UV ?= uv
VENV ?= .venv
PY ?= $(VENV)/bin/python
PIP ?= $(UV) pip
.PHONY: venv install install-ocr install-llm install-llm-models test test-ocr test-quick test-llm test-e2e clean
.PHONY: smoke prepush
.PHONY: lint format-check
.PHONY: check-tesseract check-llm
.PHONY: setup start-vlm
venv: $(VENV)/bin/python
$(VENV)/bin/python:
$(UV) venv $(VENV)
install: venv
. $(VENV)/bin/activate && $(PIP) install -r requirements.txt
install-ocr: install
. $(VENV)/bin/activate && $(PIP) install -e ".[ocr]"
@echo "OCR dependencies installed. Ensure Tesseract is installed:"
@echo " macOS: brew install tesseract"
@echo " Linux: sudo apt-get install tesseract-ocr"
test: install
. $(VENV)/bin/activate && PYTHONWARNINGS=default pytest
test-ocr: install-ocr
. $(VENV)/bin/activate && PYTHONWARNINGS=default pytest tests/test_ocr.py tests/test_phase2_features.py -v
test-quick: install
. $(VENV)/bin/activate && PYTHONWARNINGS=default pytest -q --tb=short
check-tesseract:
@which tesseract > /dev/null 2>&1 && tesseract --version | head -1 || echo "Tesseract not installed. Run: brew install tesseract (macOS) or apt-get install tesseract-ocr (Linux)"
@$(PY) -c "import pytesseract; print(f'pytesseract: OK, languages: {pytesseract.get_languages()}')" 2>/dev/null || echo "pytesseract not installed. Run: make install-ocr"
lint: install
@FILES=$$(git diff --name-only --diff-filter=ACMRTUXB origin/main...HEAD 2>/dev/null | grep -E '\\.py$$' || true); \
if [ -z "$$FILES" ]; then echo "ruff: no changed python files"; exit 0; fi; \
. $(VENV)/bin/activate && ruff check $$FILES
format-check: install
@FILES=$$(git diff --name-only --diff-filter=ACMRTUXB origin/main...HEAD 2>/dev/null | grep -E '\\.py$$' || true); \
if [ -z "$$FILES" ]; then echo "ruff-format: no changed python files"; exit 0; fi; \
. $(VENV)/bin/activate && ruff format --check $$FILES
smoke: install
. $(VENV)/bin/activate && $(PY) scripts/cursor_smoke.py --out-dir /tmp/pdf-handler-prepush-smoke
prepush: lint format-check test smoke
# LLM/Agentic AI targets (v0.9.0+)
install-llm: install
. $(VENV)/bin/activate && $(PIP) install -e ".[llm]"
@echo "LLM dependencies installed (openai)."
@echo "For local VLM (free), start the model server:"
@echo " ./scripts/run_local_vlm.sh"
@echo "For Ollama (free), install: curl -fsSL https://ollama.ai/install.sh | sh"
@echo "For Ollama model setup: make install-llm-models"
install-llm-models: install
. $(VENV)/bin/activate && $(PY) scripts/ensure_ollama_model.py
test-llm: install
. $(VENV)/bin/activate && PYTHONWARNINGS=default pytest tests/test_agentic_features.py -v
test-e2e: install
. $(VENV)/bin/activate && PYTHONWARNINGS=default pytest tests/test_agentic_features.py -v -m slow
@echo "Note: E2E tests require local model server running at localhost:8100"
check-llm:
@echo ""
@echo "==============================================================="
@echo " LLM Backend Status Check"
@echo "==============================================================="
@echo ""
@$(PY) scripts/check_llm_status.py 2>/dev/null || echo "pdf_mcp not installed. Run: make install"
# Full environment setup (one-time, cross-platform)
setup:
@chmod +x scripts/setup_environment.sh
@./scripts/setup_environment.sh
# Start local VLM server (auto-detects best GPU)
start-vlm:
@chmod +x scripts/run_local_vlm.sh
@./scripts/run_local_vlm.sh
clean:
rm -rf $(VENV) .pytest_cache **/__pycache__
