Files
devplacepy/tests/unit/embeds.py
T

176 lines
7.6 KiB
Python
Raw Normal View History

# retoor <retoor@molodetz.nl>
import itertools
import random
import re
from devplacepy.embeds import (
EMBED_JS_PATH,
MAX_MUSIC_EMBEDS,
MAX_START_SECONDS,
EmbedParser,
)
from devplacepy.rendering import render_content
VIDEO = "dQw4w9WgXcQ"
TRACK = "4D7u5KF7SP8"
ALBUM = "OLAK5uy_nZZjkBu_E4olFSb5Ey-fQ-4a0ZCqJICdQ"
PLAIN = "PLFgquLnL59alCl_2TQvOiD5Vgm1hCaGSI"
BASE = "https://www.youtube.com/embed"
EMBED_CASES = [
(f"https://music.youtube.com/watch?v={TRACK}", f"{BASE}/{TRACK}", True),
(f"https://music.youtube.com/watch?v={TRACK}&list={ALBUM}", f"{BASE}/{TRACK}?list={ALBUM}", True),
(f"https://music.youtube.com/watch?list={ALBUM}&si=abc&v={TRACK}", f"{BASE}/{TRACK}?list={ALBUM}", True),
(f"https://music.youtube.com/watch?v=bad&list={ALBUM}", f"{BASE}/videoseries?list={ALBUM}", True),
(f"https://music.youtube.com/playlist?list={ALBUM}", f"{BASE}/videoseries?list={ALBUM}", True),
(f"https://music.youtube.com/watch?v={TRACK}&list=RDAMVM{TRACK}", f"{BASE}/{TRACK}?list=RDAMVM{TRACK}", True),
(f"https://music.youtube.com/watch?v={TRACK}&t=45", f"{BASE}/{TRACK}?start=45", True),
(f"https://www.youtube.com/playlist?list={PLAIN}", f"{BASE}/videoseries?list={PLAIN}", False),
(f"https://www.youtube.com/watch?v={VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/watch?feature=share&v={VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://m.youtube.com/watch?v={VIDEO}&t=30s", f"{BASE}/{VIDEO}?start=30", False),
(f"https://youtube.com/embed/{VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube-nocookie.com/embed/{VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/v/{VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/shorts/{VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://youtu.be/{VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://youtu.be/{VIDEO}?t=30", f"{BASE}/{VIDEO}?start=30", False),
(f"http://youtu.be/{VIDEO}", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/embed/videoseries?list={PLAIN}", f"{BASE}/videoseries?list={PLAIN}", False),
(f"https://www.youtube.com/embed?listType=playlist&list={PLAIN}", f"{BASE}/videoseries?list={PLAIN}", False),
(f"https://www.youtube.com/watch?v={VIDEO}&list=WL", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/watch?v={VIDEO}&list=bad%22%3E%3Cscript%3E", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/watch?v={VIDEO}&t={MAX_START_SECONDS + 1}", f"{BASE}/{VIDEO}", False),
(f"https://www.youtube.com/watch?v={VIDEO}&t=1m30s", f"{BASE}/{VIDEO}", False),
("https://music.youtube.com/browse/MPREb_abcdefghijk", None, False),
("https://music.youtube.com/channel/UCabcdefghijklmnopqrstuv", None, False),
("https://music.youtube.com/search?q=daft+punk", None, False),
("https://music.youtube.com/library", None, False),
("https://music.youtube.com/playlist", None, False),
(f"https://music.youtube.com/embed/{TRACK}", None, False),
(f"https://evilyoutube.com/watch?v={VIDEO}", None, False),
(f"https://youtube.com.evil.example/watch?v={VIDEO}", None, False),
(f"https://evil.example/https://www.youtube.com/watch?v={VIDEO}", None, False),
(f"https://notyoutu.be/{VIDEO}", None, False),
(f"https://youtu.be/{VIDEO}/extra", None, False),
("https://www.youtube.com/watch?v=short", None, False),
("https://www.youtube.com/watch", None, False),
("https://www.youtube.com/embed/videoseries", None, False),
(f"javascript://youtube.com/watch?v={VIDEO}", None, False),
(f"ftp://youtube.com/watch?v={VIDEO}", None, False),
("not a url", None, False),
("", None, False),
]
def _iframe_srcs(rendered: str) -> list[str]:
