Spaces:
Running
Running
File size: 11,962 Bytes
80d8c84 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 | """Tests for MOD 05 — deterministic semantic validation."""
from __future__ import annotations
from replicalab.models import Protocol
from replicalab.scenarios.templates import generate_scenario
from replicalab.utils.validation import (
IssueSeverity,
ValidationResult,
validate_protocol,
)
def _easy_math_scenario():
return generate_scenario(seed=42, template="math_reasoning", difficulty="easy")
def _hard_finance_scenario():
return generate_scenario(seed=42, template="finance_trading", difficulty="hard")
def _protocol_for_scenario(scenario, **overrides) -> Protocol:
"""Build a plausible protocol from the scenario's lab inventory."""
lab = scenario.lab_manager_observation
defaults = {
"sample_size": 10,
"controls": ["baseline"],
"technique": "grid_search",
"duration_days": lab.time_limit_days,
"required_equipment": list(lab.equipment_available[:1]) if lab.equipment_available else ["compute"],
"required_reagents": list(lab.reagents_in_stock[:1]) if lab.reagents_in_stock else ["data"],
"rationale": "Uses available resources within time and budget.",
}
defaults.update(overrides)
return Protocol(**defaults)
# ---------------------------------------------------------------------------
# Basic valid / invalid
# ---------------------------------------------------------------------------
def test_valid_protocol_passes() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(scenario)
result = validate_protocol(protocol, scenario)
assert result.valid is True
assert len(result.errors) == 0
def test_zero_sample_size_is_error() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(scenario, sample_size=0)
result = validate_protocol(protocol, scenario)
assert result.valid is False
assert any(i.category == "sample_size" for i in result.errors)
def test_zero_duration_is_error() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(scenario, duration_days=0)
result = validate_protocol(protocol, scenario)
assert result.valid is False
assert any(
i.category == "duration" and i.severity is IssueSeverity.ERROR
for i in result.issues
)
# ---------------------------------------------------------------------------
# Duration vs time limit
# ---------------------------------------------------------------------------
def test_duration_exceeding_time_limit_is_error() -> None:
scenario = _easy_math_scenario()
time_limit = scenario.lab_manager_observation.time_limit_days
protocol = _protocol_for_scenario(scenario, duration_days=time_limit + 5)
result = validate_protocol(protocol, scenario)
assert result.valid is False
duration_errors = [
i for i in result.errors
if i.category == "duration" and "exceeds" in i.message
]
assert len(duration_errors) == 1
def test_duration_within_limit_passes() -> None:
scenario = _easy_math_scenario()
time_limit = scenario.lab_manager_observation.time_limit_days
protocol = _protocol_for_scenario(scenario, duration_days=time_limit)
result = validate_protocol(protocol, scenario)
duration_errors = [
i for i in result.errors if i.category == "duration" and "exceeds" in i.message
]
assert len(duration_errors) == 0
# ---------------------------------------------------------------------------
# Equipment vocabulary
# ---------------------------------------------------------------------------
def test_unknown_equipment_is_warning() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
required_equipment=["quantum_flux_capacitor"],
)
result = validate_protocol(protocol, scenario)
equip_warnings = [
i for i in result.warnings if i.category == "equipment"
]
assert len(equip_warnings) >= 1
assert "not in the lab" in equip_warnings[0].message
def test_booked_equipment_without_substitution_is_error() -> None:
scenario = _easy_math_scenario()
lab = scenario.lab_manager_observation
if not lab.equipment_booked:
# Force a booked item for the test
lab.equipment_booked.append("special_scope")
lab.equipment_available = [
e for e in lab.equipment_available if e != "special_scope"
]
booked_item = lab.equipment_booked[0]
protocol = _protocol_for_scenario(
scenario,
required_equipment=[booked_item],
)
result = validate_protocol(protocol, scenario)
equip_errors = [i for i in result.errors if i.category == "equipment"]
assert len(equip_errors) >= 1
# ---------------------------------------------------------------------------
# Reagent vocabulary
# ---------------------------------------------------------------------------
def test_out_of_stock_reagent_without_substitution_is_error() -> None:
scenario = _easy_math_scenario()
lab = scenario.lab_manager_observation
if not lab.reagents_out_of_stock:
lab.reagents_out_of_stock.append("unobtainium")
oos_item = lab.reagents_out_of_stock[0]
protocol = _protocol_for_scenario(
scenario,
required_reagents=[oos_item],
)
result = validate_protocol(protocol, scenario)
reagent_errors = [i for i in result.errors if i.category == "reagent"]
assert len(reagent_errors) >= 1
def test_unknown_reagent_is_warning() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
required_reagents=["exotic_compound_99"],
)
result = validate_protocol(protocol, scenario)
reagent_warnings = [i for i in result.warnings if i.category == "reagent"]
assert len(reagent_warnings) >= 1
# ---------------------------------------------------------------------------
# Required-element coverage
# ---------------------------------------------------------------------------
def test_required_element_warning_when_not_addressed() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
technique="unrelated_method",
rationale="No relation to any required element.",
controls=[],
required_equipment=["laptop"],
required_reagents=["water"],
)
result = validate_protocol(protocol, scenario)
element_warnings = [
i for i in result.warnings if i.category == "required_element"
]
assert len(element_warnings) >= 1
# ---------------------------------------------------------------------------
# MOD 06 — Semantic impossibility checks
# ---------------------------------------------------------------------------
def test_zero_sample_with_controls_is_error() -> None:
"""Zero sample_size + non-empty controls is self-contradictory."""
