| 1 | #!/usr/bin/env python3 |
| 2 | from __future__ import annotations |
| 3 | |
| 4 | import argparse |
| 5 | import hashlib |
| 6 | import json |
| 7 | import os |
| 8 | import re |
| 9 | import sys |
| 10 | from datetime import UTC, datetime |
| 11 | from pathlib import Path |
| 12 | from typing import Any |
| 13 | |
| 14 | from scripts.render_press_context import NO_PRESS_SENTINEL_MARKER |
| 15 | |
| 16 | try: # pragma: no cover - optional dependency on runners |
| 17 | import yaml |
| 18 | except ImportError: # pragma: no cover - exercised via fallback parser |
| 19 | yaml = None |
| 20 | |
| 21 | REQUIRED_FIELDS = [ |
| 22 | "title", |
| 23 | "date", |
| 24 | "week", |
| 25 | "year", |
| 26 | "tags", |
| 27 | "categories", |
| 28 | "repos_featured", |
| 29 | "stars_tracked", |
| 30 | "top_repo", |
| 31 | "quality_score", |
| 32 | "summary", |
| 33 | ] |
| 34 | OPTIONAL_FIELDS = ["predictions"] |
| 35 | PREDICTION_DIRECTIONS = {"up", "flat", "down"} |
| 36 | PREDICTION_CLAIM_TYPES = {"signal", "noise", "gap"} |
| 37 | PREDICTION_FIELDS = {"repo", "claim_type", "direction", "confidence"} |
| 38 | REQUIRED_HEADINGS = [ |
| 39 | "## This Week's Trends", |
| 40 | "## Where Industry Meets Code", |
| 41 | "## Signal & Noise", |
| 42 | "## Blind Spots", |
| 43 | "## The Week Ahead", |
| 44 | "## Key References", |
| 45 | "### Notable Projects", |
| 46 | "### Press & Industry", |
| 47 | ] |
| 48 | PUBLISHABLE_AI_SOURCES = {"copilot-cli"} |
| 49 | UNPUBLISHABLE_MODEL_VALUES = {"", "unknown", "unavailable", "none", "no-ai"} |
| 50 | RAW_MARKERS = [ |
| 51 | "```json", |
| 52 | '"week":', |
| 53 | '"new_repos"', |
| 54 | '"trending_repos"', |
| 55 | "traceback (most recent call last)", |
| 56 | ] |
| 57 | PLACEHOLDER_PATTERNS = [ |
| 58 | (re.compile(r"(?mi)^\s*(?:[-*]\s*)?todo\s*[:\-]"), "TODO placeholder marker"), |
| 59 | (re.compile(r"(?mi)^\s*(?:[-*]\s*)?tbd\s*[:\-]"), "TBD placeholder marker"), |
| 60 | (re.compile(r"(?i)\bplaceholder text\b"), "placeholder text"), |
| 61 | (re.compile(r"(?i)\byour analysis here\b"), "placeholder instruction"), |
| 62 | ] |
| 63 | WEEK_PATTERN = re.compile(r"^(?P<year>\d{4})-W(?P<week>\d{2})$") |
| 64 | FRONTMATTER_PATTERN = re.compile(r"^---\n(.*?)\n---\n(.*)\Z", re.DOTALL) |
| 65 | HEADING_PATTERN = re.compile(r"(?m)^(#{2,3})\s+(.+?)\s*$") |
| 66 | WORD_PATTERN = re.compile(r"\b[\w'-]+\b") |
| 67 | TOP_REPO_PATTERN = re.compile(r"^[^/\s]+/[^/\s]+$") |
| 68 | GENERIC_TITLE_PATTERNS = [ |
| 69 | re.compile(r"^Week\s+\d+.*Analysis$", re.IGNORECASE), |
| 70 | re.compile(r"^Week\s+\d+,\s*\d{4}$", re.IGNORECASE), |
| 71 | ] |
| 72 | REPO_LINK_PATTERN = re.compile(r"\[([^/\]\s]+/[^/\]\s]+)\]\(https://github\.com/\1\)") |
| 73 | SECTION_MIN_WORDS = { |
| 74 | "## This Week's Trends": 60, |
| 75 | "## Where Industry Meets Code": 40, |
| 76 | "## Signal & Noise": 50, |
| 77 | "## Blind Spots": 30, |
| 78 | "## The Week Ahead": 30, |
| 79 | } |
| 80 | EDITORIAL_TERMS = { |
| 81 | "signal", |
| 82 | "noise", |
| 83 | "gap", |
| 84 | "gaps", |
| 85 | "durable", |
| 86 | "hype", |
| 87 | "matters", |
| 88 | "evidence", |
| 89 | "trend", |
| 90 | "trends", |
| 91 | "blind", |
| 92 | "missing", |
| 93 | "practitioners", |
| 94 | "ecosystem", |
| 95 | "observability", |
| 96 | "security", |
| 97 | "testing", |
| 98 | } |
| 99 | EXPLANATORY_PATTERN = re.compile( |
| 100 | r"\b(because|why|matters|signals|reveals|driven|shows|suggests|represents|means|confirms|indicates|constitutes)\b", |
| 101 | re.IGNORECASE, |
| 102 | ) |
| 103 | CONTRADICTION_PATTERNS = [ |
| 104 | ( |
| 105 | re.compile(r"no press data was provided", re.IGNORECASE), |
| 106 | re.compile(r"\b(reported|techcrunch)\b", re.IGNORECASE), |
| 107 | "claims no press data was provided while also describing press coverage.", |
| 108 | ), |
| 109 | ( |
| 110 | re.compile(r"no meaningful developer activity", re.IGNORECASE), |
| 111 | REPO_LINK_PATTERN, |
| 112 | "claims no meaningful developer activity while citing active repositories.", |
| 113 | ), |
| 114 | ] |
| 115 | # "No press data" style statements that must NOT appear in the published body when |
| 116 | # the week's press context is actually populated. Unlike CONTRADICTION_PATTERNS these |
| 117 | # fire even when the body does not self-contradict (no other press discussion present) — |
| 118 | # they catch the false-negative where a populated press-context.md is silently dropped. |
| 119 | STALE_PRESS_CLAIM_PATTERNS = [ |
| 120 | ( |
| 121 | re.compile(r"no industry press data was available", re.IGNORECASE), |
| 122 | 'body states "No industry press data was available for this week\'s analysis." ' |
| 123 | "while a populated press context exists for this week.", |
| 124 | ), |
| 125 | ( |
| 126 | re.compile(r"no press data was provided this week", re.IGNORECASE), |
| 127 | 'Key References state "No press data was provided this week." ' |
| 128 | "while a populated press context exists for this week.", |
| 129 | ), |
| 130 | ] |
| 131 | |
| 132 | |
| 133 | def parse_args(argv: list[str] | None = None) -> argparse.Namespace: |
| 134 | parser = argparse.ArgumentParser( |
| 135 | description="Validate weekly analysis output against the analysis spec." |
| 136 | ) |
| 137 | parser.add_argument( |
| 138 | "--analysis-file", required=True, type=Path, help="Path to the rendered markdown summary." |
| 139 | ) |
| 140 | parser.add_argument( |
| 141 | "--raw-json", required=True, type=Path, help="Path to the raw weekly payload." |
| 142 | ) |
| 143 | parser.add_argument( |
| 144 | "--current-datetime", required=True, help="Current run timestamp in ISO 8601 format." |
| 145 | ) |
| 146 | parser.add_argument("--source", default="unknown", help="Analysis source label for summaries.") |
| 147 | parser.add_argument( |
| 148 | "--model", default="copilot-default", help="AI model label for provenance validation." |
| 149 | ) |
| 150 | parser.add_argument( |