return re.findall(r'<iframe src="([^"]*)"', str(rendered))
def test_parser_cases_table():
for url, expected, music in EMBED_CASES:
embed = EmbedParser.parse(url)
if expected is None:
assert embed is None, url
continue
assert embed is not None, url
assert EmbedParser.src(embed) == expected, url
assert embed.music is music, url
def test_render_matches_parser_for_every_case():
for url, expected, _ in EMBED_CASES:
srcs = _iframe_srcs(render_content(url))
if expected is None:
assert srcs == [], url
else:
assert srcs == [expected.replace("&", "&amp;")], url
def test_music_embed_markup_and_caption():
out = str(render_content(f"https://music.youtube.com/watch?v={TRACK}"))
assert 'class="embed-music"' in out
assert 'class="embed-youtube"' in out
assert 'loading="lazy"' in out
assert 'class="embed-music-caption"' in out
assert "Open in YouTube Music" in out
assert f'href="https://music.youtube.com/watch?v={TRACK}"' in out
assert 'rel="noopener noreferrer"' in out
def test_plain_youtube_keeps_old_markup():
out = str(render_content(f"https://youtu.be/{VIDEO}"))
assert 'class="embed-youtube"' in out
assert "embed-music" not in out
assert "loading=" not in out
def test_music_embed_cap_turns_overflow_into_links():
urls = [f"https://music.youtube.com/watch?v={TRACK[:-1]}{c}" for c in "abcde"]
out = str(render_content("\n\n".join(urls)))
assert out.count("<iframe") == MAX_MUSIC_EMBEDS
assert out.count('class="embed-music"') == MAX_MUSIC_EMBEDS
for url in urls[MAX_MUSIC_EMBEDS:]:
assert f'<a href="{url}"' in out
def test_music_cap_does_not_limit_plain_youtube():
urls = [f"https://youtu.be/{VIDEO[:-1]}{c}" for c in "abcdefg"]
out = str(render_content("\n\n".join(urls)))
assert out.count("<iframe") == 7
def test_music_cap_is_per_content_item():
text = f"https://music.youtube.com/watch?v={TRACK}"
for _ in range(3):
assert str(render_content(text)).count("<iframe") == 1
def test_urls_inside_code_and_links_are_not_embedded():
url = f"https://music.youtube.com/watch?v={TRACK}"
assert "<iframe" not in str(render_content(f"`{url}`"))
assert "<iframe" not in str(render_content(f"[x]({url})"))
assert "<iframe" not in str(render_content(f"```\n{url}\n```"))
def test_trailing_punctuation_does_not_break_music_url():
out = str(render_content(f"listen: https://music.youtube.com/watch?v={TRACK}."))
assert f'src="{BASE}/{TRACK}"' in out
assert out.count("<iframe") == 1
def test_only_validated_characters_reach_the_iframe_src():
rng = random.Random(7)
alphabet = "abcXYZ019_-\"'<>&%/ ;()=:."
for _ in range(2000):
junk = "".join(rng.choice(alphabet) for _ in range(rng.randint(1, 40)))
for url in (
f"https://music.youtube.com/watch?v={junk}&list={junk}",
f"https://music.youtube.com/playlist?list={junk}",
f"https://youtu.be/{junk}",
):
embed = EmbedParser.parse(url)
if embed is None:
continue
assert re.fullmatch(r"https://www\.youtube\.com/embed/[A-Za-z0-9_\-/?=&]+", EmbedParser.src(embed)), url
def test_start_parsing_is_bounded_across_the_domain():
for seconds, suffix in itertools.product(range(0, MAX_START_SECONDS + 5, 997), ("", "s")):
embed = EmbedParser.parse(f"https://youtu.be/{VIDEO}?t={seconds}{suffix}")
assert embed is not None
assert 0 <= embed.start <= MAX_START_SECONDS
assert embed.start in (0, seconds)
def test_generated_js_spec_matches_the_python_spec():
expected_path = EMBED_JS_PATH
current = expected_path.read_text(encoding="utf-8")
scratch = expected_path.with_name("embed-spec.check.js")
try:
EmbedParser.write_module(scratch)
assert scratch.read_text(encoding="utf-8") == current
finally:
scratch.unlink(missing_ok=True)