w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[ustream] Simplify channel extraction

Browse source 
the ChannelParser has been moved to a new function in utils get_meta_content
Instead of the SocialStreamParser now it uses a regex
This commit is contained in:
Jaime Marquínez Ferrándiz 2013-09-13 22:05:29 +02:00
parent 74ac9bdd82
commit a921f40799
3 changed files with 70 additions and 59 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

28
test/test_utils.py
View file

@ -11,13 +11,16 @@ import os
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
#from youtube_dl.utils import htmlentity_transform
from youtube_dl.utils import timeconvert
from youtube_dl.utils import sanitize_filename
from youtube_dl.utils import unescapeHTML
from youtube_dl.utils import orderedSet
from youtube_dl.utils import DateRange
from youtube_dl.utils import unified_strdate
from youtube_dl.utils import find_xpath_attr
from youtube_dl.utils import (
timeconvert,
sanitize_filename,
unescapeHTML,
orderedSet,
DateRange,
unified_strdate,
find_xpath_attr,
get_meta_content,
)
if sys.version_info < (3, 0):
_compat_str = lambda b: b.decode('unicode-escape')
@ -127,5 +130,16 @@ class TestUtil(unittest.TestCase):
self.assertEqual(find_xpath_attr(doc, './/node', 'x', 'a'), doc[1])
self.assertEqual(find_xpath_attr(doc, './/node', 'y', 'c'), doc[2])
def test_meta_parser(self):
testhtml = u'''
<head>
<meta name="description" content="foo &amp; bar">
<meta content='Plato' name='author'/>
</head>
'''
get_meta = lambda name: get_meta_content(name, testhtml)
self.assertEqual(get_meta('description'), u'foo & bar')
self.assertEqual(get_meta('author'), 'Plato')
if __name__ == '__main__':
unittest.main()

49
youtube_dl/extractor/ustream.py
View file

@ -4,7 +4,7 @@ import re
from .common import InfoExtractor
from ..utils import (
compat_urlparse,
compat_html_parser,
get_meta_content,
)
@ -49,40 +49,6 @@ class UstreamIE(InfoExtractor):
}
return info
# More robust than regular expressions
class ChannelParser(compat_html_parser.HTMLParser):
"""
<meta name="ustream:channel_id" content="1234">
"""
channel_id = None
def handle_starttag(self, tag, attrs):
if tag != 'meta':
return
values = dict(attrs)
if values.get('name') != 'ustream:channel_id':
return
value = values.get('content', '')
if value.isdigit():
self.channel_id = value
class SocialstreamParser(compat_html_parser.HTMLParser):
"""
<li class="content123 video" data-content-id="123" data-length="1452"
data-href="/recorded/123" data-og-url="/recorded/123">
"""
def __init__(self):
compat_html_parser.HTMLParser.__init__(self)
self.content_ids = []
def handle_starttag(self, tag, attrs):
if tag != 'li':
return
for (attr, value) in attrs:
if attr == 'data-content-id' and value.isdigit():
self.content_ids.append(value)
class UstreamChannelIE(InfoExtractor):
_VALID_URL = r'https?://www\.ustream\.tv/channel/(?P<slug>.+)'
IE_NAME = u'ustream:channel'
@ -90,21 +56,16 @@ class UstreamChannelIE(InfoExtractor):
def _real_extract(self, url):
m = re.match(self._VALID_URL, url)
slug = m.group('slug')
webpage = self._download_webpage(url, slug)
channel_id = get_meta_content('ustream:channel_id', webpage)
p = ChannelParser()
p.feed(self._download_webpage(url, slug))
p.close()
channel_id = p.channel_id
p = SocialstreamParser()
BASE = 'http://www.ustream.tv'
next_url = '/ajax/socialstream/videos/%s/1.json' % channel_id
video_ids = []
while next_url:
reply = json.loads(self._download_webpage(compat_urlparse.urljoin(BASE, next_url), channel_id))
p.feed(reply['data'])
video_ids.extend(re.findall(r'data-content-id="(\d.*)"', reply['data']))
next_url = reply['nextUrl']
p.close()
video_ids = p.content_ids
urls = ['http://www.ustream.tv/recorded/' + vid for vid in video_ids]
url_entries = [self.url_result(eurl, 'Ustream') for eurl in urls]

52
youtube_dl/utils.py
View file

@ -249,7 +249,17 @@ def htmlentity_transform(matchobj):
return (u'&%s;' % entity)
compat_html_parser.locatestarttagend = re.compile(r"""<[a-zA-Z][-.a-zA-Z0-9:_]*(?:\s+(?:(?<=['"\s])[^\s/>][^\s/=>]*(?:\s*=+\s*(?:'[^']*'|"[^"]*"|(?!['"])[^>\s]*))?\s*)*)?\s*""", re.VERBOSE) # backport bugfix
class AttrParser(compat_html_parser.HTMLParser):
class BaseHTMLParser(compat_html_parser.HTMLParser):
def __init(self):
compat_html_parser.HTMLParser.__init__(self)
self.html = None
def loads(self, html):
self.html = html
self.feed(html)
self.close()
class AttrParser(BaseHTMLParser):
"""Modified HTMLParser that isolates a tag with the specified attribute"""
def __init__(self, attribute, value):
self.attribute = attribute
@ -257,10 +267,9 @@ class AttrParser(compat_html_parser.HTMLParser):
self.result = None
self.started = False
self.depth = {}
self.html = None
self.watch_startpos = False
self.error_count = 0
compat_html_parser.HTMLParser.__init__(self)
BaseHTMLParser.__init__(self)
def error(self, message):
if self.error_count > 10 or self.started:
@ -269,11 +278,6 @@ class AttrParser(compat_html_parser.HTMLParser):
self.error_count += 1
self.goahead(1)
def loads(self, html):
self.html = html
self.feed(html)
self.close()
def handle_starttag(self, tag, attrs):
attrs = dict(attrs)
if self.started:
@ -334,6 +338,38 @@ def get_element_by_attribute(attribute, value, html):
pass
return parser.get_result()
class MetaParser(BaseHTMLParser):
"""
Modified HTMLParser that isolates a meta tag with the specified name
attribute.
"""
def __init__(self, name):
BaseHTMLParser.__init__(self)
self.name = name
self.content = None
self.result = None
def handle_starttag(self, tag, attrs):
if tag != 'meta':
return
attrs = dict(attrs)
if attrs.get('name') == self.name:
self.result = attrs.get('content')
def get_result(self):
return self.result
def get_meta_content(name, html):
"""
Return the content attribute from the meta tag with the given name attribute.
"""
parser = MetaParser(name)
try:
parser.loads(html)
except compat_html_parser.HTMLParseError:
pass
return parser.get_result()
def clean_html(html):
"""Clean an HTML snippet into a readable string"""