| 151 | "--repair-safe", |
| 152 | action="store_true", |
| 153 | help="Apply deterministic frontmatter/schema repairs before final validation.", |
| 154 | ) |
| 155 | parser.add_argument("--report-json", type=Path, help="Write a machine-readable gate report.") |
| 156 | parser.add_argument( |
| 157 | "--press-context-path", |
| 158 | type=Path, |
| 159 | default=None, |
| 160 | help="Path to the week's rendered press-context.md. When populated, the gate fails " |
| 161 | 'analyses that still claim "no press data".', |
| 162 | ) |
| 163 | parser.add_argument( |
| 164 | "--press-token-estimate", |
| 165 | type=int, |
| 166 | default=None, |
| 167 | help="Optional token estimate for the week's press context; a value > 0 marks the " |
| 168 | "press context as populated even without the rendered file.", |
| 169 | ) |
| 170 | return parser.parse_args(argv) |
| 171 | |
| 172 | |
| 173 | def parse_datetime(value: str | datetime) -> datetime: |
| 174 | if isinstance(value, datetime): |
| 175 | parsed = value |
| 176 | elif isinstance(value, str): |
| 177 | candidate = value.strip() |
| 178 | if candidate.endswith("Z"): |
| 179 | candidate = f"{candidate[:-1]}+00:00" |
| 180 | parsed = datetime.fromisoformat(candidate) |
| 181 | else: # pragma: no cover - guarded by callers |
| 182 | raise TypeError(f"Unsupported datetime value: {value!r}") |
| 183 | return parsed if parsed.tzinfo else parsed.replace(tzinfo=UTC) |
| 184 | |
| 185 | |
| 186 | def week_slug(value: datetime) -> str: |
| 187 | year, week, _ = value.astimezone(UTC).isocalendar() |
| 188 | return f"{year}-W{week:02d}" |
| 189 | |
| 190 | |
| 191 | def load_json(path: Path) -> dict[str, Any]: |
| 192 | payload = json.loads(path.read_text(encoding="utf-8")) |
| 193 | if not isinstance(payload, dict): |
| 194 | raise ValueError(f"Raw payload must be an object: {path}") |
| 195 | return payload |
| 196 | |
| 197 | |
| 198 | def strip_quotes(value: str) -> str: |
| 199 | stripped = value.strip() |
| 200 | if len(stripped) >= 2 and stripped[0] == stripped[-1] and stripped[0] in {'"', "'"}: |
| 201 | return stripped[1:-1] |
| 202 | return stripped |
| 203 | |
| 204 | |
| 205 | def parse_inline_list(value: str) -> list[str]: |
| 206 | inner = value.strip()[1:-1].strip() |
| 207 | if not inner: |
| 208 | return [] |
| 209 | items: list[str] = [] |
| 210 | for part in inner.split(","): |
| 211 | item = strip_quotes(part) |
| 212 | if not item: |
| 213 | raise ValueError(f"Malformed YAML list entry: {value}") |
| 214 | items.append(item) |
| 215 | return items |
| 216 | |
| 217 | |
| 218 | def parse_scalar(value: str) -> Any: |
| 219 | scalar = strip_quotes(value) |
| 220 | if re.fullmatch(r"-?\d+", scalar): |
| 221 | return int(scalar) |
| 222 | if re.fullmatch(r"-?\d+\.\d+", scalar): |
| 223 | return float(scalar) |
| 224 | return scalar |
| 225 | |
| 226 | |
| 227 | def parse_frontmatter_fallback(text: str) -> dict[str, Any]: |
| 228 | frontmatter: dict[str, Any] = {} |
| 229 | lines = text.splitlines() |
| 230 | index = 0 |
| 231 | while index < len(lines): |
| 232 | line = lines[index] |
| 233 | if not line.strip(): |
| 234 | index += 1 |
| 235 | continue |
| 236 | if line.startswith((" ", "\t")): |
| 237 | raise ValueError(f"Unexpected indentation in frontmatter: {line}") |
| 238 | if ":" not in line: |
| 239 | raise ValueError(f"Malformed frontmatter line: {line}") |
| 240 | key, raw_value = line.split(":", 1) |
| 241 | key = key.strip() |
| 242 | value = raw_value.strip() |
| 243 | if not key: |
| 244 | raise ValueError(f"Malformed frontmatter key: {line}") |
| 245 | if value == "": |
| 246 | items: list[Any] = [] |
| 247 | index += 1 |
| 248 | while index < len(lines): |
| 249 | candidate = lines[index] |
| 250 | if not candidate.strip(): |
| 251 | index += 1 |
| 252 | continue |
| 253 | candidate_indent = len(candidate) - len(candidate.lstrip(" \t")) |
| 254 | if candidate_indent == 0: |
| 255 | break |
| 256 | stripped = candidate.strip() |
| 257 | if not stripped.startswith("- "): |
| 258 | raise ValueError( |
| 259 | f"Unsupported multiline frontmatter value for {key}: {candidate}" |
| 260 | ) |
| 261 | item_value = stripped[2:].strip() |
| 262 | if ":" in item_value: |
| 263 | item: dict[str, Any] = {} |
| 264 | item_key, item_raw_value = item_value.split(":", 1) |
| 265 | item[item_key.strip()] = parse_scalar(item_raw_value.strip()) |
| 266 | index += 1 |
| 267 | while index < len(lines): |
| 268 | nested = lines[index] |
| 269 | if not nested.strip(): |
| 270 | index += 1 |
| 271 | continue |
| 272 | nested_indent = len(nested) - len(nested.lstrip(" \t")) |
| 273 | if nested_indent <= candidate_indent: |
| 274 | break |
| 275 | if ":" not in nested: |
| 276 | raise ValueError( |
| 277 | f"Unsupported multiline frontmatter value for {key}: {nested}" |
| 278 | ) |
| 279 | nested_key, nested_raw_value = nested.strip().split(":", 1) |
| 280 | item[nested_key.strip()] = parse_scalar(nested_raw_value.strip()) |
| 281 | index += 1 |
| 282 | items.append(item) |
| 283 | continue |
| 284 | items.append(parse_scalar(item_value)) |
| 285 | index += 1 |
| 286 | frontmatter[key] = items |
| 287 | continue |
| 288 | if value.startswith("[") and value.endswith("]"): |
| 289 | frontmatter[key] = parse_inline_list(value) |
| 290 | else: |
| 291 | frontmatter[key] = parse_scalar(value) |
| 292 | index += 1 |
| 293 | return frontmatter |
| 294 | |
| 295 | |
| 296 | def parse_frontmatter(text: str) -> dict[str, Any]: |
| 297 | if yaml is not None: |
| 298 | try: |
| 299 | data = yaml.safe_load(text) or {} |
| 300 | except Exception: |
| 301 | data = parse_frontmatter_fallback(text) |
| 302 | else: |
| 303 | if not isinstance(data, dict): |
| 304 | raise ValueError("YAML frontmatter must be a mapping.") |
| 305 | return data |
| 306 | return parse_frontmatter_fallback(text) |
| 307 | |
| 308 | |
