File size: 14,507 Bytes
ba71465 | 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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 | """Verification utilities for migrated rule quality.
This module adds:
- rule-level equivalence checks (Perl evaluator vs generated check),
- mutation testing for generated Python checks.
"""
from __future__ import annotations
import json
import random
from typing import Callable, Dict, List, Mapping, Sequence
def _to_float(value: object) -> float | None:
if value is None:
return None
try:
return float(value)
except (TypeError, ValueError):
return None
def _comparison_pairs(operator: str) -> tuple[tuple[float, float], tuple[float, float]]:
pairs = {
">": ((2.0, 1.0), (1.0, 2.0)),
"<": ((1.0, 2.0), (2.0, 1.0)),
">=": ((2.0, 2.0), (1.0, 2.0)),
"<=": ((2.0, 2.0), (3.0, 2.0)),
"==": ((2.0, 2.0), (2.0, 3.0)),
"!=": ((2.0, 3.0), (2.0, 2.0)),
}
return pairs.get(operator, ((2.0, 1.0), (1.0, 2.0)))
def _threshold_values(operator: str, threshold: float) -> tuple[float, float]:
if operator == ">":
return threshold + 1.0, threshold
if operator == "<":
return threshold - 1.0, threshold
if operator == ">=":
return threshold, threshold - 1.0
if operator == "<=":
return threshold, threshold + 1.0
if operator == "==":
return threshold, threshold + 1.0
if operator == "!=":
return threshold + 1.0, threshold
return threshold + 1.0, threshold
def _dedupe_cases(cases: Sequence[Mapping[str, object]]) -> List[Dict[str, object]]:
deduped: List[Dict[str, object]] = []
seen: set[str] = set()
for case in cases:
payload = dict(case)
key = json.dumps(payload, sort_keys=True, default=str)
if key in seen:
continue
seen.add(key)
deduped.append(payload)
return deduped
def generate_equivalence_cases(
rule: Mapping[str, object],
seed: int = 17,
random_cases: int = 18,
) -> List[Dict[str, object]]:
rng = random.Random(seed + (sum(ord(ch) for ch in str(rule.get("rule_name", ""))) % 997))
condition_type = str(rule.get("condition_type", ""))
cases: List[Dict[str, object]] = []
if condition_type == "field_comparison":
left = str(rule.get("left_operand"))
right = str(rule.get("right_operand"))
operator = str(rule.get("operator"))
true_pair, false_pair = _comparison_pairs(operator)
cases.extend(
[
{left: true_pair[0], right: true_pair[1]},
{left: false_pair[0], right: false_pair[1]},
{left: true_pair[1], right: true_pair[1]},
{left: None, right: true_pair[1]},
{left: true_pair[0], right: None},
]
)
for _ in range(random_cases):
cases.append({left: round(rng.uniform(-10, 200), 3), right: round(rng.uniform(-10, 200), 3)})
elif condition_type == "field_threshold":
left = str(rule.get("left_operand"))
operator = str(rule.get("operator"))
threshold = float(rule.get("right_operand", 0.0))
true_value, false_value = _threshold_values(operator, threshold)
cases.extend([{left: true_value}, {left: false_value}, {left: None}, {left: "nan_text"}])
for _ in range(random_cases):
cases.append({left: round(rng.uniform(threshold - 100, threshold + 100), 3)})
elif condition_type == "missing_field":
field = str(rule.get("left_operand"))
cases.extend(
[
{field: None},
{field: ""},
{field: " "},
{field: "en"},
{field: "xx"},
]
)
elif condition_type == "scaled_field_comparison":
left = str(rule.get("left_operand"))
right = str(rule.get("right_operand"))
operator = str(rule.get("operator"))
factor = float(rule.get("scale_factor", 1.0))
right_value = 10.0
target = right_value * factor
true_value, false_value = _threshold_values(operator, target)
cases.extend(
[
{left: true_value, right: right_value},
{left: false_value, right: right_value},
{left: None, right: right_value},
{left: true_value, right: None},
]
)
for _ in range(random_cases):
random_right = round(rng.uniform(0.1, 150), 3)
random_target = random_right * factor
delta = rng.uniform(-10, 10)
