Add Quantum-Bio taxonomy V2: tests/test_taxonomy.py

tests/test_taxonomy.py

@@ -0,0 +1,252 @@
+"""
+PhD Research OS — Taxonomy Tests
+==================================
+Tests for the Quantum-Bio V2 taxonomy, domain management, and confidence scoring.
+"""
+
+import os
+import sys
+import json
+import pytest
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from phd_research_os.taxonomy import (
+    TaxonomyManager, STUDY_TYPE_WEIGHTS, ALLOWED_STUDY_TYPES,
+    TAXONOMY_VERSION, PIPELINE_VERSION, LEGACY_TO_V2_MAP
+)
+from phd_research_os.db import init_db, get_db, create_claim, create_source, to_fixed, from_fixed
+
+TEST_DB = "test_taxonomy.db"
+
+
+@pytest.fixture(autouse=True)
+def setup_teardown():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    # Seed test data
+    conn = get_db(TEST_DB)
+    create_claim(conn, "Test claim", "Fact", 0.85,
+                 study_type="primary_experimental", evidence_strength=0.9)
+    create_source(conn, "10.1234/test", "Test Paper", study_type="Simulation")
+    conn.close()
+    yield
+    for suffix in ["", "-wal", "-shm"]:
+        p = TEST_DB + suffix
+        if os.path.exists(p):
+            os.remove(p)
+
+
+# ============================================================
+# Base Taxonomy Tests
+# ============================================================
+
+def test_8_study_types_defined():
+    assert len(ALLOWED_STUDY_TYPES) == 8
+
+def test_weights_are_fixed_point():
+    for st, weight in STUDY_TYPE_WEIGHTS.items():
+        assert isinstance(weight, int), f"{st} weight should be int (fixed-point)"
+        assert 0 <= weight <= 1000, f"{st} weight {weight} out of range"
+
+def test_in_vivo_highest():
+    assert STUDY_TYPE_WEIGHTS["in_vivo"] == 1000
+
+def test_perspective_lowest():
+    assert STUDY_TYPE_WEIGHTS["perspective"] == 200
+
+def test_simulation_split():
+    """First-principles > phenomenological."""
+    assert STUDY_TYPE_WEIGHTS["first_principles_simulation"] > STUDY_TYPE_WEIGHTS["phenomenological_simulation"]
+
+
+# ============================================================
+# Normalization Tests
+# ============================================================
+
+def test_normalize_legacy_types():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    assert tm.normalize_study_type("PrimaryExperimental") == "direct_physical_measurement"
+    assert tm.normalize_study_type("InVitro") == "in_vitro"
+    assert tm.normalize_study_type("Simulation") == "phenomenological_simulation"
+    assert tm.normalize_study_type("Review") == "review"
+
+def test_normalize_aliases():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    assert tm.normalize_study_type("clinical_trial") == "in_vivo"
+    assert tm.normalize_study_type("meta-analysis") == "review"
+    assert tm.normalize_study_type("case_study") == "perspective"
+
+def test_normalize_v2_identity():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    for st in ALLOWED_STUDY_TYPES:
+        assert tm.normalize_study_type(st) == st
+
+def test_normalize_case_insensitive():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    assert tm.normalize_study_type("IN_VIVO") == "in_vivo"
+    assert tm.normalize_study_type("Mathematical_Proof") == "mathematical_proof"
+
+
+# ============================================================
+# Confidence Scoring Tests
+# ============================================================
+
+def test_confidence_max():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    result = tm.score_confidence(1.0, "in_vivo", 1, True)
+    assert result["confidence"] == 1.0
+
+def test_confidence_formula():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    result = tm.score_confidence(0.8, "in_vitro", 2, True)
+    # 0.8 × 0.85 × 0.85 × 1.0 = 0.578
+    assert 0.57 <= result["confidence"] <= 0.58
+
+def test_confidence_incomplete_penalty():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    complete = tm.score_confidence(0.9, "in_vivo", 1, True)
+    incomplete = tm.score_confidence(0.9, "in_vivo", 1, False)
+    assert incomplete["confidence"] < complete["confidence"]
+    assert incomplete["completeness_penalty"] == 0.7
+
+def test_confidence_taxonomy_version_tag():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    result = tm.score_confidence(0.5, "review", 3, True)
+    assert result["taxonomy_version"] == TAXONOMY_VERSION
+
+def test_confidence_perspective_low():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    result = tm.score_confidence(1.0, "perspective", 1, True)
+    assert result["confidence"] == 0.2  # 1.0 × 0.2 × 1.0 × 1.0
+
+
+# ============================================================
+# Domain Management Tests