| 309 | def dump_frontmatter(data: dict[str, Any]) -> str: |
| 310 | if yaml is not None: |
| 311 | dumped = yaml.safe_dump(data, sort_keys=False, allow_unicode=True).strip() |
| 312 | return re.sub(r"^(date): '([^']+)'$", r"\1: \2", dumped, flags=re.MULTILINE) |
| 313 | |
| 314 | def format_scalar(value: Any) -> str: |
| 315 | if isinstance(value, str): |
| 316 | return json.dumps(value) |
| 317 | if isinstance(value, (int, float)): |
| 318 | return str(value) |
| 319 | raise TypeError(f"Unsupported frontmatter value for fallback dump: {value!r}") |
| 320 | |
| 321 | lines: list[str] = [] |
| 322 | for key, value in data.items(): |
| 323 | if isinstance(value, list): |
| 324 | if not value: |
| 325 | lines.append(f"{key}: []") |
| 326 | continue |
| 327 | lines.append(f"{key}:") |
| 328 | for item in value: |
| 329 | if isinstance(item, dict): |
| 330 | if not item: |
| 331 | lines.append(" - {}") |
| 332 | continue |
| 333 | item_fields = list(item.items()) |
| 334 | first_key, first_value = item_fields[0] |
| 335 | lines.append(f" - {first_key}: {format_scalar(first_value)}") |
| 336 | for nested_key, nested_value in item_fields[1:]: |
| 337 | lines.append(f" {nested_key}: {format_scalar(nested_value)}") |
| 338 | else: |
| 339 | lines.append(f" - {format_scalar(item)}") |
| 340 | else: |
| 341 | lines.append(f"{key}: {format_scalar(value)}") |
| 342 | return "\n".join(lines) |
| 343 | |
| 344 | |
| 345 | def extract_frontmatter(text: str) -> tuple[dict[str, Any], str]: |
| 346 | match = FRONTMATTER_PATTERN.match(text) |
| 347 | if not match: |
| 348 | raise ValueError("Analysis output is missing YAML frontmatter.") |
| 349 | frontmatter_text, body = match.groups() |
| 350 | return parse_frontmatter(frontmatter_text), body |
| 351 | |
| 352 | |
| 353 | def render_analysis(frontmatter: dict[str, Any], body: str) -> str: |
| 354 | return f"---\n{dump_frontmatter(frontmatter)}\n---\n{body}" |
| 355 | |
| 356 | |
| 357 | def expected_repo_counts(raw_payload: dict[str, Any]) -> tuple[int, int]: |
| 358 | repos: list[dict[str, Any]] = [] |
| 359 | for field in ("new_repos", "trending_repos"): |
| 360 | value = raw_payload.get(field) |
| 361 | if isinstance(value, list): |
| 362 | repos.extend(item for item in value if isinstance(item, dict)) |
| 363 | stars = sum( |
| 364 | star |
| 365 | for repo in repos |
| 366 | if isinstance((star := repo.get("stars")), int) and not isinstance(star, bool) |
| 367 | ) |
| 368 | return len(repos), stars |
| 369 | |
| 370 | |
| 371 | def repair_analysis( |
| 372 | text: str, |
| 373 | raw_payload: dict[str, Any], |
| 374 | current_datetime: str, |
| 375 | ) -> tuple[str, list[str]]: |
| 376 | frontmatter, body = extract_frontmatter(text) |
| 377 | repaired = dict(frontmatter) |
| 378 | actions: list[str] = [] |
| 379 | |
| 380 | expected_week = raw_payload.get("week") |
| 381 | week_match = WEEK_PATTERN.fullmatch(expected_week) if isinstance(expected_week, str) else None |
| 382 | if isinstance(expected_week, str) and repaired.get("week") != expected_week: |
| 383 | repaired["week"] = expected_week |
| 384 | actions.append(f"set week from raw payload ({expected_week})") |
| 385 | if week_match: |
| 386 | expected_year = int(week_match.group("year")) |
| 387 | if repaired.get("year") != expected_year: |
| 388 | repaired["year"] = expected_year |
| 389 | actions.append(f"set year from raw payload week ({expected_year})") |
| 390 | if repaired.get("date") != current_datetime: |
| 391 | repaired["date"] = current_datetime |
| 392 | actions.append("set date from current run timestamp") |
| 393 | |
| 394 | repos_featured, stars_tracked = expected_repo_counts(raw_payload) |
| 395 | if repaired.get("repos_featured") != repos_featured: |
| 396 | repaired["repos_featured"] = repos_featured |
| 397 | actions.append(f"set repos_featured from raw repo counts ({repos_featured})") |
| 398 | if repaired.get("stars_tracked") != stars_tracked: |
| 399 | repaired["stars_tracked"] = stars_tracked |
| 400 | actions.append(f"set stars_tracked from raw repo stars ({stars_tracked})") |
| 401 | |
| 402 | predictions = repaired.get("predictions") |
| 403 | if isinstance(predictions, list): |
| 404 | repaired_predictions = [] |
| 405 | changed_predictions = False |
| 406 | for index, prediction in enumerate(predictions, start=1): |
| 407 | if not isinstance(prediction, dict): |
| 408 | repaired_predictions.append(prediction) |
| 409 | continue |
| 410 | repaired_prediction = dict(prediction) |
| 411 | if "claim_type" not in repaired_prediction: |
| 412 | for alias in ("claim", "claimType", "type", "kind"): |
| 413 | alias_value = repaired_prediction.get(alias) |
| 414 | if ( |
| 415 | isinstance(alias_value, str) |
| 416 | and alias_value.strip().lower() in PREDICTION_CLAIM_TYPES |
| 417 | ): |
| 418 | repaired_prediction["claim_type"] = alias_value.strip().lower() |
| 419 | del repaired_prediction[alias] |
| 420 | changed_predictions = True |
| 421 | actions.append(f"set predictions[{index}].claim_type from {alias}") |
| 422 | break |
| 423 | claim_type = repaired_prediction.get("claim_type") |
| 424 | if ( |
| 425 | isinstance(claim_type, str) |
| 426 | and claim_type.strip().lower() in PREDICTION_CLAIM_TYPES |
| 427 | and claim_type != claim_type.strip().lower() |
| 428 | ): |
| 429 | repaired_prediction["claim_type"] = claim_type.strip().lower() |
| 430 | changed_predictions = True |
| 431 | actions.append(f"normalized predictions[{index}].claim_type") |
| 432 | direction = repaired_prediction.get("direction") |
| 433 | if ( |
| 434 | isinstance(direction, str) |
| 435 | and direction.strip().lower() in PREDICTION_DIRECTIONS |
| 436 | and direction != direction.strip().lower() |
| 437 | ): |
| 438 | repaired_prediction["direction"] = direction.strip().lower() |
| 439 | changed_predictions = True |
| 440 | actions.append(f"normalized predictions[{index}].direction") |
