From 61ca9a80b34b14fa0e5e19e35ee76e0086d49edf Mon Sep 17 00:00:00 2001 From: Philipp Hagemeister Date: Fri, 23 Jan 2015 01:21:30 +0100 Subject: [PATCH] [generic] Add support for BOMs (Fixes #4753) --- test/test_utils.py | 17 +++++++++++++++++ youtube_dl/extractor/generic.py | 3 ++- youtube_dl/utils.py | 20 ++++++++++++++++++++ 3 files changed, 39 insertions(+), 1 deletion(-) diff --git a/test/test_utils.py b/test/test_utils.py index 206760d99..bdd7f268a 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -28,6 +28,7 @@ from youtube_dl.utils import ( fix_xml_ampersands, InAdvancePagedList, intlist_to_bytes, + is_html, js_to_json, limit_length, OnDemandPagedList, @@ -417,5 +418,21 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4') self.assertTrue(age_restricted(18, 14)) self.assertFalse(age_restricted(18, 18)) + def test_is_html(self): + self.assertFalse(is_html(b'\x49\x44\x43\xaaa')) + self.assertTrue(is_html( # UTF-8 with BOM + b'\xef\xbb\xbf\xaaa')) + self.assertTrue(is_html( # UTF-16-LE + b'\xff\xfe<\x00h\x00t\x00m\x00l\x00>\x00\xe4\x00' + )) + self.assertTrue(is_html( # UTF-16-BE + b'\xfe\xff\x00<\x00h\x00t\x00m\x00l\x00>\x00\xe4' + )) + self.assertTrue(is_html( # UTF-32-BE + b'\x00\x00\xFE\xFF\x00\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4')) + self.assertTrue(is_html( # UTF-32-LE + b'\xFF\xFE\x00\x00<\x00\x00\x00h\x00\x00\x00t\x00\x00\x00m\x00\x00\x00l\x00\x00\x00>\x00\x00\x00\xe4\x00\x00\x00')) + if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 7a5bf9392..b893d8149 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -17,6 +17,7 @@ from ..utils import ( ExtractorError, float_or_none, HEADRequest, + is_html, orderedSet, parse_xml, smuggle_url, @@ -647,7 +648,7 @@ class GenericIE(InfoExtractor): # Maybe it's a direct link to a video? # Be careful not to download the whole thing! first_bytes = full_response.read(512) - if not re.match(r'^\s*<', first_bytes.decode('utf-8', 'replace')): + if not is_html(first_bytes): self._downloader.report_warning( 'URL could be a direct video link, returning it as such.') upload_date = unified_strdate( diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 764474c33..b433b591b 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -1631,3 +1631,23 @@ def age_restricted(content_limit, age_limit): if content_limit is None: return False # Content available for everyone return age_limit < content_limit + + +def is_html(first_bytes): + """ Detect whether a file contains HTML by examining its first bytes. """ + + BOMS = [ + (b'\xef\xbb\xbf', 'utf-8'), + (b'\x00\x00\xfe\xff', 'utf-32-be'), + (b'\xff\xfe\x00\x00', 'utf-32-le'), + (b'\xff\xfe', 'utf-16-le'), + (b'\xfe\xff', 'utf-16-be'), + ] + for bom, enc in BOMS: + if first_bytes.startswith(bom): + s = first_bytes[len(bom):].decode(enc, 'replace') + break + else: + s = first_bytes.decode('utf-8', 'replace') + + return re.match(r'^\s*<', s)