scripts/build_public_surface_qa.py

#!/usr/bin/env python3
"""Build the public presentation report for repo, website, and HF mirrors."""

from __future__ import annotations

import json
from datetime import datetime, timezone
from pathlib import Path


ROOT = Path(__file__).resolve().parents[1]
HF_ROOT = ROOT.parent / "hf_publish"
OUTPUT_JSON = ROOT / "docs/data/public_surface_qa.json"
OUTPUT_MD = ROOT / "PUBLIC_SURFACE_QA.md"

SURFACES = {
    "github_readme": ROOT / "README.md",
    "website_html": ROOT / "docs/index.html",
    "hf_space_card": HF_ROOT / "space/README.md",
    "hf_artifact_card": HF_ROOT / "artifacts/README.md",
    "hf_model_card": HF_ROOT / "model/README.md",
}

STATUS_REPORTS = {
    "website_integrity": ROOT / "docs/data/website_integrity.json",
    "rendered_site_check": ROOT / "docs/data/rendered_site_check.json",
    "task_surface_integrity": ROOT / "docs/data/task_surface_integrity.json",
    "source_alignment": ROOT / "docs/data/source_alignment_audit.json",
    "scale_up_status": ROOT / "docs/data/scope_claims_audit.json",
    "publication_package": ROOT / "docs/data/publication_audit.json",
    "mirror_parity": ROOT / "docs/data/mirror_parity.json",
    "live_publication": ROOT / "docs/data/live_publication_status.json",
}

DISPLAY_LABELS = {
    "public_presentation_files_exist": "Public files",
    "core_status_reports_pass": "Project reports",
    "website_has_research_seo_metadata": "Website metadata",
    "website_tabs_are_accessible_and_keyboardable": "Keyboard navigation",
    "responsive_navigation_guard_present": "Responsive navigation",
    "public_naming_consistent": "Project naming",
    "public_links_cover_repo_hf_dataset_and_ropedia": "Public links",
    "public_artifact_qa_files_are_exposed": "Artifact links",
    "public_copy_uses_reader_facing_language": "Project language",
}

BANNED_PUBLIC_STRINGS = [
    "audit" + "able",
    "in" + "ternal " + "re" + "view label",
    "pri" + "vate " + "evaluation note",
    "ChatGPT" + "-image",
    "H" + "20",
    "A" + "100",
    "Cur" + "sor",
    "public " + "dashboard and generated figures " + "deliber" + "ately follow",
    "unsupported general-result language",
    "non-public planning note",
    "Public " + "project QA",
    "public-project " + "QA",
    "readiness" + "-only",
    "not a foundation-model result",
    "unsupported " + "interpretations",
    "unsupported " + "conclusions",
    "result-scope guard",
    "Result-scope guard",
    "Publication " + "hyg" + "iene",
    "copy " + "hyg" + "iene",
    "Research progress with pri" + "vate caveats",
    "Research progress with process caveats",
    "re" + "viewer " + "scorecard",
    "block" + "er",
    "re" + "view" + "-only checklist",
    "accelerated relay",
    "selected relay",
    "chunked parallel transfer",
    "batch prefetch",
    "overlapping batch prefetch",
]


def read_text(path: Path) -> str:
    return path.read_text(encoding="utf-8", errors="ignore") if path.exists() else ""


def read_status(path: Path) -> dict:
    if not path.exists():
        return {"exists": False, "status": "missing"}
    try:
        payload = json.loads(path.read_text(encoding="utf-8"))
    except json.JSONDecodeError as exc:
        return {"exists": True, "status": "invalid_json", "error": str(exc)}
    return {
        "exists": True,
        "status": str(payload.get("status", "unknown")),
        "generated_at_utc": payload.get("generated_at_utc"),
    }


def display_path(path: Path) -> str:
    for base in (ROOT, ROOT.parent):
        try:
            return path.relative_to(base).as_posix()
        except ValueError:
            continue
    return path.name


def check(name: str, passed: bool, reason: str, **detail) -> dict:
    return {"name": name, "status": "pass" if passed else "fail", "reason": reason, **detail}


def marker_count(text: str, markers: list[str]) -> dict[str, int]:
    return {marker: text.count(marker) for marker in markers}


def build_report() -> dict:
    texts = {name: read_text(path) for name, path in SURFACES.items()}
    combined_public_text = "\n".join(texts.values())
    website = texts["website_html"]
    all_surface_files_exist = all(path.exists() for path in SURFACES.values())

    status_records = {name: read_status(path) for name, path in STATUS_REPORTS.items()}
    status_failures = {
        name: record
        for name, record in status_records.items()
        if record["status"] != "pass"
    }

    seo_markers = [
        "<title>Ropedia Xperience-10M Task Suite</title>",
        'name="description"',
        'rel="canonical"',
        'property="og:title"',
        'property="og:image"',
        'name="twitter:card"',
        'application/ld+json',
        'rel="manifest"',
        'rel="apple-touch-icon"',
    ]
    accessible_markers = [
        'role="tablist"',
        'role="tab"',
        'role="tabpanel"',
        "aria-selected",
        "aria-controls",
        "moveProjectTabFocus",
        "ArrowRight",
        "Home",
        "End",
    ]
    naming_markers = [
        "Ropedia Xperience-10M Task Suite",
        "Xperience-10M",
        "12-task",
        "Qwen3-Omni",
        "128-episode pilot",
    ]
    hf_link_markers = [
        "https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite",
        "https://huggingface.co/spaces/cy0307/ropedia-xperience-10m-task-suite",
        "https://huggingface.co/datasets/cy0307/ropedia-xperience-10m-task-suite-artifacts",
        "https://huggingface.co/cy0307/ropedia-xperience-10m-task-baselines",
        "https://huggingface.co/datasets/ropedia-ai/xperience-10m",
        "https://ropedia.com/dataset",
    ]
    artifact_markers = [
        "data/project_brief.json",
        "data/website_integrity.json",
        "data/rendered_site_check.json",
        "data/task_surface_integrity.json",
        "data/publication_audit.json",
        "data/mirror_parity.json",
        "data/public_surface_qa.json",
        "data/research_roadmap.json",
    ]

