cy0307/awesome-loop-engineering

1#!/usr/bin/env python32"""Check cross-surface counts, release metadata, and generated discovery files."""3 4from __future__ import annotations5 6import csv7import json8import re9import sys10from pathlib import Path11 12from build_hf_card import project_context, render_card13 14 15ROOT = Path(__file__).resolve().parents[1]16REQUIRED_RESOURCE_FIELDS = {17    "row_id",18    "title",19    "url",20    "canonical_url",21    "resource_type",22    "annotation",23    "key_contribution",24    "novelty",25    "impact",26    "signal",27    "signal_strength",28    "collection",29    "user_goal",30    "lifecycle_stages",31    "audience",32    "loop_layer",33    "scope_fit",34    "evidence_class",35    "evidence_tier",36    "source_status",37    "metadata_source",38    "audited_at",39}40ALLOWED_EVIDENCE_TIERS = {"A", "B", "C", "D"}41ALLOWED_SIGNAL_STRENGTHS = {"high", "medium", "contextual", "unverified"}42ALLOWED_SOURCE_STATUSES = {"ok", "local_ok", "restricted", "broken", "unreachable", "local_missing"}43ALLOWED_LOOP_LAYERS = {"model", "agent", "harness", "workflow", "operations", "evaluation", "cross-layer"}44ALLOWED_SCOPE_FITS = {"direct", "enabling", "adjacent"}45 46 47def require(path: Path, snippets: list[str], failures: list[str]) -> None:48    text = path.read_text(encoding="utf-8")49    for snippet in snippets:50        if snippet not in text:51            failures.append(f"{path.relative_to(ROOT)}: missing {snippet!r}")52 53 54def validate_resource_rows(rows: list[dict[str, str]], failures: list[str]) -> None:55    if not rows:56        failures.append("data/resources.csv: no resource rows")57        return58 59    missing_columns = REQUIRED_RESOURCE_FIELDS - set(rows[0])60    if missing_columns:61        failures.append(f"data/resources.csv: missing columns {sorted(missing_columns)}")62 63    seen_urls: set[str] = set()64    seen_ids: set[str] = set()65    for index, row in enumerate(rows, 1):66        row_label = row.get("row_id") or f"row {index}"67        missing = sorted(field for field in REQUIRED_RESOURCE_FIELDS if not row.get(field, "").strip())68        if missing:69            failures.append(f"data/resources.csv: {row_label} missing required values {missing}")70 71        normalized_url = row.get("url", "").strip().lower().rstrip("/")72        if normalized_url in seen_urls:73            failures.append(f"data/resources.csv: duplicate URL at {row_label}: {row.get('url', '')}")74        seen_urls.add(normalized_url)75 76        row_id = row.get("row_id", "")77        if row_id in seen_ids:78            failures.append(f"data/resources.csv: duplicate row_id {row_id}")79        seen_ids.add(row_id)80        expected_id = f"ale-{index:04d}"81        if row_id != expected_id:82            failures.append(f"data/resources.csv: expected {expected_id}, found {row_id or '<blank>'}")83 84        if row.get("evidence_tier") not in ALLOWED_EVIDENCE_TIERS:85            failures.append(f"data/resources.csv: {row_label} has invalid evidence_tier {row.get('evidence_tier')!r}")86        if row.get("signal_strength") not in ALLOWED_SIGNAL_STRENGTHS:87            failures.append(f"data/resources.csv: {row_label} has invalid signal_strength {row.get('signal_strength')!r}")88        if row.get("source_status") not in ALLOWED_SOURCE_STATUSES:89            failures.append(f"data/resources.csv: {row_label} has invalid source_status {row.get('source_status')!r}")90        if row.get("loop_layer") not in ALLOWED_LOOP_LAYERS:91            failures.append(f"data/resources.csv: {row_label} has invalid loop_layer {row.get('loop_layer')!r}")92        if row.get("scope_fit") not in ALLOWED_SCOPE_FITS:93            failures.append(f"data/resources.csv: {row_label} has invalid scope_fit {row.get('scope_fit')!r}")94 95        year = row.get("publication_year", "")96        if year and not re.fullmatch(r"(?:19|20)\d{2}", year):97            failures.append(f"data/resources.csv: {row_label} has invalid publication_year {year!r}")98        if row.get("resource_type") == "Paper":99            for field in ("authors", "publication_year"):100                if not row.get(field, "").strip():101                    failures.append(f"data/resources.csv: {row_label} paper is missing {field}")102        if "arxiv.org" in row.get("url", "") and not row.get("arxiv_id", "").strip():103            failures.append(f"data/resources.csv: {row_label} arXiv work is missing arxiv_id")104 105 106def main() -> int:107    context = project_context()108    count = context["RESOURCE_COUNT"]109    version = context["VERSION"]110    source_summary = (111        f"{context['REACHABLE_COUNT']} public links opened successfully, "112        f"{context['RESTRICTED_COUNT']} required access, "113        f"{context['LOCAL_COUNT']} pointed to files in this repository"114    )115    failures: list[str] = []116 117    require(118        ROOT / "README.md",119        [120            f"resources-{count}-",121            