mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-10-31 14:45:14 +00:00
Code Issues Projects Releases Wiki Activity

[core] Support decoding multiple content encodings (#7142)

Browse Source 
Authored by: coletdjnz
This commit is contained in:
coletdjnz
2023-05-27 22:40:05 +12:00
committed by GitHub
parent 3f66b6fe50
commit daafbf49b3
2 changed files with 108 additions and 29 deletions
Download Patch File Download Diff File Expand all files Collapse all files

76
test/test_http.py
View File

@@ -17,9 +17,11 @@ import tempfile
import threading import threading
import urllib.error import urllib.error
import urllib.request import urllib.request
import zlib
from test.helper import http_server_port from test.helper import http_server_port
from yt_dlp import YoutubeDL from yt_dlp import YoutubeDL
from yt_dlp.dependencies import brotli
from yt_dlp.utils import sanitized_Request, urlencode_postdata from yt_dlp.utils import sanitized_Request, urlencode_postdata
from .helper import FakeYDL from .helper import FakeYDL
@@ -148,6 +150,31 @@ class HTTPTestRequestHandler(http.server.BaseHTTPRequestHandler):
self.send_header('Location', new_url) self.send_header('Location', new_url)
self.send_header('Content-Length', '0') self.send_header('Content-Length', '0')
self.end_headers() self.end_headers()
elif self.path == '/content-encoding':
encodings = self.headers.get('ytdl-encoding', '')
payload = b'<html><video src="/vid.mp4" /></html>'
for encoding in filter(None, (e.strip() for e in encodings.split(','))):
if encoding == 'br' and brotli:
payload = brotli.compress(payload)
elif encoding == 'gzip':
buf = io.BytesIO()
with gzip.GzipFile(fileobj=buf, mode='wb') as f:
f.write(payload)
payload = buf.getvalue()
elif encoding == 'deflate':
payload = zlib.compress(payload)
elif encoding == 'unsupported':
payload = b'raw'
break
else:
self._status(415)
return
self.send_response(200)
self.send_header('Content-Encoding', encodings)
self.send_header('Content-Length', str(len(payload)))
self.end_headers()
self.wfile.write(payload)
else: else:
self._status(404) self._status(404)
@@ -302,6 +329,55 @@ class TestHTTP(unittest.TestCase):
data = ydl.urlopen(sanitized_Request(f'http://localhost:{self.http_port}/trailing_garbage')).read().decode('utf-8') data = ydl.urlopen(sanitized_Request(f'http://localhost:{self.http_port}/trailing_garbage')).read().decode('utf-8')
self.assertEqual(data, '<html><video src="/vid.mp4" /></html>') self.assertEqual(data, '<html><video src="/vid.mp4" /></html>')
@unittest.skipUnless(brotli, 'brotli support is not installed')
def test_brotli(self):
with FakeYDL() as ydl:
res = ydl.urlopen(
sanitized_Request(
f'http://127.0.0.1:{self.http_port}/content-encoding',
headers={'ytdl-encoding': 'br'}))
self.assertEqual(res.headers.get('Content-Encoding'), 'br')
self.assertEqual(res.read(), b'<html><video src="/vid.mp4" /></html>')
def test_deflate(self):
with FakeYDL() as ydl:
res = ydl.urlopen(
sanitized_Request(
f'http://127.0.0.1:{self.http_port}/content-encoding',
headers={'ytdl-encoding': 'deflate'}))
self.assertEqual(res.headers.get('Content-Encoding'), 'deflate')
self.assertEqual(res.read(), b'<html><video src="/vid.mp4" /></html>')
def test_gzip(self):
with FakeYDL() as ydl:
res = ydl.urlopen(
sanitized_Request(
f'http://127.0.0.1:{self.http_port}/content-encoding',
headers={'ytdl-encoding': 'gzip'}))
self.assertEqual(res.headers.get('Content-Encoding'), 'gzip')
self.assertEqual(res.read(), b'<html><video src="/vid.mp4" /></html>')
def test_multiple_encodings(self):
# https://www.rfc-editor.org/rfc/rfc9110.html#section-8.4
with FakeYDL() as ydl:
for pair in ('gzip,deflate', 'deflate, gzip', 'gzip, gzip', 'deflate, deflate'):
res = ydl.urlopen(
sanitized_Request(
f'http://127.0.0.1:{self.http_port}/content-encoding',
headers={'ytdl-encoding': pair}))
self.assertEqual(res.headers.get('Content-Encoding'), pair)
self.assertEqual(res.read(), b'<html><video src="/vid.mp4" /></html>')
def test_unsupported_encoding(self):
# it should return the raw content
with FakeYDL() as ydl:
res = ydl.urlopen(
sanitized_Request(
f'http://127.0.0.1:{self.http_port}/content-encoding',
headers={'ytdl-encoding': 'unsupported'}))
self.assertEqual(res.headers.get('Content-Encoding'), 'unsupported')
self.assertEqual(res.read(), b'raw')
class TestClientCert(unittest.TestCase): class TestClientCert(unittest.TestCase):
def setUp(self): def setUp(self):