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
sample_size=0,
controls=["baseline", "positive"],
)
result = validate_protocol(protocol, scenario)
assert result.valid is False
sc_errors = [i for i in result.errors if i.category == "sample_controls"]
assert len(sc_errors) >= 1
assert "controls require samples" in sc_errors[0].message
def test_controls_count_gte_sample_size_is_error() -> None:
"""More controls than samples means no experimental samples remain."""
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
sample_size=2,
controls=["baseline", "positive", "negative"],
)
result = validate_protocol(protocol, scenario)
assert result.valid is False
sc_errors = [i for i in result.errors if i.category == "sample_controls"]
assert len(sc_errors) >= 1
assert "no experimental samples remain" in sc_errors[0].message
def test_controls_equal_sample_size_is_error() -> None:
"""Exactly as many controls as samples — still no experimental arm."""
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
sample_size=2,
controls=["baseline", "positive"],
)
result = validate_protocol(protocol, scenario)
assert result.valid is False
sc_errors = [i for i in result.errors if i.category == "sample_controls"]
assert len(sc_errors) >= 1
def test_duplicate_controls_is_warning() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(
scenario,
sample_size=20,
controls=["baseline", "baseline", "positive"],
)
result = validate_protocol(protocol, scenario)
dup_warnings = [i for i in result.warnings if i.category == "duplicate_controls"]
assert len(dup_warnings) == 1
assert "baseline" in dup_warnings[0].message
def test_duplicate_equipment_is_warning() -> None:
scenario = _easy_math_scenario()
lab = scenario.lab_manager_observation
equip = lab.equipment_available[0] if lab.equipment_available else "compute"
protocol = _protocol_for_scenario(
scenario,
required_equipment=[equip, equip],
)
result = validate_protocol(protocol, scenario)
dup_warnings = [i for i in result.warnings if i.category == "duplicate_equipment"]
assert len(dup_warnings) == 1
def test_duplicate_reagents_is_warning() -> None:
scenario = _easy_math_scenario()
lab = scenario.lab_manager_observation
reagent = lab.reagents_in_stock[0] if lab.reagents_in_stock else "data"
protocol = _protocol_for_scenario(
scenario,
required_reagents=[reagent, reagent],
)
result = validate_protocol(protocol, scenario)
dup_warnings = [i for i in result.warnings if i.category == "duplicate_reagents"]
assert len(dup_warnings) == 1
def test_valid_protocol_still_passes_after_mod06() -> None:
"""Ensure MOD 06 checks don't break valid protocols."""
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(scenario, sample_size=10, controls=["baseline"])
result = validate_protocol(protocol, scenario)
assert result.valid is True
assert len(result.errors) == 0
# No MOD 06 categories should appear as errors
mod06_errors = [
i for i in result.errors
if i.category in ("sample_controls", "duplicate_controls", "duplicate_equipment", "duplicate_reagents")
]
assert len(mod06_errors) == 0
def test_no_controls_is_warning() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(scenario, controls=[])
result = validate_protocol(protocol, scenario)
control_warnings = [
i for i in result.warnings if i.category == "controls"
]
assert len(control_warnings) == 1
# ---------------------------------------------------------------------------
# Result structure
# ---------------------------------------------------------------------------
def test_validation_result_never_raises() -> None:
"""validate_protocol should always return a result, never crash."""
scenario = _easy_math_scenario()
protocol = Protocol(
sample_size=0,
controls=[],
technique="nothing",
duration_days=0,
required_equipment=["nonexistent"],
required_reagents=["nonexistent"],
rationale="Completely invalid.",
)
result = validate_protocol(protocol, scenario)
assert isinstance(result, ValidationResult)
assert result.valid is False
assert len(result.issues) > 0
def test_validation_result_json_round_trip() -> None:
scenario = _easy_math_scenario()
protocol = _protocol_for_scenario(scenario, sample_size=0)
result = validate_protocol(protocol, scenario)
dumped = result.model_dump_json()
restored = ValidationResult.model_validate_json(dumped)
assert restored.valid == result.valid
assert len(restored.issues) == len(result.issues)
|