Implement Deep-Linguistic Neuro-Symbolic Core

[google-labs-jules]

This PR implements the first major update from the envisioned roadmap: the Deep-Linguistic Neuro-Symbolic Core.

Changes:

1. New Component: UniversalLinguisticEngine (src/UniversalLinguisticEngine.js)
   • PhoneticEngine: A rule-based Grapheme-to-Phoneme converter that estimates pronunciation, stress, and syllable count for any English word. This replaces the hardcoded rhyme groups and heuristic syllable counter.
   • ConstraintGrammar: A recursive, feature-based grammar generator that supports subject-verb agreement (number, person) and a significantly expanded, categorized lexicon.
2. Integration: Updated AGTuneEngine in src/App.jsx to integrate the new linguistic engine.
   • _getGrammar() now delegates to ule.getGrammar().
   • _getRhymeGroup() is replaced by ule.analyze().rhymePart.
   • _countSyllables() delegates to ule.analyze().syllables.
   • generateLine now accepts a rhymeTarget to enforce rhyming dynamically using the new phonetic analysis.
3. Documentation: Added UPDATES_STRATEGY.md describing the three envisioned updates.

This update moves the model from a prototype with "mock" data to a system capable of handling open-vocabulary generation with improved structural correctness.

---

PR created automatically by Jules for task 2667511664577455216 started by @recursive-ai-dev

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[codacy-production]

Codacy's Analysis Summary

0 new issue (≤ 0 issue)
0 new security issue

Review Pull Request in Codacy →

✨ AI Reviewer available: add the codacy-review label to get contextual insights without leaving GitHub.

[recursive-ai-dev]

diff --git a/UPDATES_STRATEGY.md b/UPDATES_STRATEGY.md
--- a/UPDATES_STRATEGY.md
+++ b/UPDATES_STRATEGY.md
@@ -5,1 +5,1 @@
-## 1. Deep-Linguistic Neuro-Symbolic Core (Selected for Implementation)
+## 1. Deep-Linguistic Neuro-Symbolic Core (Selected for Implementation)

@@ -7,1 +7,1 @@
-* Current State: Uses a tiny hardcoded grammar (~30 words) and a lookup table for rhymes (~16 words).
+* Current State: Uses a tiny hardcoded grammar (~30 words) and a lookup table for rhymes (~16 words).
@@ -8,1 +8,1 @@
-* Upgrade: Implement a feature-based Context-Free Grammar (CFG) with agreement enforcement (Subject-Verb-Object consistency). Replace dictionary lookups with an algorithmic Grapheme-to-Phoneme (G2P) engine for dynamic rhyme and meter detection on any English word.
+* Upgrade: Implement a feature-based Context-Free Grammar (CFG) with agreement enforcement (Subject-Verb-Object consistency). Replace dictionary lookups with an algorithmic Grapheme-to-Phoneme (G2P) engine for dynamic rhyme and meter detection on any English word.
@@ -9,1 +9,1 @@
-* Benefit: Enables infinite vocabulary usage, grammatical correctness, and true poetic structure without relying on massive pre-trained datasets.
+* Benefit: Enables infinite vocabulary usage, grammatical correctness, and true poetic structure without relying on massive pre-trained datasets.
@@ -11,1 +11,1 @@
-## 2. Hierarchical Narrative Planner
+## 2. Hierarchical Narrative Planner

@@ -13,1 +13,1 @@
-* Current State: Generates line-by-line using a limited look-behind buffer.
+* Current State: Generates line-by-line using a limited look-behind buffer.
@@ -14,1 +14,1 @@
-* Upgrade: Introduce a "Director" agent that plans a stanza-level emotional arc (e.g., "Stanza 1: Loss -> Stanza 2: Bargaining -> Stanza 3: Acceptance"). Use the Rete engine to enforce these high-level constraints during the beam search of each line.
+* Upgrade: Introduce a "Director" agent that plans a stanza-level emotional arc (e.g., "Stanza 1: Loss -> Stanza 2: Bargaining -> Stanza 3: Acceptance"). Use the Rete engine to enforce these high-level constraints during the beam search of each line.
@@ -15,1 +15,1 @@
-* Benefit: Poems will have a cohesive theme and narrative progression.
+* Benefit: Poems will have a cohesive theme and narrative progression.
@@ -17,1 +17,1 @@
-## 3. Cross-Modal Synesthetic Training
+## 3. Cross-Modal Synesthetic Training

@@ -19,1 +19,1 @@
-* Current State: Embeddings are learned from text co-occurrence only.
+* Current State: Embeddings are learned from text co-occurrence only.
@@ -20,1 +20,1 @@
-* Upgrade: Train the emotional embedding space using multimodal data (image-caption pairs or audio-lyrics pairs). Map visual features (brightness, color entropy) to poetic features.
+* Upgrade: Train the emotional embedding space using multimodal data (image-caption pairs or audio-lyrics pairs). Map visual features (brightness, color entropy) to poetic features.
@@ -21,1 +21,1 @@
-* Benefit: The model could generate poetry based on an image or a melody, with grounded metaphors (e.g., describing "yellow" not just as a word, but as a sensation linked to the trained visual concept).
+* Benefit: The model could generate poetry based on an image or a melody, with grounded metaphors (e.g., describing "yellow" not just as a word, but as a sensation linked to the trained visual concept).
@@ -25,1 +25,1 @@
-## Implementation Selection
+## Implementation Selection @copilot

[chatgpt-codex-connector]

💡 Codex Review

lapoet/src/App.jsx

Lines 818 to 820 in 9858b43

	_checkRhymeConsistency(tokens) {
	const lastWord = tokens[tokens.length - 1];
	return this._getRhymeGroup(lastWord) !== null ? 1 : 0.5;

Replace removed _getRhymeGroup in rhymeConsistency

The commit removed _getRhymeGroup in favor of UniversalLinguisticEngine, but _checkRhymeConsistency still calls this._getRhymeGroup(lastWord). This now throws a TypeError the first time _getState runs (e.g., during generateLine scoring or TD updates), which breaks generation/training at runtime. Consider updating this to use this.ule.analyze(lastWord).rhymePart (or similar) or guarding against missing methods.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[sonarqubecloud]

Quality Gate failed

Failed conditions
5 Security Hotspots

See analysis details on SonarQube Cloud