webmail-client/backend/tests/threading/test_threading.py

"""JWZ threading test vectors (SPEC §7.4). All must pass."""

from __future__ import annotations

from tests.threading.env import ThreadingEnv


async def test_linear_chain(env: ThreadingEnv) -> None:
    a = await env.add(message_id="a@x.com", subject="Hello")
    b = await env.add(
        message_id="b@x.com",
        subject="Re: Hello",
        in_reply_to="a@x.com",
        references=["a@x.com"],
        minutes=1,
    )
    c = await env.add(
        message_id="c@x.com",
        subject="Re: Hello",
        in_reply_to="b@x.com",
        references=["a@x.com", "b@x.com"],
        minutes=2,
    )
    assert a.thread_id == b.thread_id == c.thread_id


async def test_chain_out_of_order(env: ThreadingEnv) -> None:
    # C arrives first, then A, then B — all must end in one thread.
    c = await env.add(
        message_id="c@x.com", subject="Re: Hi", references=["a@x.com", "b@x.com"], minutes=2
    )
    a = await env.add(message_id="a@x.com", subject="Hi", minutes=0)
    b = await env.add(message_id="b@x.com", subject="Re: Hi", references=["a@x.com"], minutes=1)
    assert a.thread_id == b.thread_id == c.thread_id


async def test_reply_missing_in_reply_to_but_has_references(env: ThreadingEnv) -> None:
    a = await env.add(message_id="a@x.com", subject="Topic")
    b = await env.add(
        message_id="b@x.com", subject="Re: Topic", references=["a@x.com"], minutes=1
    )  # no In-Reply-To
    assert a.thread_id == b.thread_id


async def test_thread_merge_on_late_parent(env: ThreadingEnv) -> None:
    # Two children of A arrive first (each referencing only A) forming separate
    # threads; the late parent A must merge them into one.
    b = await env.add(message_id="b@x.com", subject="Re: T", references=["a@x.com"], minutes=1)
    c = await env.add(message_id="c@x.com", subject="Re: T", references=["a@x.com"], minutes=2)
    # b and c both reference a but not each other -> initially could be separate threads
    a = await env.add(message_id="a@x.com", subject="T", minutes=0)
    await env.session.refresh(b)
    await env.session.refresh(c)
    assert a.thread_id == b.thread_id == c.thread_id


async def test_broken_client_subject_only_joins(env: ThreadingEnv) -> None:
    # A client that drops References; joins by subject + shared participant within 7 days.
    a = await env.add(
        message_id="a@x.com",
        subject="Project update",
        from_email="alice@example.com",
        to_email="bob@example.com",
    )
    b = await env.add(
        message_id="b@x.com",
        subject="Re: Project update",
        from_email="bob@example.com",
        to_email="alice@example.com",
        minutes=60,
    )  # no references, shares participants
    assert a.thread_id == b.thread_id


async def test_auto_replies_to_different_originals_do_not_join(env: ThreadingEnv) -> None:
    # Two out-of-office auto-replies, same subject and same recipient, but each
    # replying to a DIFFERENT (absent) original. They must not be subject-merged.
    a = await env.add(
        message_id="autoA@tvp.pl",
        subject="Automatyczna odpowiedź: Silent Hill",
        from_email="robert@tvp.pl",
        to_email="me@gmail.com",
        in_reply_to="orig1@mail.gmail.com",
        references=["orig1@mail.gmail.com"],
    )
    b = await env.add(
        message_id="autoB@wfdif.com.pl",
        subject="Automatyczna odpowiedź: Silent Hill",
        from_email="dyrektor@wfdif.com.pl",
        to_email="me@gmail.com",
        in_reply_to="orig2@mail.gmail.com",
        references=["orig2@mail.gmail.com"],
        minutes=5,
    )
    assert a.thread_id != b.thread_id


async def test_replies_to_same_absent_parent_join(env: ThreadingEnv) -> None:
    # Two replies that reference the SAME (absent) original — and not each other —
    # still thread together via the shared parent reference.
    a = await env.add(
        message_id="r1@x.com",
        subject="Re: Topic",
        in_reply_to="orig@x.com",
        references=["orig@x.com"],
    )
    b = await env.add(
        message_id="r2@x.com",
        subject="Re: Topic",
        in_reply_to="orig@x.com",
        references=["orig@x.com"],
        minutes=3,
    )
    assert a.thread_id == b.thread_id


async def test_distinct_topic_same_subject_does_not_join(env: ThreadingEnv) -> None:
    a = await env.add(
        message_id="a@x.com",
        subject="hi",
        from_email="alice@example.com",
        to_email="bob@example.com",
    )
    b = await env.add(
        message_id="b@x.com",
        subject="hi",
        from_email="carol@example.com",
        to_email="dave@example.com",
        minutes=5,
    )
    assert a.thread_id != b.thread_id


async def test_cross_account_merge(env: ThreadingEnv) -> None:
    # A message in account A references a message in account B -> one thread (I-3).
    a = await env.add(message_id="a@x.com", subject="Shared", account="a")
    b = await env.add(
        message_id="b@x.com", subject="Re: Shared", references=["a@x.com"], account="b", minutes=1
    )
    assert a.thread_id == b.thread_id
    assert a.account_id != b.account_id


async def test_duplicate_message_id_same_account_merges(env: ThreadingEnv) -> None:
    # Same message in two mailboxes of ONE account (e.g. Gmail INBOX + All Mail):
    # the shared Message-ID is authoritative -> one thread. (Threading-level view;
    # the indexer decides authoritative vs synthetic per SPEC §7.1 — see indexer tests.)
    a1 = await env.add(message_id="dup@x.com", subject="Dup", account="a")
    a2 = await env.add(message_id="dup@x.com", subject="Dup", account="a", minutes=1)
    assert a1.thread_id == a2.thread_id


async def test_synthetic_ids_do_not_join(env: ThreadingEnv) -> None:
    # Two messages that were given synthetic ids (§7.1 duplicate handling) with no
    # references must not merge, even with the same subject far apart in time.
    a = await env.add(message_id="deadbeef@synthetic.local", subject="Welcome", minutes=0)
    b = await env.add(
        message_id="cafebabe@synthetic.local", subject="Welcome", account="b", minutes=60 * 24 * 30
    )
    assert a.thread_id != b.thread_id


async def test_synthetic_id_isolation(env: ThreadingEnv) -> None:
    # Two messages lacking Message-ID and References with empty subjects never join.
    a = await env.add(message_id=None, subject="")
    b = await env.add(message_id=None, subject="", minutes=1)
    assert a.thread_id != b.thread_id


async def test_denormalization_counts(env: ThreadingEnv) -> None:
    from app.models import Thread

    a = await env.add(message_id="a@x.com", subject="Count", flags=["\\Seen"])
    await env.add(message_id="b@x.com", subject="Re: Count", references=["a@x.com"], minutes=1)
    thread = await env.session.get(Thread, a.thread_id)
    assert thread is not None
    assert thread.message_count == 2
    assert thread.unread_count == 1  # only b is unread
    assert thread.snippet == "Re: Count"  # newest message's snippet