"""Tests for scripts/aggregate_directives.py.

These tests exercise the script via subprocess to verify the contract:
 - reads v1.md files only (never meta.md)
 - emits clean bodies separated by '====' header
 - exits 1 on missing files
 - emits to stdout by default; -o writes to file

The importable aggregate_directives() function tests cover the API surface used
by the MCP server (scripts/mcp_server.py:aggregate_directives tool).
"""
from __future__ import annotations

import subprocess
import sys
from pathlib import Path

import pytest

REPO_ROOT = Path(__file__).resolve().parent.parent
SCRIPT = REPO_ROOT / "scripts" / "aggregate_directives.py"
PRESET = REPO_ROOT / "conductor" / "directives" / "presets" / "current_baseline.md"


def _run(args: list[str], cwd: Path | None = None) -> subprocess.CompletedProcess:
 return subprocess.run(
 [sys.executable, str(SCRIPT), *args],
 capture_output=True,
 text=True,
 encoding="utf-8",
 cwd=cwd or REPO_ROOT,
 )


def test_runs_against_current_baseline() -> None:
 result = _run([str(PRESET)])
 assert result.returncode == 0, "script should succeed; stderr=" + result.stderr
 body = result.stdout
 assert "atomic_per_task_commits" in body
 assert "ban_dict_any" in body
 assert "knowledge_harvest_pattern" in body
 assert body.count("========================================") >= 51 * 2


def test_meta_md_pollution_fix() -> None:
 result = _run([str(PRESET)])
 body = result.stdout
 assert "**Why this iteration:**" not in body
 assert "**Lifted:**" not in body


def test_missing_file_exits_nonzero(tmp_path: Path) -> None:
 bad = tmp_path / "bad_preset.md"
 bad.write_text("- missing_thing: conductor/directives/missing_thing/v1.md\n", encoding="utf-8")
 result = _run([str(bad)])
 assert result.returncode != 0
 assert "missing_thing" in result.stderr


def test_output_flag_writes_to_file(tmp_path: Path) -> None:
 out = tmp_path / "out.md"
 result = _run([str(PRESET), "-o", str(out)])
 assert result.returncode == 0
 assert out.exists()
 content = out.read_text(encoding="utf-8")
 assert "ban_dict_any" in content
 assert content.count("========================================") >= 51 * 2


def test_no_meta_md_reads() -> None:
 result = _run([str(PRESET)])
 body = result.stdout
 assert "Provenance" not in body
 assert "## v1" not in body
 assert "# ban_dict_any" not in body


DIRECTIVES_DIR = REPO_ROOT / "conductor" / "directives"


def _iter_v1_files() -> list[Path]:
 return sorted(DIRECTIVES_DIR.glob("*/v1.md"))


def test_every_v1_has_top_level_heading() -> None:
 v1_files = _iter_v1_files()
 assert len(v1_files) >= 66, "expected at least 66 directive v1.md files; found " + str(len(v1_files))
 missing: list[str] = []
 for path in v1_files:
  first_line = path.read_text(encoding="utf-8").splitlines()[0] if path.read_text(encoding="utf-8") else ""
  if not first_line.startswith("# "):
   missing.append(path.parent.name)
 assert not missing, "v1.md files missing a top-level '# ' rule-statement heading: " + ", ".join(missing)


def test_v1_heading_is_not_meta_provenance_format() -> None:
 for path in _iter_v1_files():
  first_line = path.read_text(encoding="utf-8").splitlines()[0]
  assert not first_line.strip().endswith("- v1"), path.parent.name + " uses the meta.md provenance heading format in v1.md body"


def _import_aggregate_directives_module():
 import importlib.util
 spec = importlib.util.spec_from_file_location("aggregate_directives_under_test", SCRIPT)
 if spec is None or spec.loader is None:
  raise RuntimeError("failed to build import spec for " + str(SCRIPT))
 mod = importlib.util.module_from_spec(spec)
 spec.loader.exec_module(mod)
 return mod


def test_importable_function_returns_clean_body() -> None:
 mod = _import_aggregate_directives_module()
 body = mod.aggregate_directives(str(PRESET))
 assert "atomic_per_task_commits" in body
 assert "ban_dict_any" in body
 assert "knowledge_harvest_pattern" in body
 assert body.count("========================================") >= 51 * 2


def test_importable_function_no_meta_md_pollution() -> None:
 mod = _import_aggregate_directives_module()
 body = mod.aggregate_directives(str(PRESET))
 assert "**Why this iteration:**" not in body
 assert "**Source:**" not in body
 assert "**Lifted:**" not in body
 assert "## v1" not in body
 assert "Provenance" not in body


def test_importable_function_max_chars_truncates_with_suffix() -> None:
 mod = _import_aggregate_directives_module()
 body = mod.aggregate_directives(str(PRESET), max_chars=2000)
 assert len(body) > 2000, "truncated body must include suffix text exceeding the cap"
 assert "truncated" in body.lower()


def test_importable_function_max_chars_zero_is_unlimited() -> None:
 mod = _import_aggregate_directives_module()
 full = mod.aggregate_directives(str(PRESET))
 same = mod.aggregate_directives(str(PRESET), max_chars=0)
 assert same == full


def test_importable_function_max_chars_above_size_is_unlimited() -> None:
 mod = _import_aggregate_directives_module()
 full = mod.aggregate_directives(str(PRESET))
 capped = mod.aggregate_directives(str(PRESET), max_chars=len(full) + 5000)
 assert capped == full


def test_importable_function_missing_preset_raises() -> None:
 mod = _import_aggregate_directives_module()
 bad = REPO_ROOT / "conductor" / "directives" / "presets" / "_nonexistent_preset_xyz.md"
 with pytest.raises(FileNotFoundError) as exc_info:
  mod.aggregate_directives(str(bad))
 assert "preset file not found" in str(exc_info.value)


def test_importable_function_missing_v1_in_preset_raises(tmp_path: Path) -> None:
 mod = _import_aggregate_directives_module()
 bad = tmp_path / "bad_preset.md"
 bad.write_text("- missing_v1_xyz: conductor/directives/missing_v1_xyz/v1.md\n", encoding="utf-8")
 with pytest.raises(FileNotFoundError) as exc_info:
  mod.aggregate_directives(str(bad))
 assert "missing_v1_xyz" in str(exc_info.value)


def test_importable_function_accepts_relative_preset_path() -> None:
 mod = _import_aggregate_directives_module()
 rel = "conductor/directives/presets/current_baseline.md"
 body = mod.aggregate_directives(rel, project_root=str(REPO_ROOT))
 assert "ban_dict_any" in body