -123,8 +124,10 @@ module Moult
         t = policy.duplication_max_mass
         outcome("new_code_duplication_ceiling", "<= #{t}", obs, diagnostic, "structural_duplication") do |list|
           violating = list.select { |o| o.mass > t }
+          # Observations are per occurrence, but the reason still counts GROUPS.
           detail = phrase(violating, "no clone group touching the diff exceeds mass #{t}",
-            "clone group(s) touching the diff exceed mass #{t}")
+            "clone group(s) touching the diff exceed mass #{t}",
+            count: violating.map(&:clone_group).uniq.size)
           [violating, list.map(&:mass).max, detail, ->(o) { o.mass }]
         end
       end
@@ -142,7 +145,7 @@ module Moult
           rule: rule, evaluated: true, observed: observed, threshold: threshold,
           passed: violating.empty?,
           reasons: [Reason.new(rule: rule.to_sym, detail: detail)],
-          findings: violating.map { |o| Contribution.new(category: category, path: o.path, symbol_id: o.symbol_id, line: o.line, value: value.call(o)) }
+          findings: violating.map { |o| Contribution.new(category: category, path: o.path, symbol_id: o.symbol_id, line: o.line, value: value.call(o), clone_group: (o.clone_group if o.respond_to?(:clone_group))) }
         )
       end
 
@@ -155,9 +158,11 @@ module Moult
         )
       end
 
-      # "no X ..." when nothing violates, "<n> X ..." otherwise.
-      def phrase(violating, none, some)
-        violating.empty? ? none : "#{violating.size} #{some}"
+      # "no X ..." when nothing violates, "<n> X ..." otherwise. +count+ overrides
+      # the rendered n where one violation spans several observations (duplication
+      # counts groups, not occurrences).
+      def phrase(violating, none, some, count: violating.size)
+        violating.empty? ? none : "#{count} #{some}"
       end
 
       def verdict_reasons(verdict, failed)

data/lib/moult/gate.rb

@@ -127,18 +127,21 @@ module Moult
       end
     end
 
-    # One observation per clone GROUP touching the diff (mass is a group property);
-    # attributed to its first in-diff occurrence.
+    # One observation per in-diff OCCURRENCE, so every site of a clone is visible
+    # downstream. Mass stays a group property shared by all of them, and the
+    # occurrences of one group share its clone_group join key.
     def scope_duplication(run, diff)
       return nil unless run.ok?
 
-      run.value.findings.filter_map do |finding|
-        occ = finding.occurrences.find { |o| diff.in_diff?(path: o.path, start_line: o.line, end_line: o.line) }
-        next unless occ
+      run.value.findings.flat_map do |finding|
+        finding.occurrences.filter_map do |occ|
+          next unless diff.in_diff?(path: occ.path, start_line: occ.line, end_line: occ.line)
 
-        Evaluation::DuplicationObs.new(
-          symbol_id: occ.symbol_id, path: occ.path, line: occ.line, mass: finding.mass
-        )
+          Evaluation::DuplicationObs.new(
+            symbol_id: occ.symbol_id, path: occ.path, line: occ.line,
+            mass: finding.mass, clone_group: finding.clone_group
+          )
+        end
       end
     end
 

data/lib/moult/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Moult
-  VERSION = "0.3.0"
+  VERSION = "0.4.1"
 end

data/schema/duplication.schema.json

@@ -111,6 +111,10 @@
           "type": "integer",
           "minimum": 0
         },
+        "clone_group": {
+          "description": "Clone-group join key, \"<kind>:<structural-hash>\" where kind is identical|similar. Every occurrence of this finding shares it. Stable within a report only (the hash comes from the detector backend).",
+          "type": ["string", "null"]
+        },
         "reasons": {
           "description": "The factors that raised or lowered the confidence, in application order.",
           "type": "array",

data/schema/gate.schema.json

@@ -130,6 +130,10 @@
         "value": {
           "description": "The observed signal that tripped the rule (a confidence, an ABC score, a clone mass, or a severity). A graded/classified value, never a certainty.",
           "type": ["number", "string", "null"]
+        },
+        "clone_group": {
+          "description": "Clone-group join key, \"<kind>:<structural-hash>\" where kind is identical|similar. Contributions sharing a value are occurrences of the same clone group. Stable within a report only. Null for non-duplication findings.",
+          "type": ["string", "null"]
         }
       }
     },

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: moult
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.4.1
 platform: ruby
 authors:
-- The Moult authors
+- Brandyn Britton
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-07-02 00:00:00.000000000 Z
+date: 2026-07-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: prism
@@ -58,7 +58,7 @@ description: Moult sheds dead code. `moult hotspots` ranks files by a complexity
   a rubydex definition graph, with Rails entrypoint awareness. Every finding carries
   a confidence and its reasons, never a claim of certain death.
 email:
-- contact@moult.dev
+- brandynbb96@gmail.com
 executables:
 - moult
 extensions: []