61
yt_dlp/utils/_utils.py
View File

@@ -1361,6 +1361,23 @@ class YoutubeDLHandler(urllib.request.HTTPHandler):
return data return data
return brotli.decompress(data) return brotli.decompress(data)
@staticmethod
def gz(data):
gz = gzip.GzipFile(fileobj=io.BytesIO(data), mode='rb')
try:
return gz.read()
except OSError as original_oserror:
# There may be junk add the end of the file
# See http://stackoverflow.com/q/4928560/35070 for details
for i in range(1, 1024):
try:
gz = gzip.GzipFile(fileobj=io.BytesIO(data[:-i]), mode='rb')
return gz.read()
except OSError:
continue
else:
raise original_oserror
def http_request(self, req): def http_request(self, req):
# According to RFC 3986, URLs can not contain non-ASCII characters, however this is not # According to RFC 3986, URLs can not contain non-ASCII characters, however this is not
# always respected by websites, some tend to give out URLs with non percent-encoded # always respected by websites, some tend to give out URLs with non percent-encoded
@@ -1394,35 +1411,21 @@ class YoutubeDLHandler(urllib.request.HTTPHandler):
def http_response(self, req, resp): def http_response(self, req, resp):
old_resp = resp old_resp = resp
# gzip
if resp.headers.get('Content-encoding', '') == 'gzip': # Content-Encoding header lists the encodings in order that they were applied [1].
content = resp.read() # To decompress, we simply do the reverse.
gz = gzip.GzipFile(fileobj=io.BytesIO(content), mode='rb') # [1]: https://datatracker.ietf.org/doc/html/rfc9110#name-content-encoding
try: decoded_response = None
uncompressed = io.BytesIO(gz.read()) for encoding in (e.strip() for e in reversed(resp.headers.get('Content-encoding', '').split(','))):
except OSError as original_ioerror: if encoding == 'gzip':
# There may be junk add the end of the file decoded_response = self.gz(decoded_response or resp.read())
# See http://stackoverflow.com/q/4928560/35070 for details elif encoding == 'deflate':
for i in range(1, 1024): decoded_response = self.deflate(decoded_response or resp.read())
try: elif encoding == 'br' and brotli:
gz = gzip.GzipFile(fileobj=io.BytesIO(content[:-i]), mode='rb') decoded_response = self.brotli(decoded_response or resp.read())
uncompressed = io.BytesIO(gz.read())
except OSError: if decoded_response is not None:
continue resp = urllib.request.addinfourl(io.BytesIO(decoded_response), old_resp.headers, old_resp.url, old_resp.code)
break
else:
raise original_ioerror
resp = urllib.request.addinfourl(uncompressed, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
# deflate
if resp.headers.get('Content-encoding', '') == 'deflate':
gz = io.BytesIO(self.deflate(resp.read()))
resp = urllib.request.addinfourl(gz, old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg
# brotli
if resp.headers.get('Content-encoding', '') == 'br':
resp = urllib.request.addinfourl(
io.BytesIO(self.brotli(resp.read())), old_resp.headers, old_resp.url, old_resp.code)
resp.msg = old_resp.msg resp.msg = old_resp.msg
# Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see # Percent-encode redirect URL of Location HTTP header to satisfy RFC 3986 (see
# https://github.com/ytdl-org/youtube-dl/issues/6457). # https://github.com/ytdl-org/youtube-dl/issues/6457).