| 441 | repaired_predictions.append(repaired_prediction) |
| 442 | if changed_predictions: |
| 443 | repaired["predictions"] = repaired_predictions |
| 444 | |
| 445 | if not actions: |
| 446 | return text, actions |
| 447 | return render_analysis(repaired, body), actions |
| 448 | |
| 449 | |
| 450 | def validate_string_field(frontmatter: dict[str, Any], field: str, errors: list[str]) -> None: |
| 451 | value = frontmatter.get(field) |
| 452 | if value is None: |
| 453 | return |
| 454 | if not isinstance(value, str) or not value.strip(): |
| 455 | errors.append(f"{field} must be a non-empty string.") |
| 456 | |
| 457 | |
| 458 | def validate_integer_field( |
| 459 | frontmatter: dict[str, Any], field: str, errors: list[str], *, minimum: int = 0 |
| 460 | ) -> None: |
| 461 | value = frontmatter.get(field) |
| 462 | if value is None: |
| 463 | return |
| 464 | if isinstance(value, bool) or not isinstance(value, int): |
| 465 | errors.append(f"{field} must be an integer.") |
| 466 | return |
| 467 | if value < minimum: |
| 468 | errors.append(f"{field} must be at least {minimum}.") |
| 469 | |
| 470 | |
| 471 | def validate_string_list( |
| 472 | frontmatter: dict[str, Any], |
| 473 | field: str, |
| 474 | errors: list[str], |
| 475 | *, |
| 476 | minimum: int | None = None, |
| 477 | maximum: int | None = None, |
| 478 | includes: str | None = None, |
| 479 | ) -> None: |
| 480 | value = frontmatter.get(field) |
| 481 | if value is None: |
| 482 | return |
| 483 | if not isinstance(value, list) or any( |
| 484 | not isinstance(item, str) or not item.strip() for item in value |
| 485 | ): |
| 486 | errors.append(f"{field} must be an array of strings.") |
| 487 | return |
| 488 | if minimum is not None and len(value) < minimum: |
| 489 | errors.append(f"{field} must contain at least {minimum} items.") |
| 490 | if maximum is not None and len(value) > maximum: |
| 491 | errors.append(f"{field} must contain at most {maximum} items.") |
| 492 | if includes is not None and includes not in value: |
| 493 | errors.append(f"{field} must include {includes!r}.") |
| 494 | |
| 495 | |
| 496 | def validate_predictions(frontmatter: dict[str, Any], errors: list[str]) -> None: |
| 497 | predictions = frontmatter.get("predictions") |
| 498 | if predictions is None: |
| 499 | return |
| 500 | if not isinstance(predictions, list): |
| 501 | errors.append("predictions must be an array.") |
| 502 | return |
| 503 | for index, prediction in enumerate(predictions, start=1): |
| 504 | if not isinstance(prediction, dict): |
| 505 | errors.append(f"predictions[{index}] must be an object.") |
| 506 | continue |
| 507 | repo = prediction.get("repo") |
| 508 | claim_type = prediction.get("claim_type") |
| 509 | direction = prediction.get("direction") |
| 510 | confidence = prediction.get("confidence") |
| 511 | if not isinstance(repo, str) or not TOP_REPO_PATTERN.fullmatch(repo.strip()): |
| 512 | errors.append(f"predictions[{index}].repo must use owner/repo format.") |
| 513 | if ( |
| 514 | not isinstance(claim_type, str) |
| 515 | or claim_type.strip().lower() not in PREDICTION_CLAIM_TYPES |
| 516 | ): |
| 517 | errors.append(f"predictions[{index}].claim_type must be one of signal, noise, gap.") |
| 518 | if not isinstance(direction, str) or direction.strip().lower() not in PREDICTION_DIRECTIONS: |
| 519 | errors.append(f"predictions[{index}].direction must be one of up, flat, down.") |
| 520 | if isinstance(confidence, bool) or not isinstance(confidence, (int, float)): |
| 521 | errors.append(f"predictions[{index}].confidence must be numeric.") |
| 522 | elif not 0 <= float(confidence) <= 1: |
| 523 | errors.append(f"predictions[{index}].confidence must be between 0 and 1.") |
| 524 | extra_fields = sorted(set(prediction) - PREDICTION_FIELDS) |
| 525 | if extra_fields: |
| 526 | errors.append(f"predictions[{index}] has unexpected fields: {', '.join(extra_fields)}") |
| 527 | |
| 528 | |
| 529 | def find_missing_headings(body: str) -> list[str]: |
| 530 | headings = [f"{level} {title.strip()}" for level, title in HEADING_PATTERN.findall(body)] |
| 531 | missing: list[str] = [] |
| 532 | position = -1 |
| 533 | for required in REQUIRED_HEADINGS: |
| 534 | try: |
| 535 | position = headings.index(required, position + 1) |
| 536 | except ValueError: |
| 537 | missing.append(required) |
| 538 | return missing |
| 539 | |
| 540 | |
| 541 | def section_text(body: str, heading: str) -> str: |
| 542 | heading_match = re.search(rf"(?m)^{re.escape(heading)}\s*$", body) |
| 543 | if heading_match is None: |
| 544 | return "" |
| 545 | next_heading = re.search(r"(?m)^##\s+", body[heading_match.end() :]) |
| 546 | end = heading_match.end() + next_heading.start() if next_heading else len(body) |
| 547 | return body[heading_match.end() : end].strip() |
| 548 | |
| 549 | |
| 550 | def raw_repo_names(raw_payload: dict[str, Any]) -> set[str]: |
| 551 | names: set[str] = set() |
| 552 | for field in ("new_repos", "trending_repos"): |
| 553 | repos = raw_payload.get(field) |
| 554 | if not isinstance(repos, list): |
| 555 | continue |
| 556 | for repo in repos: |
| 557 | if isinstance(repo, dict) and isinstance(repo.get("full_name"), str): |
| 558 | names.add(repo["full_name"].strip()) |
| 559 | return {name for name in names if TOP_REPO_PATTERN.fullmatch(name)} |
| 560 | |
| 561 | |
| 562 | def compute_objective_quality( |
| 563 | text: str, raw_payload: dict, press_context_available: bool |
| 564 | ) -> tuple[int, dict]: |
| 565 | _, body = extract_frontmatter(text) |
| 566 | words = len(WORD_PATTERN.findall(body)) |
| 567 | depth = round(min(15, max(0, (words - 200) / 1000 * 15))) |
| 568 | |
| 569 | available_repos = raw_repo_names(raw_payload) |
| 570 | cited_repos = set(REPO_LINK_PATTERN.findall(body)).intersection(available_repos) |
| 571 | evidence_target = min(10, len(available_repos)) |
| 572 | evidence = ( |
