# Ignore NeurIPS papers directory
*_papers/
*_paper_hashes.json
paper_sections_index*

# Data directories (PDFs, indexes, text store)
data/
tests/test_data/

# Ignore environment files
.env
.env.*

# Python virtual environment
.venv/
venv/
ENV/

# Python cache files
__pycache__/
*.py[cod]
*$py.class

# Package build artifacts
*.egg-info/
build/
dist/

# Index files (large)
*.zip
*.index
test_index*/
paper_index_gte_temp/

# Log files
*.log

# macOS system files
.DS_Store
.DS_Store?
._*
.Spotlight-V100
.Trashes
ehthumbs.db
Thumbs.db # Exclude PDFs from git tracking
tests/test_data/citation_test_pdfs/*.pdf

# Temporary and generated files
.coverage
coverage.xml
test_confidence_display.py

# Test generation script (one-time use, requires full corpus)
generate_citation_test.py

# IDE settings
.vscode/
usage.db

# Topic-pipeline run outputs. The generator and the published artifact are
# tracked; everything a run writes alongside them is not. These regenerate on
# every run and one ledger is half a megabyte.
experiments/emerging_topics/gap_ledger*.json
experiments/emerging_topics/research_topics_*.json
experiments/emerging_topics/SUGGESTED_TOPICS*.md
experiments/emerging_topics/embed_cache.npz
experiments/emerging_topics/*.log
