"""Synthetic checks for source-safe local import discovery."""

from hashlib import sha256

from django.test import override_settings

from ingestion.guided_imports import discover_imports


HEADER = b"Data,Dependencia,Historico,Documento,Tipo,Valor\n"


def test_discovery_returns_only_safe_metadata_for_direct_supported_files(tmp_path):
    bank = HEADER + b"01/02/2026,,SYNTHETIC BANK ROW,,,12.34\n"
    full = b"1 - SYNTHETIC OWNER\n" + b"15.02.2026" + b"SYNTHETIC CARD ROW".ljust(37) + b"BR 12,34 0,00\n"
    nested = tmp_path / "nested"
    nested.mkdir()
    (tmp_path / "bank.csv").write_bytes(bank)
    (tmp_path / "card.txt").write_bytes(full)
    (nested / "hidden.csv").write_bytes(bank)

    with override_settings(IMPORT_ROOT=tmp_path):
        rows = discover_imports()

    assert [(row.name, row.profile, row.sections, row.code) for row in rows] == [
        ("bank.csv", "BANK_CSV_V1", ("1",), None),
        ("card.txt", "OUROCARD_FULL_V1", ("1",), None),
    ]
    assert [row.document_hash for row in rows] == [sha256(bank).hexdigest(), sha256(full).hexdigest()]
    assert all(not hasattr(row, "source_text") for row in rows)


def test_discovery_reports_safe_codes_for_direct_unsupported_or_invalid_files(tmp_path):
    (tmp_path / "unknown.ofx").write_bytes(b"SYNTHETIC OFX")
    (tmp_path / "statement.pdf").write_bytes(b"%PDF-SYNTHETIC")
    (tmp_path / "invalid.txt").write_bytes(b"SYNTHETIC UNRECOGNIZED TEXT")
    (tmp_path / "oversize.csv").write_bytes(b"x" * (10 * 1024 * 1024 + 1))

    with override_settings(IMPORT_ROOT=tmp_path):
        rows = discover_imports()

    assert [(row.name, row.profile, row.sections, row.code) for row in rows] == [
        ("invalid.txt", None, (), "UNRECOGNIZED_LAYOUT"),
        ("oversize.csv", None, (), "INPUT_TOO_LARGE"),
        ("statement.pdf", None, (), "UNSUPPORTED_FILE"),
        ("unknown.ofx", None, (), "UNSUPPORTED_FILE"),
    ]
    assert all(not hasattr(row, "source_text") for row in rows)


def test_discovery_ignores_symlinks_and_non_files(tmp_path):
    bank = HEADER + b"01/02/2026,,SYNTHETIC BANK ROW,,,12.34\n"
    target = tmp_path / "target.csv"
    target.write_bytes(bank)
    (tmp_path / "linked.csv").symlink_to(target)
    (tmp_path / "directory.csv").mkdir()

    with override_settings(IMPORT_ROOT=tmp_path):
        rows = discover_imports()

    assert [row.name for row in rows] == ["target.csv"]