Files
OSGKeyboard/.gitignore
T
Rocky 2392cc8ac7 chore(repo): ignore new generated v6 artifacts
- Add Generated/, CorpusRegistry, v6 boundary/blind evaluation
  corpora, and sealed blessing benchmark review queues to the
  generated-artifact ignore list so they stay out of the source
  tree after reruns.
2026-08-29 11:51:45 +08:00

115 lines
2.6 KiB
Plaintext

# Xcode
build/
DerivedData/
*.xcodeproj
!project.yml
*.xcworkspace
xcuserdata/
*.xcuserstate
*.moved-aside
*.hmap
*.ipa
*.dSYM.zip
*.dSYM
# Swift Package Manager
.build/
.swiftpm/
Package.resolved
# CocoaPods (unused but safe)
Pods/
*.xcworkspace
# Carthage
Carthage/Build/
# fastlane
fastlane/report.xml
fastlane/Preview.html
fastlane/screenshots/**/*.png
fastlane/test_output
# macOS
.DS_Store
.AppleDouble
.LSOverride
._*
# IDE
.idea/
.vscode/
*.swp
*.swo
# Local config (API keys, signing, etc.)
*.local
Signing.local.xcconfig
.env
.env.*
# Generated by XcodeGen
*.xcodeproj/
# Local DerivedData (when using -derivedDataPath in repo)
.derivedData/
.derivedData-*/
.spm-*/
# Lexicon build cache (downloaded SogouPopularDict TSV)
.cache/
# Custom LM compile output (Mac-only; device-side prepare uses .bin from bundle)
OSGKeyboard/Resources/CustomLanguageModel/v1/compiled/
# Reproducible Create ML candidates (selected models live in Shared resources)
ModelTraining/ClipboardSemantics/Candidates/
# Reproducible open-data corpus artifacts (regenerated from pinned sources)
ModelTraining/ClipboardSemantics/Generated/
ModelTraining/ClipboardSemantics/open-training-corpus.jsonl
ModelTraining/ClipboardSemantics/combined-training-corpus.jsonl
ModelTraining/ClipboardSemantics/v6-boundary-training-supplement.jsonl
ModelTraining/ClipboardSemantics/v6-blind-evaluation-corpus.jsonl
ModelTraining/ClipboardSemantics/Consensus/*.jsonl
ModelTraining/ClipboardSemantics/CorpusRegistry/*.jsonl
ModelTraining/ClipboardSemantics/CorpusRegistry/Labels/
ModelTraining/ClipboardSemantics/CorpusRegistry/*-report.json
ModelTraining/ClipboardSemantics/VerifierCandidates/
ModelTraining/ClipboardSemantics/IterativeResearch/**/rounds.json
ModelTraining/ClipboardSemantics/IterativeResearch/**/final-report.json
# Blind review queues can contain restricted evaluation text and reviewer IDs.
ModelTraining/ClipboardSemantics/BlessingBenchmark/*.jsonl
ModelTraining/ClipboardSemantics/BlessingBenchmark/manifest.json
ModelTraining/ClipboardSemantics/BlessingBenchmark/final-manifest.json
# Python bytecode
__pycache__/
*.pyc
# Vendored mlx-audio-swift (fetched by Scripts/ensure-mlx-audio-swift.sh)
ThirdParty/mlx-audio-swift/
# Local debug capture logs and screenshots
.flow-capture.log
.flow-oslog-capture.log
.flow-*.log
.osg_*.png
# Standalone experiment: 灵动岛录音机制验证 demo (throwaway)
AudioIntentProbe/
# Local crash-log pulls and one-off diff/working notes (not for the repo)
.refs/
.crash-pull/
all-changes-vs-main.patch
CHANGES.md
# Local scratch: DerivedData clones, stress logs, QA frames, AI notes, audits
.tmp/
.ai-skills-layout-shots/
.workbuddy-ai/
reports/