local_test_with_api.sh

#!/usr/bin/env bash
# ═══════════════════════════════════════════════════════════════
# purpose-agent v3.0.0 — Local Test Runner (including cloud model)
#
# Run on your own machine (no HF credits needed):
#   git clone https://huggingface.co/Rohan03/purpose-agent pa && cd pa
#   pip install -e .
#   export OPENROUTER_API_KEY="sk-or-v1-..."
#   bash local_test_with_api.sh
# ═══════════════════════════════════════════════════════════════
set -e

PASS=0; FAIL=0; TOTAL=0

run_test() {
    local name="$1"; local cmd="$2"
    echo ""
    echo "══════════════════════════════════════════════════════════"
    echo "  Running: $name"
    echo "══════════════════════════════════════════════════════════"
    if eval "$cmd"; then
        echo "  ✅ $name PASSED"; PASS=$((PASS+1))
    else
        echo "  ❌ $name FAILED"; FAIL=$((FAIL+1))
    fi
    TOTAL=$((TOTAL+1))
}

echo "╔══════════════════════════════════════════════════════════╗"
echo "║  purpose-agent v3.0.0 — Full Test Suite (Local + Cloud)║"
echo "╚══════════════════════════════════════════════════════════╝"

# Pre-flight
echo "═══ Pre-flight ═══"
python -c "
import purpose_agent as pa
print(f'  v{pa.__version__} — {len(pa.__all__)} exports')
assert pa.__version__ == '3.0.0', f'Version: {pa.__version__}'
assert len(pa.__all__) >= 110, f'Exports: {len(pa.__all__)}'
missing = [n for n in pa.__all__ if not hasattr(pa, n)]
assert not missing, f'Missing: {missing}'
print('  ✅ All exports importable')
"

# Apply fixes
echo ""
echo "═══ Applying test fixes ═══"
python apply_fixes.py

# Layer 1: Unit Tests
run_test "test_core" "python tests/test_core.py"
run_test "test_public_api_211" "python tests/compat/test_public_api_211.py"
run_test "test_first_principles" "python tests/test_first_principles.py"
run_test "test_hardening" "python tests/test_hardening.py"
run_test "test_sre_regression" "python tests/test_sre_regression.py"

# Layer 2: Feature Tests
run_test "test_sprint1_events" "python tests/test_sprint1_events.py"
run_test "test_sprint2_checkpoint" "python tests/test_sprint2_checkpoint.py"
run_test "test_sprint3_homeostasis" "python tests/test_sprint3_homeostasis.py"
run_test "test_sprint4_8_protocols" "python tests/test_sprint4_8_protocols.py"
run_test "test_track_c" "python tests/test_track_c.py"
run_test "test_track_d" "python tests/test_track_d.py"

# Layer 3: Integration
run_test "validate_quick" "python benchmarks/validate.py --quick"
run_test "benchmark_v3" "python -m purpose_agent.benchmark_v3"

# Layer 4: Cloud Model Tests
if [ -n "$OPENROUTER_API_KEY" ]; then
    run_test "prod_test (real model)" "python tests/prod_test.py"
else
    echo "  ⚠️  OPENROUTER_API_KEY not set — skipping cloud tests"
fi

# Report
echo ""
echo "╔══════════════════════════════════════════════════════════╗"
echo "║  RESULTS: $PASS/$TOTAL passed, $FAIL failed"
if [ $FAIL -eq 0 ]; then
echo "║  ✅ ALL TESTS PASSED — READY TO PUBLISH"
echo "║  Next: python build_and_publish.py"
else
echo "║  ❌ $FAIL FAILURES — FIX BEFORE PUBLISHING"
fi
echo "╚══════════════════════════════════════════════════════════╝"
exit $FAIL