cases.append({left: round(random_target + delta, 3), right: random_right})
elif condition_type == "affine_field_comparison":
left = str(rule.get("left_operand"))
right = str(rule.get("right_operand"))
operator = str(rule.get("operator"))
factor = float(rule.get("scale_factor", 1.0))
offset = float(rule.get("offset", 0.0))
right_value = 10.0
target = (factor * right_value) + offset
true_value, false_value = _threshold_values(operator, target)
cases.extend(
[
{left: true_value, right: right_value},
{left: false_value, right: right_value},
{left: None, right: right_value},
{left: true_value, right: None},
]
)
for _ in range(random_cases):
random_right = round(rng.uniform(0.1, 150), 3)
random_target = (factor * random_right) + offset
delta = rng.uniform(-10, 10)
cases.append({left: round(random_target + delta, 3), right: random_right})
elif condition_type == "sum_fields_comparison":
left_operands = list(rule.get("left_operands", []))
if len(left_operands) >= 2:
left_a = str(left_operands[0])
left_b = str(left_operands[1])
right = str(rule.get("right_operand"))
operator = str(rule.get("operator"))
right_offset = float(rule.get("right_offset", 0.0))
right_value = 20.0
target = right_value + right_offset
true_sum, false_sum = _threshold_values(operator, target)
cases.extend(
[
{left_a: true_sum / 2.0, left_b: true_sum / 2.0, right: right_value},
{left_a: false_sum / 2.0, left_b: false_sum / 2.0, right: right_value},
{left_a: None, left_b: 2.0, right: right_value},
{left_a: 2.0, left_b: None, right: right_value},
]
)
for _ in range(random_cases):
random_right = round(rng.uniform(0.1, 120), 3)
random_target = random_right + right_offset
left_sum = random_target + rng.uniform(-20, 20)
left_part = round(rng.uniform(0, max(left_sum, 0.1)), 3)
cases.append(
{
left_a: left_part,
left_b: round(left_sum - left_part, 3),
right: random_right,
}
)
elif condition_type == "compound_threshold_and":
clauses = list(rule.get("clauses", []))
passing: Dict[str, object] = {}
failing: Dict[str, object] = {}
for idx, clause in enumerate(clauses):
field = str(clause.get("left_operand"))
operator = str(clause.get("operator"))
threshold = float(clause.get("right_operand", 0.0))
true_value, false_value = _threshold_values(operator, threshold)
passing[field] = true_value
failing[field] = true_value
if idx == 0:
failing[field] = false_value
if passing:
cases.append(passing)
if failing:
cases.append(failing)
return _dedupe_cases(cases)
def evaluate_rule_equivalence(
rule: Mapping[str, object],
perl_evaluator: Callable[[Mapping[str, object]], bool],
check_fn: Callable[[Mapping[str, object]], object],
seed: int = 17,
cases: Sequence[Mapping[str, object]] | None = None,
max_counterexamples: int = 5,
) -> Dict[str, object]:
sample_cases = list(cases) if cases is not None else generate_equivalence_cases(rule, seed=seed)
tag = str(rule.get("tag"))
matches = 0
mismatches = 0
counterexamples: List[Dict[str, object]] = []
for case in sample_cases:
product = dict(case)
expected = tag if bool(perl_evaluator(product)) else None
try:
actual = check_fn(product)
except Exception as exc: # noqa: BLE001
actual = f"EXCEPTION:{exc.__class__.__name__}"
if actual == expected:
matches += 1
else:
mismatches += 1
if len(counterexamples) < max_counterexamples:
counterexamples.append({"input": product, "expected": expected, "actual": actual})
total = len(sample_cases)
rate = (matches / total) if total else 1.0
return {
"equivalence_cases": total,
"equivalence_matches": matches,
"equivalence_mismatches": mismatches,
"equivalence_match_rate": round(rate, 4),
"equivalence_status": "PASS" if mismatches == 0 else "FAIL",
"counterexamples": counterexamples,
}
def _mutate_once(code: str, old: str, new: str) -> str | None:
if old not in code:
return None
mutated = code.replace(old, new, 1)
if mutated == code:
