from __future__ import annotations

from pathlib import Path

from yt_scraper.parse import Segment, parse_json3, parse_vtt, merge_adjacent

FIXTURES = Path(__file__).parent / "fixtures"


def test_parse_json3_basic():
    raw = (FIXTURES / "sample.json3").read_text(encoding="utf-8")
    segs = parse_json3(raw)
    assert len(segs) >= 2
    assert segs[0].start == 0.0
    assert "Hola" in segs[0].text
    assert "segunda" in segs[0].text
    assert segs[-1].start == 6.0
    assert "Tercera" in segs[-1].text


def test_parse_json3_skips_empty_events():
    raw = (FIXTURES / "sample.json3").read_text(encoding="utf-8")
    segs = parse_json3(raw)
    texts = [s.text for s in segs]
    assert all(t.strip() for t in texts)


def test_parse_vtt_basic():
    raw = (FIXTURES / "sample.vtt").read_text(encoding="utf-8")
    segs = parse_vtt(raw)
    assert len(segs) == 2
    assert segs[0].start == 0.0
    assert "Hola" in segs[0].text
    assert "tag" in segs[0].text
    assert segs[1].start == 6.0
    assert "Tercer" in segs[1].text


def test_parse_vtt_strips_tags():
    raw = (FIXTURES / "sample.vtt").read_text(encoding="utf-8")
    segs = parse_vtt(raw)
    assert "<c." not in segs[1].text


def test_merge_adjacent():
    segs = [
        Segment(start=0.0, end=1.0, text="a"),
        Segment(start=1.0, end=2.0, text="b"),
        Segment(start=5.0, end=6.0, text="c"),
    ]
    merged = merge_adjacent(segs, max_gap=0.0)
    assert len(merged) == 2
    assert "a b" in merged[0].text
    assert merged[1].text == "c"


def test_parse_json3_invalid_json():
    assert parse_json3("not json") == []
    assert parse_json3('{"events": []}') == []