| 573 | 0 if evidence_target == 0 else round(min(10, len(cited_repos) / evidence_target * 10)) |
| 574 | ) |
| 575 | |
| 576 | press_citations = 0 |
| 577 | press = 0 |
| 578 | if press_context_available: |
| 579 | key_references = section_text(body, "## Key References") |
| 580 | press_section = section_text(key_references, "### Press & Industry") |
| 581 | # section_text() only terminates on the next level-2 heading, so trim at the next |
| 582 | # level-3 subsection to avoid counting URLs from later ### blocks as press citations. |
| 583 | next_subsection = re.search(r"(?m)^###\s+", press_section) |
| 584 | if next_subsection: |
| 585 | press_section = press_section[: next_subsection.start()] |
| 586 | urls = set(re.findall(r"https?://[^\s)\]]+", press_section)) |
| 587 | external_urls = { |
| 588 | url |
| 589 | for url in urls |
| 590 | if not re.match( |
| 591 | r"https?://(?:[^/\s]+\.)?(?:github\.com|githubusercontent\.com)(?:[/:?#]|$)", |
| 592 | url, |
| 593 | re.IGNORECASE, |
| 594 | ) |
| 595 | } |
| 596 | press_citations = len(external_urls) |
| 597 | press = round(min(15, press_citations / 3 * 15)) |
| 598 | |
| 599 | # Identical content with cited press must score strictly above its press-less variant. |
| 600 | score = min(100, 60 + depth + evidence + press) |
| 601 | return score, { |
| 602 | "base": 60, |
| 603 | "depth": depth, |
| 604 | "evidence": evidence, |
| 605 | "press": press, |
| 606 | "words": words, |
| 607 | "repo_citations": len(cited_repos), |
| 608 | "press_citations": press_citations, |
| 609 | "press_available": press_context_available, |
| 610 | } |
| 611 | |
| 612 | |
| 613 | def set_frontmatter_quality_score(text: str, score: int) -> str: |
| 614 | match = FRONTMATTER_PATTERN.match(text) |
| 615 | if not match: |
| 616 | raise ValueError("Analysis output is missing YAML frontmatter.") |
| 617 | frontmatter_text, body = match.groups() |
| 618 | rewritten_frontmatter, replacements = re.subn( |
| 619 | r"(?m)^quality_score:.*$", |
| 620 | f"quality_score: {score}", |
| 621 | frontmatter_text, |
| 622 | count=1, |
| 623 | ) |
| 624 | if replacements == 0: |
| 625 | rewritten_frontmatter = f"{frontmatter_text}\nquality_score: {score}" |
| 626 | return f"---\n{rewritten_frontmatter}\n---\n{body}" |
| 627 | |
| 628 | |
| 629 | def raw_artifact_week_errors(raw_payload: dict[str, Any], expected_week: Any) -> list[str]: |
| 630 | if not isinstance(expected_week, str): |
| 631 | return [] |
| 632 | timestamp = raw_payload.get("generated_at") or raw_payload.get("crawled_at") |
| 633 | if not isinstance(timestamp, str) or not timestamp.strip(): |
| 634 | return [] |
| 635 | try: |
| 636 | parsed = parse_datetime(timestamp) |
| 637 | except (TypeError, ValueError) as exc: |
| 638 | return [f"raw evidence timestamp is invalid: {exc}"] |
| 639 | artifact_week = week_slug(parsed) |
| 640 | if artifact_week != expected_week: |
| 641 | return [ |
| 642 | f"raw evidence timestamp week mismatch: expected {expected_week}, found {artifact_week}." |
| 643 | ] |
| 644 | return [] |
| 645 | |
| 646 | |
| 647 | def evidence_citation_errors(body: str, raw_payload: dict[str, Any]) -> list[str]: |
| 648 | errors: list[str] = [] |
| 649 | repos = raw_repo_names(raw_payload) |
| 650 | linked_repos = set(REPO_LINK_PATTERN.findall(body)) |
| 651 | if repos and not linked_repos.intersection(repos): |
| 652 | errors.append( |
| 653 | "evidence citations must include at least one repository link from the raw payload." |
| 654 | ) |
| 655 | unresolved_links = sorted(linked_repos - repos) if repos else [] |
| 656 | if unresolved_links: |
| 657 | preview = ", ".join(unresolved_links[:10]) |
| 658 | suffix = f" (+{len(unresolved_links) - 10} more)" if len(unresolved_links) > 10 else "" |
| 659 | errors.append( |
| 660 | f"repository links must resolve to the current raw evidence inventory: {preview}{suffix}." |
| 661 | ) |
| 662 | if repos and "## Key References" in body: |
| 663 | notable = section_text(body, "## Key References") |
| 664 | notable_links = set(REPO_LINK_PATTERN.findall(notable)) |
| 665 | if not notable_links.intersection(repos): |
| 666 | errors.append("Key References must cite at least one raw-payload repository link.") |
| 667 | errors.extend(raw_artifact_week_errors(raw_payload, raw_payload.get("week"))) |
| 668 | return errors |
| 669 | |
| 670 | |
| 671 | def editorial_quality_errors(body: str) -> list[str]: |
| 672 | errors: list[str] = [] |
| 673 | prose = "\n".join(line for line in body.splitlines() if not line.lstrip().startswith("#")) |
| 674 | lower_body = prose.lower() |
| 675 | terms_found = { |
| 676 | term for term in EDITORIAL_TERMS if re.search(rf"\b{re.escape(term)}\b", lower_body) |
| 677 | } |
| 678 | if len(terms_found) < 3: |
| 679 | errors.append( |
| 680 | "editorial analysis must use trend/evidence judgment language, not generic summary prose." |
| 681 | ) |
| 682 | for heading, minimum in SECTION_MIN_WORDS.items(): |
| 683 | text = section_text(body, heading) |
| 684 | if not text: |
| 685 | continue |
| 686 | count = len(WORD_PATTERN.findall(text)) |
| 687 | if count < minimum: |
| 688 | errors.append( |
| 689 | f"{heading} section is too thin for publish-quality analysis; found {count} words, expected at least {minimum}." |
| 690 | ) |
| 691 | for heading in ("## This Week's Trends", "## Signal & Noise", "## Blind Spots"): |
| 692 | text = section_text(body, heading) |
| 693 | linked_repos = REPO_LINK_PATTERN.findall(text) |
| 694 | section_terms = { |
| 695 | term for term in EDITORIAL_TERMS if re.search(rf"\b{re.escape(term)}\b", text.lower()) |
| 696 | } |
| 697 | has_reasoning_or_evidence = ( |
| 698 | EXPLANATORY_PATTERN.search(text) or linked_repos or len(section_terms) >= 2 |
| 699 | ) |
| 700 | if text and not has_reasoning_or_evidence: |
| 701 | errors.append(f"{heading} must explain why the pattern matters, not only name it.") |
