#!/usr/bin/env python3
"""Require math evidence (receipt or Lean change) alongside math-track edits.

This script enforces the math-first contract: if a commit or PR touches files
in a math-track surface, at least one evidence file must also be present.

Usage:
    # CI mode: compare against a base ref
    python3 scripts/math-first/require_math_evidence.py --from-git-diff origin/main

    # Pre-commit mode: check staged files
    python3 scripts/math-first/require_math_evidence.py --staged

Exits 0 if:
  - No math-track files are present in the changeset, OR
  - At least one evidence file is also in the changeset

Exits 1 if math-track files are present but no evidence file accompanies them.
"""

from __future__ import annotations

import argparse
import subprocess
import sys
from pathlib import Path

# Math-track surface prefixes: files under these paths require evidence
MATH_TRACK_PREFIXES = (
    "0-Core-Formalism/lean/Semantics/",
    "6-Documentation/docs/distilled/",
    "shared-data/data/stack_solidification/",
)

# Evidence patterns: if any changed file matches, the requirement is satisfied
EVIDENCE_PREFIXES = (
    "shared-data/artifacts/deepseek_review/",
    "0-Core-Formalism/lean/Semantics/",
)
EVIDENCE_FILES = (
    "claims.yaml",
)


def get_changed_files_git_diff(base_ref: str) -> list[str]:
    """Get changed files by diffing against a base ref."""
    result = subprocess.run(
        ["git", "diff", "--name-only", base_ref + "...HEAD"],
        capture_output=True, text=True, check=False,
    )
    if result.returncode != 0:
        # Fall back to two-dot diff
        result = subprocess.run(
            ["git", "diff", "--name-only", base_ref],
            capture_output=True, text=True, check=True,
        )
    return [f for f in result.stdout.strip().split("\n") if f]


def get_changed_files_staged() -> list[str]:
    """Get staged files for pre-commit hook."""
    result = subprocess.run(
        ["git", "diff", "--cached", "--name-only"],
        capture_output=True, text=True, check=True,
    )
    return [f for f in result.stdout.strip().split("\n") if f]


def is_math_track(path: str) -> bool:
    return any(path.startswith(p) for p in MATH_TRACK_PREFIXES)


def is_evidence(path: str) -> bool:
    if any(path.startswith(p) for p in EVIDENCE_PREFIXES):
        return True
    if path in EVIDENCE_FILES:
        return True
    return False


def main() -> int:
    parser = argparse.ArgumentParser(description=__doc__)
    group = parser.add_mutually_exclusive_group(required=True)
    group.add_argument("--from-git-diff", metavar="BASE_REF",
                       help="Compare HEAD against BASE_REF")
    group.add_argument("--staged", action="store_true",
                       help="Check staged files (pre-commit mode)")
    args = parser.parse_args()

    if args.staged:
        changed = get_changed_files_staged()
    else:
        changed = get_changed_files_git_diff(args.from_git_diff)

    if not changed:
        print("No changed files")
        return 0

    math_files = [f for f in changed if is_math_track(f)]
    if not math_files:
        print("No math-track files in changeset — skipping evidence check")
        return 0

    evidence_files = [f for f in changed if is_evidence(f)]
    if evidence_files:
        print(f"OK: {len(math_files)} math-track file(s) with "
              f"{len(evidence_files)} evidence file(s)")
        return 0

    # Math-track files present but no evidence
    print("FAIL: math-track files changed but no evidence accompanies them:")
    for f in math_files:
        print(f"  - {f}")
    print("\nPlease include at least one of:")
    print("  - A DeepSeek review receipt under shared-data/artifacts/deepseek_review/")
    print("  - A Lean source file under 0-Core-Formalism/lean/Semantics/")
    print("  - An update to claims.yaml")
    return 1


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