from __future__ import annotations

import io
import json
from pathlib import Path

from yt_scraper.extract import extract_via_watch_page


WATCH_PAGE = (
    "var ytInitialPlayerResponse = "
    + json.dumps({
        "playabilityStatus": {"status": "OK"},
        "videoDetails": {
            "title": "Members only test",
            "author": "Canal",
            "lengthSeconds": "42",
            "viewCount": "7",
            "shortDescription": "desc",
            "keywords": ["a", "b"],
        },
        "microformat": {"playerMicroformatRenderer": {"publishDate": "2026-09-01"}},
        "captions": {"playerCaptionsTracklistRenderer": {"captionTracks": [
            {"languageCode": "es", "kind": "asr", "baseUrl": "https://captions.test/es"},
        ]}},
    })
    + ";"
)

JSON3 = json.dumps({
    "events": [
        {"tStartMs": 0, "dDurationMs": 1000, "segs": [{"utf8": "hola "}, {"utf8": "mundo"}]},
        {"tStartMs": 1000, "dDurationMs": 500, "segs": [{"utf8": "segundo"}]},
    ]
})


class _FakeResponse(io.BytesIO):
    def __enter__(self):
        return self

    def __exit__(self, *a):
        return False


def test_watch_page_fallback_builds_segments(tmp_path: Path, monkeypatch):
    from yt_scraper import extract as ex

    def fake_urlopen(cookies_file, url, **kw):
        if url.startswith("https://www.youtube.com/"):
            return _FakeResponse(WATCH_PAGE.encode("utf-8"))
        assert "fmt=json3" in url
        return _FakeResponse(JSON3.encode("utf-8"))

    monkeypatch.setattr(ex, "_session_urlopen", fake_urlopen)
    monkeypatch.setattr(ex.GLOBAL_PACER, "wait", lambda **kw: None)

    data = extract_via_watch_page(
        "https://www.youtube.com/watch?v=xyz", "cookies/fake.txt", {"es": "any"}
    )

    assert data is not None
    assert data.subtitle is not None and data.subtitle.lang == "es"
    # parse_auto_dump merges caption events with no gap between them.
    assert len(data.segments) == 1
    assert data.segments[0].text == "hola mundo segundo"
    assert data.info["title"] == "Members only test"
    assert data.info["upload_date"] == "20260901"
    assert data.info["duration"] == 42
    assert data.skip_reason is None


def test_watch_page_fallback_without_tracks_returns_none(tmp_path: Path, monkeypatch):
    from yt_scraper import extract as ex

    page = WATCH_PAGE.replace('"captionTracks": [{', '"captionTracks": [{')
    page = "var ytInitialPlayerResponse = " + json.dumps({
        "playabilityStatus": {"status": "OK"},
        "videoDetails": {"title": "x"},
    }) + ";"

    monkeypatch.setattr(ex, "_session_urlopen",
                        lambda *a, **kw: _FakeResponse(page.encode("utf-8")))
    monkeypatch.setattr(ex.GLOBAL_PACER, "wait", lambda **kw: None)

    assert extract_via_watch_page(
        "https://www.youtube.com/watch?v=xyz", "cookies/fake.txt", {"es": "any"}
    ) is None