SlabUploader/tools/check_requirements.py

#!/usr/bin/env python3
"""Enforce docs/STRUCTURE.md.

Run from the repo root:

    python tools/check_requirements.py

Exit 0 = the docs structure is consistent. Exit 1 = it is not.
Checks are listed in docs/STRUCTURE.md.
"""
from __future__ import annotations

import re
import sys
from pathlib import Path

ROOT = Path(__file__).resolve().parent.parent
DOCS = ROOT / "docs"
STRUCTURE = DOCS / "STRUCTURE.md"
REQUIREMENTS = DOCS / "REQUIREMENTS.md"
MILESTONES = DOCS / "MILESTONES.md"

EXEMPT_PREFIXES = (("docs", "archive"), ("docs", "_templates"))

ROLES = "index|intent|meta|requirements|scope|capability|contract|practice"
REGISTRY_ROW = re.compile(rf"^\|\s*`(docs/[^`]+\.md)`\s*\|\s*({ROLES})\s*\|")
NAMESPACE_ROW = re.compile(r"^\|\s*`([A-Z]+)`\s*\|")
REQUIREMENT_ROW = re.compile(
    r"^\|\s*([A-Z]+)-(\d+)\s*\|\s*(.+?)\s*\|\s*(.+?)\s*\|\s*(.+?)\s*\|\s*$"
)
MILESTONE_IDS = re.compile(r"^-\s*IDs:\s*(.*)$")
ID_TOKEN = re.compile(r"^[A-Z]+-\d+$")

SCOPES = {"shipped", "M3", "later"}
PLACEHOLDER = {"", "—", "-", "*(none yet)*"}


def main() -> int:
    errors: list[str] = []

    if not STRUCTURE.exists():
        print("FAIL check_requirements: docs/STRUCTURE.md is missing")
        return 1

    structure = STRUCTURE.read_text(encoding="utf-8")

    registered: dict[str, str] = {}
    for line in structure.splitlines():
        match = REGISTRY_ROW.match(line)
        if match:
            registered[match.group(1)] = match.group(2)

    namespaces = {
        match.group(1)
        for line in structure.splitlines()
        if (match := NAMESPACE_ROW.match(line))
    }
    if not namespaces:
        errors.append("STRUCTURE.md declares no ID namespaces")

    # 1. every non-exempt doc is registered
    for path in sorted(DOCS.rglob("*.md")):
        rel = path.relative_to(ROOT).as_posix()
        if path.relative_to(ROOT).parts[:2] in EXEMPT_PREFIXES:
            continue
        if rel not in registered:
            errors.append(f"unregistered doc: {rel}")

    # 2. every registered doc exists
    for rel in registered:
        if not (ROOT / rel).exists():
            errors.append(f"registered doc missing on disk: {rel}")

    # 3-5. requirement rows
    ids: dict[str, str] = {}
    if not REQUIREMENTS.exists():
        errors.append("docs/REQUIREMENTS.md is missing")
    else:
        for number, line in enumerate(
            REQUIREMENTS.read_text(encoding="utf-8").splitlines(), 1
        ):
            match = REQUIREMENT_ROW.match(line)
            if not match:
                continue
            prefix, digit, _statement, scope, test = match.groups()
            rid = f"{prefix}-{digit}"
            loc = f"docs/REQUIREMENTS.md:{number}"
            if prefix not in namespaces:
                errors.append(f"{loc} unknown namespace '{prefix}'")
            if rid in ids:
                errors.append(f"{loc} duplicate id {rid} (first seen at {ids[rid]})")
            else:
                ids[rid] = loc
            if scope not in SCOPES:
                errors.append(f"{loc} {rid} invalid scope '{scope}'")
            if scope in {"shipped", "M3"}:
                if test.strip() in PLACEHOLDER:
                    errors.append(f"{loc} {rid} scope {scope} needs a Test path")
                elif not (ROOT / test.strip()).exists():
                    errors.append(f"{loc} {rid} test path not found: {test.strip()}")

    # 6. requirement rows appear only in REQUIREMENTS.md
    for path in sorted(DOCS.rglob("*.md")):
        if path.relative_to(ROOT).parts[:2] in EXEMPT_PREFIXES or path in {
            REQUIREMENTS,
            STRUCTURE,
        }:
            continue
        rel = path.relative_to(ROOT).as_posix()
        for number, line in enumerate(
            path.read_text(encoding="utf-8").splitlines(), 1
        ):
            if REQUIREMENT_ROW.match(line):
                errors.append(f"{rel}:{number} requirement row outside REQUIREMENTS.md")

    # 7. milestone IDs resolve
    if not MILESTONES.exists():
        errors.append("docs/MILESTONES.md is missing")
    else:
        for number, line in enumerate(
            MILESTONES.read_text(encoding="utf-8").splitlines(), 1
        ):
            match = MILESTONE_IDS.match(line)
            if not match:
                continue
            for token in match.group(1).split(","):
                token = token.strip().strip("`").strip()
                if token in PLACEHOLDER:
                    continue
                if not ID_TOKEN.match(token):
                    errors.append(f"MILESTONES.md:{number} not an ID: '{token}'")
                elif token not in ids:
                    errors.append(f"MILESTONES.md:{number} unknown ID {token}")

    if errors:
        print(f"FAIL check_requirements: {len(errors)} problem(s)")
        for error in errors:
            print(f"  - {error}")
        return 1

    print(
        f"PASS check_requirements: {len(registered)} docs, "
        f"{len(namespaces)} namespaces, {len(ids)} requirements"
    )
    return 0


if __name__ == "__main__":
    sys.exit(main())