| 702 | return errors |
| 703 | |
| 704 | |
| 705 | def contradiction_errors(body: str) -> list[str]: |
| 706 | errors: list[str] = [] |
| 707 | for negative_pattern, positive_pattern, message in CONTRADICTION_PATTERNS: |
| 708 | for match in negative_pattern.finditer(body): |
| 709 | window_start = max(0, match.start() - 300) |
| 710 | window_end = min(len(body), match.end() + 300) |
| 711 | window = body[window_start:window_end] |
| 712 | if positive_pattern.search(window.replace(match.group(0), "", 1)): |
| 713 | errors.append(f"contradictory claim: {message}") |
| 714 | break |
| 715 | return errors |
| 716 | |
| 717 | |
| 718 | def press_context_is_populated( |
| 719 | press_context_path: Path | None, token_estimate: int | None = None |
| 720 | ) -> bool: |
| 721 | """Return True when the week has real press context to write from. |
| 722 | |
| 723 | render_press_context.py emits a non-empty "No press data available for this week." |
| 724 | sentinel when press is absent, so a bare size/non-empty check is insufficient: the |
| 725 | sentinel is treated as an empty press context. |
| 726 | |
| 727 | When a press_context_path is provided its content is authoritative: sentinel |
| 728 | content wins over a positive token_estimate, so a genuinely press-less week is |
| 729 | classified as *not* populated even if token_estimate > 0. The token_estimate |
| 730 | fallback only applies when no usable path is provided (path is None, missing, |
| 731 | empty, or unreadable). |
| 732 | """ |
| 733 | if press_context_path is not None: |
| 734 | try: |
| 735 | if press_context_path.exists() and press_context_path.stat().st_size > 0: |
| 736 | content = press_context_path.read_text(encoding="utf-8").strip() |
| 737 | if content: |
| 738 | return not NO_PRESS_SENTINEL_MARKER.search(content) |
| 739 | except OSError: |
| 740 | pass |
| 741 | return token_estimate is not None and token_estimate > 0 |
| 742 | |
| 743 | |
| 744 | def stale_press_claim_errors(body: str, *, press_context_available: bool) -> list[str]: |
| 745 | """Fail analyses that claim "no press data" while press context is populated. |
| 746 | |
| 747 | This is defense-in-depth against the 2026-W30 regression where a populated |
| 748 | press-context.md was silently dropped and the body shipped |
| 749 | "No industry press data was available for this week's analysis.". |
| 750 | """ |
| 751 | if not press_context_available: |
| 752 | return [] |
| 753 | errors: list[str] = [] |
| 754 | for pattern, message in STALE_PRESS_CLAIM_PATTERNS: |
| 755 | if pattern.search(body): |
| 756 | errors.append(f"stale press claim: {message}") |
| 757 | return errors |
| 758 | |
| 759 | |
| 760 | def ai_provenance_errors(source: str, model: str) -> list[str]: |
| 761 | errors: list[str] = [] |
| 762 | normalized_source = source.strip() |
| 763 | normalized_model = model.strip() |
| 764 | if normalized_source not in PUBLISHABLE_AI_SOURCES: |
| 765 | errors.append(f"AI provenance source is not publishable: {normalized_source or 'unknown'}.") |
| 766 | if normalized_model.lower() in UNPUBLISHABLE_MODEL_VALUES: |
| 767 | errors.append(f"AI provenance model is not publishable: {normalized_model or 'unknown'}.") |
| 768 | return errors |
| 769 | |
| 770 | |
| 771 | def categorize_gate_error(error: str) -> str: |
| 772 | if error.startswith("AI provenance"): |
| 773 | return "ai_provenance" |
| 774 | if error.startswith( |
| 775 | ("evidence citations", "Key References", "raw evidence", "repository links") |
| 776 | ): |
| 777 | return "evidence_citation" |
| 778 | if ( |
| 779 | error.startswith(("editorial analysis", "contradictory claim", "stale press claim")) |
| 780 | or "section is too thin" in error |
| 781 | or "must explain why" in error |
| 782 | ): |
| 783 | return "editorial_quality" |
| 784 | if "quality_score" in error or "generic week/year" in error or "placeholder" in error: |
| 785 | return "editorial_quality" |
| 786 | return "structural_schema" |
| 787 | |
| 788 | |
| 789 | def build_gate_results(errors: list[str]) -> dict[str, dict[str, Any]]: |
| 790 | gates = { |
| 791 | "structural_schema": {"passed": True, "errors": []}, |
| 792 | "ai_provenance": {"passed": True, "errors": []}, |
| 793 | "evidence_citation": {"passed": True, "errors": []}, |
| 794 | "editorial_quality": {"passed": True, "errors": []}, |
| 795 | } |
| 796 | for error in errors: |
| 797 | category = categorize_gate_error(error) |
| 798 | gates[category]["passed"] = False |
| 799 | gates[category]["errors"].append(error) |
| 800 | return gates |
| 801 | |
| 802 | |
| 803 | def validate_publish_quality( |
| 804 | text: str, |
| 805 | raw_payload: dict[str, Any], |
| 806 | *, |
| 807 | source: str, |
| 808 | model: str, |
| 809 | press_context_available: bool = False, |
| 810 | ) -> tuple[list[str], dict[str, dict[str, Any]]]: |
| 811 | try: |
| 812 | _, body = extract_frontmatter(text) |
| 813 | except ValueError: |
| 814 | body = "" |
| 815 | errors: list[str] = [] |
| 816 | errors.extend(ai_provenance_errors(source, model)) |
| 817 | if body: |
| 818 | errors.extend(evidence_citation_errors(body, raw_payload)) |
| 819 | errors.extend(editorial_quality_errors(body)) |
| 820 | errors.extend(contradiction_errors(body)) |
| 821 | errors.extend( |
| 822 | stale_press_claim_errors(body, press_context_available=press_context_available) |
| 823 | ) |
| 824 | return errors, build_gate_results(errors) |
| 825 | |
| 826 | |
| 827 | def validate_analysis( |
| 828 | text: str, raw_payload: dict[str, Any], current_datetime: str |
| 829 | ) -> tuple[list[str], int]: |
| 830 | errors: list[str] = [] |
| 831 | try: |
| 832 | frontmatter, body = extract_frontmatter(text) |
| 833 | except ValueError as exc: |
| 834 | return [str(exc)], 0 |
| 835 | |
| 836 | missing_fields = [field for field in REQUIRED_FIELDS if field not in frontmatter] |
