Add fatal=False parameter to _download_* functions.

This allows us to simplify the calls in the youtube extractor even further.
This commit is contained in:
Philipp Hagemeister 2013-12-09 01:49:01 +01:00
parent baa7b1978b
commit 7cc3570e53
2 changed files with 57 additions and 40 deletions

View file

@ -154,27 +154,38 @@ class InfoExtractor(object):
def IE_NAME(self): def IE_NAME(self):
return type(self).__name__[:-2] return type(self).__name__[:-2]
def _request_webpage(self, url_or_request, video_id, note=None, errnote=None): def _request_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
""" Returns the response handle """ """ Returns the response handle """
if note is None: if note is None:
self.report_download_webpage(video_id) self.report_download_webpage(video_id)
elif note is not False: elif note is not False:
self.to_screen(u'%s: %s' % (video_id, note)) if video_id is None:
self.to_screen(u'%s' % (note,))
else:
self.to_screen(u'%s: %s' % (video_id, note))
try: try:
return self._downloader.urlopen(url_or_request) return self._downloader.urlopen(url_or_request)
except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
if errnote is None: if errnote is None:
errnote = u'Unable to download webpage' errnote = u'Unable to download webpage'
raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2], cause=err) errmsg = u'%s: %s' % (errnote, compat_str(err))
if fatal:
raise ExtractorError(errmsg, sys.exc_info()[2], cause=err)
else:
self._downloader.report_warning(errmsg)
return False
def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None): def _download_webpage_handle(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
""" Returns a tuple (page content as string, URL handle) """ """ Returns a tuple (page content as string, URL handle) """
# Strip hashes from the URL (#1038) # Strip hashes from the URL (#1038)
if isinstance(url_or_request, (compat_str, str)): if isinstance(url_or_request, (compat_str, str)):
url_or_request = url_or_request.partition('#')[0] url_or_request = url_or_request.partition('#')[0]
urlh = self._request_webpage(url_or_request, video_id, note, errnote) urlh = self._request_webpage(url_or_request, video_id, note, errnote, fatal)
if urlh is False:
assert not fatal
return False
content_type = urlh.headers.get('Content-Type', '') content_type = urlh.headers.get('Content-Type', '')
webpage_bytes = urlh.read() webpage_bytes = urlh.read()
m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type) m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
@ -209,9 +220,14 @@ class InfoExtractor(object):
content = webpage_bytes.decode(encoding, 'replace') content = webpage_bytes.decode(encoding, 'replace')
return (content, urlh) return (content, urlh)
def _download_webpage(self, url_or_request, video_id, note=None, errnote=None): def _download_webpage(self, url_or_request, video_id, note=None, errnote=None, fatal=True):
""" Returns the data of the page as a string """ """ Returns the data of the page as a string """
return self._download_webpage_handle(url_or_request, video_id, note, errnote)[0] res = self._download_webpage_handle(url_or_request, video_id, note, errnote, fatal)
if res is False:
return res
else:
content, _ = res
return content
def _download_xml(self, url_or_request, video_id, def _download_xml(self, url_or_request, video_id,
note=u'Downloading XML', errnote=u'Unable to download XML'): note=u'Downloading XML', errnote=u'Unable to download XML'):

View file

@ -42,19 +42,11 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
# If True it will raise an error if no login info is provided # If True it will raise an error if no login info is provided
_LOGIN_REQUIRED = False _LOGIN_REQUIRED = False
def report_lang(self):
"""Report attempt to set language."""
self.to_screen(u'Setting language')
def _set_language(self): def _set_language(self):
request = compat_urllib_request.Request(self._LANG_URL) return bool(self._download_webpage(
try: self._LANG_URL, None,
self.report_lang() note=u'Setting language', errnote='unable to set language',
self._download_webpage(self._LANG_URL, None, False) fatal=False))
except ExtractorError as err:
self._downloader.report_warning(u'unable to set language: %s' % compat_str(err.cause))
return False
return True
def _login(self): def _login(self):
(username, password) = self._get_login_info() (username, password) = self._get_login_info()
@ -64,8 +56,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True) raise ExtractorError(u'No login info available, needed for using %s.' % self.IE_NAME, expected=True)
return False return False
login_page = self._download_webpage(self._LOGIN_URL, None, False, login_page = self._download_webpage(
u'Unable to fetch login page') self._LOGIN_URL, None,
note=u'Downloading login page',
errnote=u'unable to fetch login page', fatal=False)
if login_page is False:
return
galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"', galx = self._search_regex(r'(?s)<input.+?name="GALX".+?value="(.+?)"',
login_page, u'Login GALX parameter') login_page, u'Login GALX parameter')
@ -95,26 +91,28 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
# chokes on unicode # chokes on unicode
login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items()) login_form = dict((k.encode('utf-8'), v.encode('utf-8')) for k,v in login_form_strs.items())
login_data = compat_urllib_parse.urlencode(login_form).encode('ascii') login_data = compat_urllib_parse.urlencode(login_form).encode('ascii')
request = compat_urllib_request.Request(self._LOGIN_URL, login_data)
try: req = compat_urllib_request.Request(self._LOGIN_URL, login_data)
self.report_login() login_results = self._download_webpage(
login_results = self._download_webpage(request, None, False) req, None,
if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None: note=u'Logging in', errnote=u'unable to log in', fatal=False)
self._downloader.report_warning(u'unable to log in: bad username or password') if login_results is False:
return False return False
except ExtractorError as err: if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
self._downloader.report_warning(u'unable to log in: %s' % compat_str(err.cause)) self._downloader.report_warning(u'unable to log in: bad username or password')
return False return False
return True return True
def _confirm_age(self): def _confirm_age(self):
age_form = { age_form = {
'next_url': '/', 'next_url': '/',
'action_confirm': 'Confirm', 'action_confirm': 'Confirm',
} }
request = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form)) req = compat_urllib_request.Request(self._AGE_URL, compat_urllib_parse.urlencode(age_form))
self.report_age_confirmation()
self._download_webpage(request, None, False, u'Unable to confirm age') self._download_webpage(
req, None,
note=u'Confirming age', errnote=u'Unable to confirm age')
return True return True
def _real_initialize(self): def _real_initialize(self):
@ -1736,11 +1734,14 @@ class YoutubeSearchIE(SearchInfoExtractor):
while (50 * pagenum) < limit: while (50 * pagenum) < limit:
result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1) result_url = self._API_URL % (compat_urllib_parse.quote_plus(query), (50*pagenum)+1)
data = self._download_webpage(result_url, u'query "%s"' % query, data_json = self._download_webpage(
u'Downloading page %s' % pagenum, u'Unable to download API page') result_url, video_id=u'query "%s"' % query,
api_response = json.loads(data)['data'] note=u'Downloading page %s' % (pagenum + 1),
errnote=u'Unable to download API page')
data = json.loads(data_json)
api_response = data['data']
if not 'items' in api_response: if 'items' not in api_response:
raise ExtractorError(u'[youtube] No video results') raise ExtractorError(u'[youtube] No video results')
new_ids = list(video['id'] for video in api_response['items']) new_ids = list(video['id'] for video in api_response['items'])