feat(expo): refresh app icon and adaptive icon config

[TriForMine]

Summary by CodeRabbit

• New Features

  • Scan panel now captures camera images before scanning
  • Scanner now accepts image-based input for analysis

• Bug Fixes

  • Improved error handling and validation in encode and scan operations
  • Added fallback messages for encoding failures

• Chores

  • Updated Android adaptive icon configuration
  • Enhanced native library fetching for improved compatibility

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: e99b744e-2200-44f7-8d4e-e00c08e5cff7

📥 Commits

Reviewing files that changed from the base of the PR and between 9f3c491 and 94bb3f3.

⛔ Files ignored due to path filters (2)

• apps/expo-glyphnet/assets/images/adaptive-icon.png is excluded by !**/*.png
• apps/expo-glyphnet/assets/images/icon.png is excluded by !**/*.png

📒 Files selected for processing (6)

• apps/expo-glyphnet/app.json
• apps/expo-glyphnet/scripts/eas-fetch-jni-libs.sh
• apps/expo-glyphnet/src/adapters/scanner/types.ts
• apps/expo-glyphnet/src/features/encode/EncodePanel.tsx
• apps/expo-glyphnet/src/features/scan/ScanPanel.tsx
• crates/glyphnet-jni/src/lib.rs

---

📝 Walkthrough

Walkthrough

This PR adds camera image capture for scanning with improved error handling. It extends the ScanRequest contract, implements camera-based photo capture in ScanPanel, refactors the JNI handler to support both base64 images and legacy RGBA data, improves encode error detection, hardens the JNI build artifact extraction with ABI detection, and consolidates Android icon configuration.

Changes

Image Capture Scanning Feature

Layer / File(s)	Summary
Scanner request type extension apps/expo-glyphnet/src/adapters/scanner/types.ts	ScanRequest extends with optional imageBase64, width, height, and rgbaBase64 to accommodate either direct image data or structured RGBA input.
Camera image capture and ScanPanel integration apps/expo-glyphnet/src/features/scan/ScanPanel.tsx	Camera ref is added to ScanPanel, image is captured with takePictureAsync in base64 format at 0.9 quality, validated, and passed to the scanning adapter as imageBase64.
JNI handler dual-mode image input crates/glyphnet-jni/src/lib.rs	scanStillNative now accepts either imageBase64 (decoded and loaded directly) or falls back to width/height/rgbaBase64, with distinct error handling for each path.

Encode and Scan Error Handling

Layer / File(s)	Summary
EncodePanel error detection and rendering apps/expo-glyphnet/src/features/encode/EncodePanel.tsx	runEncode distinguishes SVG responses from errors by checking for <svg prefix, attempts JSON error extraction on non-SVG responses, and renders error messages or "No output yet" instead of assuming all responses are valid SVGs.

JNI Build Artifact Extraction

Layer / File(s)	Summary
JSON parser tool selection and GitHub API parsing apps/expo-glyphnet/scripts/eas-fetch-jni-libs.sh	Script detects available JSON parsers (node or python3) and uses the selected tool to parse workflow ID, run ID, and artifact ID from GitHub Actions API responses instead of hard-coding Python.
ABI-aware artifact extraction and install apps/expo-glyphnet/scripts/eas-fetch-jni-libs.sh	Extraction replaces wholesale directory copy with ABI-specific detection (arm64-v8a, armeabi-v7a, x86_64), validates presence, and copies only matching ABI folders into the JNI destination.

Android Icon Configuration

Layer / File(s)	Summary
Android adaptiveIcon configuration update apps/expo-glyphnet/app.json	adaptiveIcon.foregroundImage now references a consolidated ./assets/images/adaptive-icon.png file; redundant icon image entries are removed.

🎯 3 (Moderate) | ⏱️ ~20 minutes

A camera wakes and frames the glyph,
Two paths now speak to JNI's gift—
RGBA waits, or images fly,
Errors are caught before they hide.
Build scripts choose their tools with care. 🐰✨

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/expo-icon-refresh

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Copilot]

Copilot encountered an error and was unable to review this pull request. You can try again by re-requesting a review.

[github-actions]

Burst reliability (loss sweep)

✅ PR gate status: PASS

• Gating rows: 4 pass / 0 fail
• Non-gating rows: none

Drop rate	Gating	PR success	Base success	Delta	PR median frames	Base median frames
10%	Yes	100.0%	100.0%	0.0pp	14	14
20%	Yes	100.0%	100.0%	0.0pp	16	16
30%	Yes	100.0%	100.0%	0.0pp	19	19
40%	Yes	50.0%	50.0%	0.0pp	18	18

[github-actions]

Scanner perf (ribbon + matrix fixtures)

✅ PR gate status: PASS

• Ribbon budget: 25.000 ms (allowed 75.000 ms)
• Matrix budget: 20.000 ms (allowed 60.000 ms)
• Tolerance: 200.0%
• Ribbon gating benches: 3 pass / 0 fail
• Matrix gating benches: 1 pass / 0 fail

Benchmark	Gating	PR median (ms)	Base median (ms)	Delta (ms)	Delta %
scan_generated_ribbon_canvas_small (RibbonPrint)	Yes	44.848	44.730	0.118	0.26%
scan_generated_ribbon_canvas_medium (RibbonPrint)	Yes	6.672	6.658	0.013	0.20%
scan_generated_ribbon_canvas_large (RibbonPrint)	Yes	23.694	23.729	-0.035	-0.15%
scan_generated_matrix_roi (MatrixCompat)	Yes	2.919	2.908	0.011	0.36%
scan_generated_matrix_canvas (MatrixCompat)	No	174.690	174.671	0.019	0.01%
scan_real_debugger_screenshot (RibbonPrint)	No	263.235	263.127	0.108	0.04%

Matrix subset

Benchmark	Gating	PR median (ms)	Base median (ms)	Delta (ms)	Delta %
scan_generated_matrix_roi (MatrixCompat)	Yes	2.919	2.908	0.011	0.36%
scan_generated_matrix_canvas (MatrixCompat)	No	174.690	174.671	0.019	0.01%