f"patterns-{context['PATTERN_COUNT']}-",122            f"contracts-{context['CONTRACT_COUNT']}-",123            f"starters-{context['RUNNABLE_COUNT']}-",124            f"**{count} resources**",125            source_summary,126            f"**{context['RUNNABLE_COUNT']} runtime starters**",127            f"{context['EXECUTABLE_COUNT']} dependency-light executables plus {context['RUNTIME_TEMPLATE_COUNT']} copy/paste runtime templates",128            f"{context['LANGUAGE_COUNT']} language entry points",129        ],130        failures,131    )132    require(133        ROOT / "docs" / "index.html",134        [135            f'content="Explore {count} resources',136            f">{count}</b><span>resources</span>",137            f"Filter {count} resources",138            f"{context['REACHABLE_COUNT']} public links opened successfully, {context['RESTRICTED_COUNT']} required access",139            f'"version": "{version}"',140        ],141        failures,142    )143    require(ROOT / "meta" / "social-preview.html", [f">{count}</b>"], failures)144    distribution_text = (ROOT / "meta" / "DISTRIBUTION.md").read_text(encoding="utf-8")145    share_match = re.search(r"awesome-loop-engineering/(x-v\d+-\d+\.html)", distribution_text)146    if share_match is None:147        failures.append("meta/DISTRIBUTION.md: missing current x share-page link")148    else:149        require(150            ROOT / "docs" / share_match.group(1),151            [f"{count} resources", f"{count} papers"],152            failures,153        )154    require(155        ROOT / "posts" / "launch.md",156        [157            f"# Awesome Loop Engineering v{version}",158            f"{count} resources",159            f"{context['MODEL_COUNT']} model-layer resources",160            f"{context['FIELD_COUNT']}-field CSV",161            f"{context['LANGUAGE_COUNT']} language entry points",162        ],163        failures,164    )165    require(ROOT / "posts" / "launch.zh-CN.md", [f"# Awesome Loop Engineering v{version}", f"{count} 篇论文"], failures)166    require(ROOT / "meta" / "DISTRIBUTION.md", [f"v{version}", f"{count} resources"], failures)167    require(168        ROOT / "design-qa.md",169        [170            f"returns {context['MODEL_COUNT']} of {count} resources ({context['MODEL_PAPER_COUNT']} papers plus Awesome Loop Models)",171            f"{context['REACHABLE_COUNT']} public sources reachable, {context['RESTRICTED_COUNT']} access-restricted, {context['LOCAL_COUNT']} repository-native",172        ],173        failures,174    )175    require(ROOT / "data" / "README.md", [f"Download all {count} resources"], failures)176 177    for translation in sorted(ROOT.glob("README.*.md")):178        require(translation, [count], failures)179 180    with (ROOT / "data" / "resources.csv").open(encoding="utf-8", newline="") as handle:181        resource_rows = list(csv.DictReader(handle))182    csv_count = len(resource_rows)183    if csv_count != int(count):184        failures.append(f"data/resources.csv: expected {count} rows, found {csv_count}")185    validate_resource_rows(resource_rows, failures)186 187    with (ROOT / "data" / "resource_source_audit.csv").open(encoding="utf-8", newline="") as handle:188        audit_rows = list(csv.DictReader(handle))189    if len(audit_rows) != int(count):190        failures.append(f"data/resource_source_audit.csv: expected {count} rows, found {len(audit_rows)}")191    audit_statuses: dict[str, int] = {}192    for row in audit_rows:193        status = row.get("audit_status", "")194        audit_statuses[status] = audit_statuses.get(status, 0) + 1195    expected_statuses = {196        "ok": int(context["REACHABLE_COUNT"]),197        "restricted": int(context["RESTRICTED_COUNT"]),198        "local_ok": int(context["LOCAL_COUNT"]),199    }200    if audit_statuses != expected_statuses:201        failures.append(202            "data/resource_source_audit.csv: status counts do not match the generated resource exports "203            f"({audit_statuses!r} != {expected_statuses!r})"204        )205 206    site_payload = json.loads((ROOT / "docs" / "assets" / "resources.json").read_text(encoding="utf-8"))207    if site_payload.get("count") != int(count) or len(site_payload.get("resources", [])) != int(count):208        failures.append("docs/assets/resources.json: count or resources array is stale")209 210    _, card_failures = render_card()211    failures.extend(card_failures)212 213    if failures:214        print("Project consistency check failed:", file=sys.stderr)215        for failure in failures:216            print(f"- {failure}", file=sys.stderr)217        return 1218 219    print(220        f"Validated {count} resources, {context['MODEL_COUNT']} model-layer resources, "221        f"{context['PATTERN_COUNT']} patterns, {context['CONTRACT_COUNT']} contracts, "222        f"{context['RUNNABLE_COUNT']} runtime starters, and release v{version}."223    )224    return 0225 226 227if __name__ == "__main__":228    raise SystemExit(main())229