    banned_hits = [
        {"marker": marker, "count": combined_public_text.count(marker)}
        for marker in BANNED_PUBLIC_STRINGS
        if marker in combined_public_text
    ]

    checks = [
        check(
            "public_presentation_files_exist",
            all_surface_files_exist,
            "Repo README, website HTML, and three Hugging Face cards should all be present in the publication workspace.",
            missing=[str(path) for path in SURFACES.values() if not path.exists()],
        ),
        check(
            "core_status_reports_pass",
            not status_failures,
            "The public project surface depends on the existing project reports already passing.",
            reports=status_records,
            failures=status_failures,
        ),
        check(
            "website_has_research_seo_metadata",
            all(marker in website for marker in seo_markers),
            "The website should expose search/social metadata and structured project metadata.",
            marker_counts=marker_count(website, seo_markers),
        ),
        check(
            "website_tabs_are_accessible_and_keyboardable",
            'role="tablist"' in website
            and website.count("data-tab-key=") == 6
            and website.count("data-panel-target=") >= 4
            and website.count('role="tab"') >= website.count("data-tab-key=") + website.count("data-panel-target=")
            and website.count('role="tabpanel"') >= 19
            and "moveProjectTabFocus" in website
            and "initContentTabs" in website
            and "ArrowRight" in website
            and "Home" in website
            and "End" in website,
            "The long research dashboard should be navigable as real tabs, including keyboard support.",
            marker_counts=marker_count(website, accessible_markers),
        ),
        check(
            "responsive_navigation_guard_present",
            "@media (max-width: 1120px)" in website
            and ".nav-links { display: none; }" in website
            and "scroll-margin-top" in website,
            "Tablet/mobile navigation should not overflow and deep links should land below sticky navigation.",
        ),
        check(
            "public_naming_consistent",
            all(marker in combined_public_text for marker in naming_markers),
            "Public copy should consistently present the project as Ropedia Xperience-10M, with the Qwen3-Omni scale-up status.",
            marker_counts=marker_count(combined_public_text, naming_markers),
        ),
        check(
            "public_links_cover_repo_hf_dataset_and_ropedia",
            all(marker in combined_public_text for marker in hf_link_markers),
            "Public cards should link the repo, Space, artifacts, model baselines, upstream dataset, and Ropedia dataset page.",
            marker_counts=marker_count(combined_public_text, hf_link_markers),
        ),
        check(
            "public_artifact_qa_files_are_exposed",
            all(marker in combined_public_text for marker in artifact_markers),
            "Readers should be able to find website reference, release package, mirror, and public presentation files from public copy.",
            marker_counts=marker_count(combined_public_text, artifact_markers),
        ),
        check(
            "public_copy_uses_reader_facing_language",
            not banned_hits,
            "Project language is clear and avoids hardware details or irrelevant implementation details.",
            banned_hits=banned_hits,
        ),
    ]
    status = "pass" if all(item["status"] == "pass" for item in checks) else "fail"
    return {
        "title": "Ropedia Xperience-10M Public Project Surface",
        "status": status,
        "generated_at_utc": datetime.now(timezone.utc).isoformat(timespec="seconds"),
        "scope": "Repo README, GitHub Pages HTML, Hugging Face Space card, artifact dataset card, and model card.",
        "checks": checks,
        "surface_files": {name: display_path(path) for name, path in SURFACES.items()},
        "scope_note": "This report covers the public repo, website, Hugging Face cards, and package contents. Multi-episode model metrics are tracked by the training and evaluation reports.",
    }


def markdown(report: dict) -> str:
    lines = [
        "# Public Project Surface",
        "",
        "This generated report checks whether the public repo, website, and Hugging Face cards read like one cohesive research project.",
        "",
        f"Current status: **{report['status']}**",
        "",
        report["scope_note"],
        "",
        "## Checks",
        "",
        "| Area | Status | What it covers |",
        "| --- | --- | --- |",
    ]
    for item in report["checks"]:
        label = DISPLAY_LABELS.get(item["name"], item["name"].replace("_", " ").title())
        lines.append(f"| {label} | `{item['status']}` | {item['reason']} |")
    lines.extend([
        "",
        "## Scope",
        "",
        "| Surface | File |",
        "| --- | --- |",
    ])
    for name, path in report["surface_files"].items():
        lines.append(f"| {name} | `{path}` |")
    lines.extend([
        "",
        "## Regenerate",
        "",
        "```bash",
        "python scripts/build_public_surface_qa.py",
        "```",
        "",
    ])
    return "\n".join(lines)


def main() -> int:
    report = build_report()
    OUTPUT_JSON.parent.mkdir(parents=True, exist_ok=True)
    OUTPUT_JSON.write_text(json.dumps(report, indent=2) + "\n", encoding="utf-8")
    OUTPUT_MD.write_text(markdown(report), encoding="utf-8")
    print(f"{report['status'].upper()}: wrote {OUTPUT_JSON}")
    print(f"{report['status'].upper()}: wrote {OUTPUT_MD}")
    if report["status"] != "pass":
        for item in report["checks"]:
            if item["status"] != "pass":
                print(f"- {item['name']}: {item['reason']}")
        return 1
    return 0


if __name__ == "__main__":
    raise SystemExit(main())