| 837 | if missing_fields: |
| 838 | errors.append(f"Missing frontmatter fields: {', '.join(missing_fields)}") |
| 839 | |
| 840 | extra_fields = sorted(set(frontmatter) - set(REQUIRED_FIELDS) - set(OPTIONAL_FIELDS)) |
| 841 | if extra_fields: |
| 842 | errors.append(f"Unexpected frontmatter fields: {', '.join(extra_fields)}") |
| 843 | |
| 844 | validate_string_field(frontmatter, "title", errors) |
| 845 | title = frontmatter.get("title") |
| 846 | if isinstance(title, str) and title.strip(): |
| 847 | if any(pattern.fullmatch(title.strip()) for pattern in GENERIC_TITLE_PATTERNS): |
| 848 | errors.append("title must not use a generic week/year placeholder format.") |
| 849 | validate_string_field(frontmatter, "week", errors) |
| 850 | validate_string_field(frontmatter, "top_repo", errors) |
| 851 | validate_string_field(frontmatter, "summary", errors) |
| 852 | validate_string_list(frontmatter, "tags", errors, minimum=3, maximum=8) |
| 853 | validate_string_list(frontmatter, "categories", errors, includes="weekly") |
| 854 | validate_integer_field(frontmatter, "year", errors) |
| 855 | validate_integer_field(frontmatter, "repos_featured", errors) |
| 856 | validate_integer_field(frontmatter, "stars_tracked", errors) |
| 857 | validate_integer_field(frontmatter, "quality_score", errors) |
| 858 | validate_predictions(frontmatter, errors) |
| 859 | |
| 860 | quality_score = frontmatter.get("quality_score") |
| 861 | if isinstance(quality_score, int) and quality_score < 60: |
| 862 | errors.append("quality_score must be at least 60.") |
| 863 | |
| 864 | expected_week = raw_payload.get("week") |
| 865 | week_match = WEEK_PATTERN.fullmatch(expected_week) if isinstance(expected_week, str) else None |
| 866 | expected_year = int(week_match.group("year")) if week_match else None |
| 867 | if frontmatter.get("week") != expected_week: |
| 868 | errors.append(f"week must match raw payload week {expected_week!r}.") |
| 869 | if expected_year is not None and frontmatter.get("year") != expected_year: |
| 870 | errors.append(f"year must match the raw payload week year ({expected_year}).") |
| 871 | |
| 872 | date_value = frontmatter.get("date") |
| 873 | if date_value is None: |
| 874 | pass |
| 875 | else: |
| 876 | try: |
| 877 | analysis_datetime = parse_datetime(date_value) |
| 878 | run_datetime = parse_datetime(current_datetime) |
| 879 | except (TypeError, ValueError) as exc: |
| 880 | errors.append(f"date must be a valid ISO 8601 timestamp: {exc}") |
| 881 | else: |
| 882 | if analysis_datetime.astimezone(UTC) != run_datetime.astimezone(UTC): |
| 883 | errors.append("date must match the current run timestamp.") |
| 884 | if isinstance(expected_week, str) and week_slug(analysis_datetime) != expected_week: |
| 885 | errors.append(f"date must fall within raw payload week {expected_week}.") |
| 886 | |
| 887 | top_repo = frontmatter.get("top_repo") |
| 888 | if isinstance(top_repo, str) and top_repo and not TOP_REPO_PATTERN.fullmatch(top_repo): |
| 889 | errors.append("top_repo must use owner/repo format.") |
| 890 | |
| 891 | missing_headings = find_missing_headings(body) |
| 892 | if missing_headings: |
| 893 | for heading in missing_headings: |
| 894 | errors.append(f"Missing required section heading: {heading}") |
| 895 | |
| 896 | word_count = len(WORD_PATTERN.findall(body)) |
| 897 | if word_count < 200: |
| 898 | errors.append(f"Analysis body must be at least 200 words; found {word_count}.") |
| 899 | |
| 900 | lower_body = body.lower() |
| 901 | for marker in RAW_MARKERS: |
| 902 | if marker in lower_body: |
| 903 | errors.append(f"Analysis body contains prohibited marker: {marker}") |
| 904 | for pattern, description in PLACEHOLDER_PATTERNS: |
| 905 | if pattern.search(body): |
| 906 | errors.append(f"Analysis body contains prohibited placeholder marker: {description}") |
| 907 | |
| 908 | return errors, word_count |
| 909 | |
| 910 | |
| 911 | def fail(errors: list[str], summary_path: str | None) -> None: |
| 912 | if summary_path: |
| 913 | with open(summary_path, "a", encoding="utf-8") as handle: |
| 914 | handle.write("## Analysis quality gate failed\n") |
| 915 | for error in errors: |
| 916 | handle.write(f"- {error}\n") |
| 917 | for error in errors: |
| 918 | print(error, file=sys.stderr) |
| 919 | raise SystemExit(1) |
| 920 | |
| 921 | |
| 922 | def report_success(path: Path, source: str, word_count: int, summary_path: str | None) -> None: |
| 923 | message = f"✅ Analysis quality gate passed for {path.name} via {source} ({word_count} words)." |
| 924 | print(message) |
| 925 | if summary_path: |
| 926 | with open(summary_path, "a", encoding="utf-8") as handle: |
| 927 | handle.write("## Analysis quality gate passed\n") |
| 928 | handle.write(f"- File: `{path}`\n") |
| 929 | handle.write(f"- Source: `{source}`\n") |
| 930 | handle.write(f"- Word count: `{word_count}`\n") |
| 931 | |
| 932 | |
| 933 | def write_gate_report( |
| 934 | path: Path | None, |
| 935 | *, |
| 936 | analysis_file: Path, |
| 937 | source: str, |
| 938 | model: str, |
| 939 | errors_before: list[str], |
| 940 | errors_after: list[str], |
| 941 | repair_actions: list[str], |
| 942 | word_count: int, |
| 943 | gate_results: dict[str, dict[str, Any]], |
| 944 | quality_breakdown: dict[str, Any] | None = None, |
| 945 | ) -> None: |
| 946 | if path is None: |
| 947 | return |
| 948 | path.parent.mkdir(parents=True, exist_ok=True) |
| 949 | payload = { |
| 950 | "analysis_file": analysis_file.as_posix(), |
| 951 | "source": source, |
| 952 | "model": model, |
| 953 | "passed": not errors_after, |
| 954 | "word_count": word_count, |
| 955 | "quality_breakdown": quality_breakdown, |
| 956 | "gates": gate_results, |
| 957 | "failure_summary": build_failure_summary(errors_after, gate_results), |
| 958 | "errors_before_repair": errors_before, |
| 959 | "repair_actions": repair_actions, |