return None
return mutated
def build_mutants(rule: Mapping[str, object], python_code: str) -> List[Dict[str, object]]:
mutants: List[Dict[str, object]] = []
condition_type = str(rule.get("condition_type", ""))
operator = str(rule.get("operator", ""))
operator_swap = {">": ">=", "<": "<=", ">=": ">", "<=": "<", "==": "!=", "!=": "=="}
swapped = operator_swap.get(operator)
if swapped:
mutated = _mutate_once(python_code, f" {operator} ", f" {swapped} ")
if mutated is not None:
mutants.append({"name": f"operator_{operator}_to_{swapped}", "code": mutated})
if condition_type == "missing_field":
mutated = _mutate_once(python_code, 'or str(value).strip() == ""', 'and str(value).strip() == ""')
if mutated is not None:
mutants.append({"name": "missing_logic_or_to_and", "code": mutated})
if condition_type == "sum_fields_comparison":
mutated = _mutate_once(python_code, "left_sum = left_a_value + left_b_value", "left_sum = left_a_value - left_b_value")
if mutated is not None:
mutants.append({"name": "sum_to_difference", "code": mutated})
scale_factor = rule.get("scale_factor")
if isinstance(scale_factor, (int, float)):
old = str(float(scale_factor))
new = str(round(float(scale_factor) + 0.3, 6))
mutated = _mutate_once(python_code, old, new)
if mutated is not None:
mutants.append({"name": "scale_factor_perturbed", "code": mutated})
offset = rule.get("offset")
if isinstance(offset, (int, float)):
old = str(float(offset))
new = str(round(float(offset) + 1.0, 6))
mutated = _mutate_once(python_code, old, new)
if mutated is not None:
mutants.append({"name": "offset_perturbed", "code": mutated})
threshold = rule.get("right_operand")
if condition_type == "field_threshold" and isinstance(threshold, (int, float)):
old = str(float(threshold))
new = str(round(float(threshold) + 1.0, 6))
mutated = _mutate_once(python_code, old, new)
if mutated is not None:
mutants.append({"name": "threshold_perturbed", "code": mutated})
deduped: List[Dict[str, object]] = []
seen: set[str] = set()
for mutant in mutants:
code = str(mutant["code"])
if code in seen:
continue
seen.add(code)
deduped.append(mutant)
return deduped[:8]
def evaluate_mutation_suite(
rule: Mapping[str, object],
perl_evaluator: Callable[[Mapping[str, object]], bool],
python_code: str,
function_name: str,
seed: int = 17,
) -> Dict[str, object]:
cases = generate_equivalence_cases(rule, seed=seed)
mutants = build_mutants(rule, python_code)
total = 0
killed = 0
survived: List[str] = []
for mutant in mutants:
namespace: Dict[str, object] = {}
try:
exec(str(mutant["code"]), {}, namespace)
fn = namespace.get(function_name)
if not callable(fn):
continue
except Exception:
continue
total += 1
result = evaluate_rule_equivalence(
rule=rule,
perl_evaluator=perl_evaluator,
check_fn=fn,
seed=seed,
cases=cases,
max_counterexamples=1,
)
if int(result["equivalence_mismatches"]) > 0:
killed += 1
else:
survived.append(str(mutant["name"]))
score = (killed / total) if total else 1.0
return {
"mutation_total": total,
"mutation_killed": killed,
"mutation_survived": total - killed,
"mutation_score": round(score, 4),
"mutation_survived_mutants": survived,
}
def run_rule_verification(
rule: Mapping[str, object],
perl_evaluator: Callable[[Mapping[str, object]], bool],
check_fn: Callable[[Mapping[str, object]], object],
python_code: str,
function_name: str,
seed: int = 17,
) -> Dict[str, object]:
equivalence = evaluate_rule_equivalence(
rule=rule,
perl_evaluator=perl_evaluator,
check_fn=check_fn,
seed=seed,
)
mutation = evaluate_mutation_suite(
rule=rule,
perl_evaluator=perl_evaluator,
python_code=python_code,
function_name=function_name,
seed=seed,
)
verification_score = float(equivalence["equivalence_match_rate"]) * float(mutation["mutation_score"])
return {
**equivalence,
**mutation,
"verification_score": round(verification_score, 4),
}
|