From cb7404456ffb5ed439cfbe8fd4c4839abd67bc29 Mon Sep 17 00:00:00 2001
From: Snider <snider@host.uk.com>
Date: Thu, 19 Feb 2026 15:50:45 +0000
Subject: [PATCH] feat(grammar): add SignalData for disambiguation signals

Load noun_determiner, verb_auxiliary, and verb_infinitive word lists
from gram.signal in locale JSON. Reserve Priors field for future
corpus-derived per-word disambiguation priors.

Co-Authored-By: Virgil <virgil@lethean.io>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 grammar_test.go | 33 +++++++++++++++++++++++++++++++++
 loader.go       | 32 ++++++++++++++++++++++++++++++++
 locales/en.json | 17 +++++++++++++++++
 types.go        |  9 +++++++++
 4 files changed, 91 insertions(+)

diff --git a/grammar_test.go b/grammar_test.go
index 6a62526..c6eeea3 100644
--- a/grammar_test.go
+++ b/grammar_test.go
@@ -346,6 +346,39 @@ func TestActionFailed(t *testing.T) {
 	}
 }
 
+func TestGrammarData_Signals(t *testing.T) {
+	svc, err := New()
+	if err != nil {
+		t.Fatalf("New() failed: %v", err)
+	}
+	SetDefault(svc)
+
+	data := GetGrammarData("en")
+	if data == nil {
+		t.Fatal("GetGrammarData(\"en\") returned nil")
+	}
+	if len(data.Signals.NounDeterminers) == 0 {
+		t.Error("Signals.NounDeterminers is empty")
+	}
+	if len(data.Signals.VerbAuxiliaries) == 0 {
+		t.Error("Signals.VerbAuxiliaries is empty")
+	}
+	if len(data.Signals.VerbInfinitive) == 0 {
+		t.Error("Signals.VerbInfinitive is empty")
+	}
+
+	// Spot-check known values
+	found := false
+	for _, d := range data.Signals.NounDeterminers {
+		if d == "the" {
+			found = true
+		}
+	}
+	if !found {
+		t.Error("NounDeterminers missing 'the'")
+	}
+}
+
 func TestTemplateFuncs(t *testing.T) {
 	funcs := TemplateFuncs()
 	expected := []string{"title", "lower", "upper", "past", "gerund", "plural", "pluralForm", "article", "quote"}
diff --git a/loader.go b/loader.go
index 1173fef..898afa5 100644
--- a/loader.go
+++ b/loader.go
@@ -154,6 +154,38 @@ func flattenWithGrammar(prefix string, data map[string]any, out map[string]Messa
 				}
 			}
 
+			// Signal data for disambiguation
+			if grammar != nil && fullKey == "gram.signal" {
+				if nd, ok := v["noun_determiner"]; ok {
+					if arr, ok := nd.([]any); ok {
+						for _, item := range arr {
+							if s, ok := item.(string); ok {
+								grammar.Signals.NounDeterminers = append(grammar.Signals.NounDeterminers, s)
+							}
+						}
+					}
+				}
+				if va, ok := v["verb_auxiliary"]; ok {
+					if arr, ok := va.([]any); ok {
+						for _, item := range arr {
+							if s, ok := item.(string); ok {
+								grammar.Signals.VerbAuxiliaries = append(grammar.Signals.VerbAuxiliaries, s)
+							}
+						}
+					}
+				}
+				if vi, ok := v["verb_infinitive"]; ok {
+					if arr, ok := vi.([]any); ok {
+						for _, item := range arr {
+							if s, ok := item.(string); ok {
+								grammar.Signals.VerbInfinitive = append(grammar.Signals.VerbInfinitive, s)
+							}
+						}
+					}
+				}
+				continue
+			}
+
 			// Article configuration
 			if grammar != nil && fullKey == "gram.article" {
 				if indef, ok := v["indefinite"].(map[string]any); ok {
diff --git a/locales/en.json b/locales/en.json
index 92feb45..27989b3 100644
--- a/locales/en.json
+++ b/locales/en.json
@@ -110,6 +110,23 @@
 			"label": ":",
 			"progress": "..."
 		},
+		"signal": {
+			"noun_determiner": [
+				"the", "a", "an",
+				"this", "that", "these", "those",
+				"my", "your", "his", "her", "its", "our", "their",
+				"every", "each", "some", "any", "no",
+				"many", "few", "several", "all", "both"
+			],
+			"verb_auxiliary": [
+				"is", "are", "was", "were",
+				"has", "had", "have",
+				"do", "does", "did",
+				"will", "would", "could", "should",
+				"can", "may", "might", "shall", "must"
+			],
+			"verb_infinitive": ["to"]
+		},
 		"number": {
 			"thousands": ",",
 			"decimal": ".",
diff --git a/types.go b/types.go
index 1f5ba9a..a001bb4 100644
--- a/types.go
+++ b/types.go
@@ -192,6 +192,7 @@ type GrammarData struct {
 	Articles ArticleForms         // article configuration
 	Words    map[string]string    // base word translations
 	Punct    PunctuationRules     // language-specific punctuation
+	Signals  SignalData           // disambiguation signal word lists
 }
 
 // VerbForms holds verb conjugations.
@@ -221,6 +222,14 @@ type PunctuationRules struct {
 	ProgressSuffix string // "..."
 }
 
+// SignalData holds word lists used for disambiguation signals.
+type SignalData struct {
+	NounDeterminers []string                      // Words that precede nouns: "the", "a", "this", "my", ...
+	VerbAuxiliaries []string                      // Auxiliaries/modals before verbs: "is", "was", "will", ...
+	VerbInfinitive  []string                      // Infinitive markers: "to"
+	Priors          map[string]map[string]float64 // Reserved: per-word priors {"commit": {"verb": 0.4, "noun": 0.6}}
+}
+
 // --- Number Formatting ---
 
 // NumberFormat defines locale-specific number formatting rules.