| 960 | "errors_after_repair": errors_after, |
| 961 | "failure_class": classify_gate_errors(errors_after), |
| 962 | } |
| 963 | path.write_text(json.dumps(payload, indent=2, sort_keys=True) + "\n", encoding="utf-8") |
| 964 | |
| 965 | |
| 966 | def build_failure_summary( |
| 967 | errors: list[str], gate_results: dict[str, dict[str, Any]] |
| 968 | ) -> dict[str, Any]: |
| 969 | categories = sorted( |
| 970 | category for category, result in gate_results.items() if not result.get("passed") |
| 971 | ) |
| 972 | return { |
| 973 | "failure_class": classify_gate_errors(errors), |
| 974 | "failure_categories": categories, |
| 975 | "error_count": len(errors), |
| 976 | "retryable": bool(errors) |
| 977 | and not any(category == "ai_provenance" for category in categories), |
| 978 | } |
| 979 | |
| 980 | |
| 981 | def classify_gate_errors(errors: list[str]) -> str: |
| 982 | if not errors: |
| 983 | return "passed" |
| 984 | categories = {categorize_gate_error(error) for error in errors} |
| 985 | if len(categories) == 1: |
| 986 | return next(iter(categories)) |
| 987 | if all( |
| 988 | error.startswith( |
| 989 | ("date must", "week must", "year must", "repos_featured must", "stars_tracked must") |
| 990 | ) |
| 991 | or ".claim_type must" in error |
| 992 | for error in errors |
| 993 | ): |
| 994 | return "metadata_schema" |
| 995 | if any( |
| 996 | error.startswith("Missing required section heading") or "body" in error for error in errors |
| 997 | ): |
| 998 | return "content_structure" |
| 999 | return "quality_gate" |
| 1000 | |
| 1001 | |
| 1002 | def gate_report_fingerprint(path: Path) -> str: |
| 1003 | try: |
| 1004 | report = json.loads(path.read_text(encoding="utf-8")) |
| 1005 | except (OSError, json.JSONDecodeError): |
| 1006 | return "" |
| 1007 | if not isinstance(report, dict): |
| 1008 | return "" |
| 1009 | errors = report.get("errors_after_repair") or report.get("errors_before_repair") or [] |
| 1010 | if not isinstance(errors, list): |
| 1011 | return "" |
| 1012 | return hashlib.sha256(json.dumps(errors, sort_keys=True).encode("utf-8")).hexdigest() |
| 1013 | |
| 1014 | |
| 1015 | def main(argv: list[str] | None = None) -> int: |
| 1016 | args = parse_args(argv) |
| 1017 | summary_path = os.environ.get("GITHUB_STEP_SUMMARY") |
| 1018 | |
| 1019 | if not args.analysis_file.exists(): |
| 1020 | fail([f"Missing analysis output: {args.analysis_file}"], summary_path) |
| 1021 | |
| 1022 | text = args.analysis_file.read_text(encoding="utf-8") |
| 1023 | raw_payload = load_json(args.raw_json) |
| 1024 | press_context_available = press_context_is_populated( |
| 1025 | args.press_context_path, args.press_token_estimate |
| 1026 | ) |
| 1027 | errors_before, word_count = validate_analysis(text, raw_payload, args.current_datetime) |
| 1028 | publish_errors_before, _ = validate_publish_quality( |
| 1029 | text, |
| 1030 | raw_payload, |
| 1031 | source=args.source, |
| 1032 | model=args.model, |
| 1033 | press_context_available=press_context_available, |
| 1034 | ) |
| 1035 | combined_errors_before = errors_before + [ |
| 1036 | error for error in publish_errors_before if error not in errors_before |
| 1037 | ] |
| 1038 | errors = errors_before |
| 1039 | repair_actions: list[str] = [] |
| 1040 | if errors and args.repair_safe: |
| 1041 | try: |
| 1042 | repaired_text, repair_actions = repair_analysis( |
| 1043 | text, raw_payload, args.current_datetime |
| 1044 | ) |
| 1045 | except Exception as exc: # noqa: BLE001 - repair is best-effort; validation/reporting must continue. |
| 1046 | repair_actions = [f"repair skipped: {exc}"] |
| 1047 | else: |
| 1048 | if repair_actions and repaired_text != text: |
| 1049 | args.analysis_file.write_text(repaired_text, encoding="utf-8") |
| 1050 | text = repaired_text |
| 1051 | errors, word_count = validate_analysis(text, raw_payload, args.current_datetime) |
| 1052 | print( |
| 1053 | f"::notice::Analysis gate applied safe repairs: {', '.join(repair_actions)}", |
| 1054 | file=sys.stderr, |
| 1055 | ) |
| 1056 | objective_score: int | None = None |
| 1057 | quality_breakdown: dict | None = None |
| 1058 | try: |
| 1059 | objective_score, quality_breakdown = compute_objective_quality( |
| 1060 | text, raw_payload, press_context_available |
| 1061 | ) |
| 1062 | rewritten = set_frontmatter_quality_score(text, objective_score) |
| 1063 | except ValueError: |
| 1064 | # Missing/invalid frontmatter is already reported by validate_analysis(); let the |
| 1065 | # gate fail cleanly with those errors instead of raising an uncaught exception. |
| 1066 | rewritten = text |
| 1067 | if rewritten != text: |
| 1068 | args.analysis_file.write_text(rewritten, encoding="utf-8") |
| 1069 | text = rewritten |
| 1070 | errors, word_count = validate_analysis(text, raw_payload, args.current_datetime) |
| 1071 | publish_errors, _ = validate_publish_quality( |
| 1072 | text, |
| 1073 | raw_payload, |
| 1074 | source=args.source, |
| 1075 | model=args.model, |
| 1076 | press_context_available=press_context_available, |
| 1077 | ) |
| 1078 | errors = errors + [error for error in publish_errors if error not in errors] |
| 1079 | gate_results = build_gate_results(errors) |
| 1080 | write_gate_report( |
| 1081 | args.report_json, |
| 1082 | analysis_file=args.analysis_file, |
| 1083 | source=args.source, |
| 1084 | model=args.model, |
| 1085 | errors_before=combined_errors_before, |
| 1086 | errors_after=errors, |
| 1087 | repair_actions=repair_actions, |
| 1088 | word_count=word_count, |
| 1089 | gate_results=gate_results, |
| 1090 | quality_breakdown=quality_breakdown, |
| 1091 | ) |
| 1092 | if errors: |
| 1093 | fail(errors, summary_path) |
| 1094 | |
| 1095 | report_success(args.analysis_file, args.source, word_count, summary_path) |
| 1096 | return 0 |
| 1097 | |
| 1098 | |
| 1099 | if __name__ == "__main__": |
| 1100 | raise SystemExit(main()) |