net-kingdom/tests/test_prune_storage_share_dumps.py

"""Synthetic tests for the list-first Storage Share dump pruning (no network, no credentials)."""

import hashlib
import importlib.util
import io
import json
import urllib.error
import urllib.parse
from pathlib import Path

import pytest

ROOT = Path(__file__).resolve().parents[1]
spec = importlib.util.spec_from_file_location(
    "prune", ROOT / "tools/bootstrap-recovery/prune-storage-share-dumps.py"
)
m = importlib.util.module_from_spec(spec)
spec.loader.exec_module(m)
FOLDER = "r01-net-kingdom-pg"


def names(days, prefix="net-kingdom-pg-privacyidea_db"):
    return {f"{prefix}-202610{d:02d}T023000Z.dump.age": 1000 + d for d in days}


class Response(io.BytesIO):
    def __init__(self, status, url, body=b""):
        super().__init__(body)
        self.status, self.url = status, url

    def __enter__(self):
        return self

    def __exit__(self, *a):
        return False


class Share:
    def __init__(self, files, extra=()):
        self.files, self.extra, self.deleted = dict(files), list(extra), []

    def open(self, request, timeout):
        url, method = request.full_url, request.get_method()
        assert request.headers["Authorization"].startswith("Basic ")
        if method == "PROPFIND":
            rows = "".join(
                f"<d:response><d:href>{m.BASE}{FOLDER}/{urllib.parse.quote(n)}</d:href><d:propstat><d:prop>"
                f"<d:resourcetype/><d:getcontentlength>{s}</d:getcontentlength></d:prop>"
                f"<d:status>HTTP/1.1 200 OK</d:status></d:propstat></d:response>"
                for n, s in [*self.files.items(), *self.extra]
            )
            body = (
                f'<d:multistatus xmlns:d="DAV:"><d:response><d:href>{m.BASE}{FOLDER}/</d:href></d:response>'
                f"{rows}</d:multistatus>"
            ).encode()
            return Response(207, url, body)
        name = urllib.parse.unquote(url.rsplit("/", 1)[1])
        if method == "DELETE":
            del self.files[name]
            self.deleted.append(name)
            return Response(204, url)
        if method == "HEAD":
            if name in self.files:
                return Response(200, url)
            raise urllib.error.HTTPError(url, 404, "Not Found", None, None)
        raise AssertionError(method)


def make_plan(tmp_path, share, *extra):
    path = tmp_path / "plan.json"
    assert (
        m.main(
            ["plan", "--folder", FOLDER, "--plan", str(path), *extra],
            client=share,
            secret="pw",
        )
        == 0
    )
    return path, hashlib.sha256(path.read_bytes()).hexdigest()


def test_plan_keeps_newest_daily_and_weekly_per_database(tmp_path, capsys):
    share = Share(
        {**names(range(1, 29)), **names(range(1, 29), "net-kingdom-pg-other_db")}
    )
    path, _ = make_plan(tmp_path, share)
    plan = json.loads(path.read_text())
    for prefix in ("net-kingdom-pg-privacyidea_db", "net-kingdom-pg-other_db"):
        kept = sorted(n for n in plan["keep"] if n.startswith(prefix + "-"))
        assert (
            f"{prefix}-20261028T023000Z.dump.age" in kept
            and f"{prefix}-20261022T023000Z.dump.age" in kept
        )
        assert 7 <= len(kept) <= 11
    assert not share.deleted and set(plan["keep"]) | set(plan["delete"]) == set(
        share.files
    )
    assert "plan sha256" in capsys.readouterr().out


def test_delete_needs_the_plan_hash_and_removes_only_planned_files(tmp_path):
    share = Share(names(range(1, 29)))
    path, digest = make_plan(tmp_path, share)
    plan = json.loads(path.read_text())
    assert (
        m.main(
            [
                "delete",
                "--plan",
                str(path),
                "--confirm",
                "0" * 64,
                "--receipt",
                str(tmp_path / "r0.json"),
            ],
            client=share,
            secret="pw",
        )
        == 1
    )
    assert not share.deleted
    assert (
        m.main(
            [
                "delete",
                "--plan",
                str(path),
                "--confirm",
                digest,
                "--receipt",
                str(tmp_path / "r.json"),
            ],
            client=share,
            secret="pw",
        )
        == 0
    )
    assert sorted(share.deleted) == sorted(plan["delete"]) and set(share.files) == set(
        plan["keep"]
    )
    receipt = json.loads((tmp_path / "r.json").read_text())
    assert (
        receipt["kept_all_present"]
        and receipt["deleted"] == len(plan["delete"])
        and "pw" not in json.dumps(receipt)
    )


def test_delete_refuses_when_the_folder_changed_since_the_plan(tmp_path):
    share = Share(names(range(1, 29)))
    path, digest = make_plan(tmp_path, share)
    victim = min(json.loads(path.read_text())["delete"])
    share.files[victim] += 1
    assert (
        m.main(
            [
                "delete",
                "--plan",
                str(path),
                "--confirm",
                digest,
                "--receipt",
                str(tmp_path / "r.json"),
            ],
            client=share,
            secret="pw",
        )
        == 1
    )
    assert not share.deleted


@pytest.mark.parametrize("folder", ["net-kingdom-identity", "openbao", "r01-x/..", ""])
def test_only_r01_dump_folders_are_accepted(tmp_path, folder):
    assert (
        m.main(
            ["plan", "--folder", folder, "--plan", str(tmp_path / "p.json")],
            client=Share({}),
            secret="pw",
        )
        == 1
    )


def test_unexpected_entries_stop_the_plan(tmp_path):
    share = Share(names(range(1, 9)), extra=[("notes.txt", 5)])
    assert (
        m.main(
            ["plan", "--folder", FOLDER, "--plan", str(tmp_path / "p.json")],
            client=share,
            secret="pw",
        )
        == 1
    )
    assert not (tmp_path / "p.json").exists()