def scrape(self, url):
"""Scrapes a url by passing it through youtube-dl"""
if not is_youtube(url):
return
# FIXME: Sometimes youtube-dl takes a *long* time to run. This
# needs to give indication of progress.
try:
output = subprocess.check_output(
['youtube-dl', '-j', url],
stderr=subprocess.STDOUT
)
except subprocess.CalledProcessError as cpe:
raise ScraperError('youtube-dl said "{0}".'.format(cpe.output))
except OSError:
raise ScraperError('youtube-dl not installed or not on PATH.')
# Each line is a single JSON object.
items = []
for line in output.splitlines():
items.append(json.loads(line))
items = [self.transform_item(item) for item in items]
return items
def test_is_youtube():
data = [
('http://www.youtube.com/watch?v=N29XAFjiKf4', True),
('http://youtu.be/N29XAFjiKf4', True),
]
for url, expected in data:
assert is_youtube(url) == expected
def test_is_youtube():
data = [
('http://www.youtube.com/watch?v=N29XAFjiKf4', True),
('http://youtu.be/N29XAFjiKf4', True),
]
for url, expected in data:
eq_(is_youtube(url), expected)
def scrape(self, url):
"""Scrapes a url by passing it through youtube-dl"""
if not is_youtube(url):
return
# FIXME: Sometimes youtube-dl takes a *long* time to run. This
# needs to give indication of progress.
try:
output = subprocess.check_output(['youtube-dl', '-j', url],
stderr=subprocess.STDOUT)
except subprocess.CalledProcessError as cpe:
raise ScraperError('youtube-dl said "{0}".'.format(cpe.output))
except OSError:
raise ScraperError('youtube-dl not installed or not on PATH.')
# Each line is a single JSON object.
items = []
for line in output.splitlines():
items.append(json.loads(line))
items = [self.transform_item(item) for item in items]
return items