+# ============================================================
+
+def test_default_domains_seeded():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    domains = tm.list_domains()
+    domain_ids = [d["domain_id"] for d in domains]
+    assert "quantum_bio" in domain_ids
+    assert "biosensors" in domain_ids
+    assert "materials_science" in domain_ids
+
+def test_create_custom_domain():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.create_domain("my_field", "My Research Field", "Custom taxonomy")
+    domain = tm.get_domain("my_field")
+    assert domain is not None
+    assert domain["name"] == "My Research Field"
+
+def test_add_custom_study_type():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.create_domain("test_domain", "Test", "Test domain")
+    tm.add_study_type("test_domain", "custom_assay", 0.75, "A custom assay type")
+    
+    domain = tm.get_domain("test_domain")
+    assert "custom_assay" in domain["custom_study_types"]
+    assert domain["custom_study_types"]["custom_assay"]["weight"] == 750
+
+def test_custom_type_affects_scoring():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.create_domain("test_scoring", "Test", "Test")
+    tm.add_study_type("test_scoring", "ultra_precise", 0.99, "Ultra-precise measurement")
+    
+    result = tm.score_confidence(1.0, "ultra_precise", 1, True, domain_id="test_scoring")
+    assert 0.98 <= result["confidence"] <= 1.0
+
+def test_remove_study_type():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.create_domain("rm_test", "Test", "Test")
+    tm.add_study_type("rm_test", "temp_type", 0.5, "Temporary")
+    assert tm.remove_study_type("rm_test", "temp_type")
+    
+    domain = tm.get_domain("rm_test")
+    assert "temp_type" not in domain["custom_study_types"]
+
+def test_cannot_delete_base_taxonomy():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    assert not tm.delete_domain("quantum_bio")
+
+def test_soft_delete_domain():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.create_domain("deletable", "Deletable", "Will be deleted")
+    assert tm.delete_domain("deletable")
+    
+    active = tm.list_domains(active_only=True)
+    assert not any(d["domain_id"] == "deletable" for d in active)
+    
+    all_domains = tm.list_domains(active_only=False)
+    assert any(d["domain_id"] == "deletable" for d in all_domains)
+
+
+# ============================================================
+# Migration Tests
+# ============================================================
+
+def test_migration_idempotent():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    result1 = tm.migrate_to_v2()
+    result2 = tm.migrate_to_v2()
+    assert result2["already_migrated"]
+
+def test_migration_normalizes_types():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    result = tm.migrate_to_v2()
+    
+    conn = get_db(TEST_DB)
+    claims = conn.execute("SELECT study_type FROM claims WHERE study_type IS NOT NULL").fetchall()
+    conn.close()
+    
+    for claim in claims:
+        st = dict(claim).get("study_type", "")
+        if st:
+            assert st in ALLOWED_STUDY_TYPES or st in ["primary_experimental"], f"Unexpected type: {st}"
+
+def test_rollback():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.migrate_to_v2()
+    result = tm.rollback_to_v1()
+    assert result["rows_reverted"] >= 0
+    assert not result["errors"]
+
+
+# ============================================================
+# Audit Log Tests
+# ============================================================
+
+def test_audit_log_records_actions():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    tm.create_domain("audit_test", "Audit Test", "Testing audit")
+    
+    log = tm.get_audit_log()
+    assert len(log) >= 1
+    actions = [e["action"] for e in log]
+    assert "create_domain" in actions
+
+
+# ============================================================
+# Cache Tests
+# ============================================================
+
+def test_cache_key_versioned():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    key1 = tm.generate_cache_key("abc123")
+    key2 = tm.generate_cache_key("abc123")
+    assert key1 == key2  # Deterministic
+
+    key3 = tm.generate_cache_key("different_hash")
+    assert key1 != key3
+
+def test_cache_validation():
+    tm = TaxonomyManager(db_path=TEST_DB)
+    valid_entry = {"taxonomy_version": TAXONOMY_VERSION, "pipeline_version": PIPELINE_VERSION}
+    assert tm.validate_cache_entry(valid_entry)
+    
+    stale_entry = {"taxonomy_version": "old_v1", "pipeline_version": "1.0.0"}
+    assert not tm.validate_cache_entry(stale_entry)
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])