aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/anvato.py4
-rw-r--r--youtube_dl/extractor/brightcove.py5
-rw-r--r--youtube_dl/extractor/common.py97
-rw-r--r--youtube_dl/extractor/dplay.py4
-rw-r--r--youtube_dl/extractor/go.py2
-rw-r--r--youtube_dl/extractor/limelight.py4
-rw-r--r--youtube_dl/extractor/tvplay.py6
7 files changed, 98 insertions, 24 deletions
diff --git a/youtube_dl/extractor/anvato.py b/youtube_dl/extractor/anvato.py
index 7a29cd2c6..f6a78eb5d 100644
--- a/youtube_dl/extractor/anvato.py
+++ b/youtube_dl/extractor/anvato.py
@@ -277,7 +277,9 @@ class AnvatoIE(InfoExtractor):
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
- self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
+ self._initialize_geo_bypass({
+ 'countries': smuggled_data.get('geo_countries'),
+ })
mobj = re.match(self._VALID_URL, url)
access_key, video_id = mobj.group('access_key_or_mcp', 'id')
diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
index 0e4eaef65..ab62e54d6 100644
--- a/youtube_dl/extractor/brightcove.py
+++ b/youtube_dl/extractor/brightcove.py
@@ -669,7 +669,10 @@ class BrightcoveNewIE(AdobePassIE):
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
- self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
+ self._initialize_geo_bypass({
+ 'countries': smuggled_data.get('geo_countries'),
+ 'ip_blocks': smuggled_data.get('geo_ip_blocks'),
+ })
account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index a9939b0fd..3ef5af13c 100644
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -346,6 +346,11 @@ class InfoExtractor(object):
geo restriction bypass mechanism right away in order to bypass
geo restriction, of course, if the mechanism is not disabled. (experimental)
+ _GEO_IP_BLOCKS attribute may contain a list of presumably geo unrestricted
+ IP blocks in CIDR notation for this extractor. One of these IP blocks
+ will be used by geo restriction bypass mechanism similarly
+ to _GEO_COUNTRIES. (experimental)
+
NB: both these geo attributes are experimental and may change in future
or be completely removed.
@@ -358,6 +363,7 @@ class InfoExtractor(object):
_x_forwarded_for_ip = None
_GEO_BYPASS = True
_GEO_COUNTRIES = None
+ _GEO_IP_BLOCKS = None
_WORKING = True
def __init__(self, downloader=None):
@@ -392,12 +398,15 @@ class InfoExtractor(object):
def initialize(self):
"""Initializes an instance (authentication, etc)."""
- self._initialize_geo_bypass(self._GEO_COUNTRIES)
+ self._initialize_geo_bypass({
+ 'countries': self._GEO_COUNTRIES,
+ 'ip_blocks': self._GEO_IP_BLOCKS,
+ })
if not self._ready:
self._real_initialize()
self._ready = True
- def _initialize_geo_bypass(self, countries):
+ def _initialize_geo_bypass(self, geo_bypass_context):
"""
Initialize geo restriction bypass mechanism.
@@ -408,28 +417,82 @@ class InfoExtractor(object):
HTTP requests.
This method will be used for initial geo bypass mechanism initialization
- during the instance initialization with _GEO_COUNTRIES.
+ during the instance initialization with _GEO_COUNTRIES and
+ _GEO_IP_BLOCKS.
- You may also manually call it from extractor's code if geo countries
+ You may also manually call it from extractor's code if geo bypass
information is not available beforehand (e.g. obtained during
- extraction) or due to some another reason.
+ extraction) or due to some other reason. In this case you should pass
+ this information in geo bypass context passed as first argument. It may
+ contain following fields:
+
+ countries: List of geo unrestricted countries (similar
+ to _GEO_COUNTRIES)
+ ip_blocks: List of geo unrestricted IP blocks in CIDR notation
+ (similar to _GEO_IP_BLOCKS)
+
"""
if not self._x_forwarded_for_ip:
- country_code = self._downloader.params.get('geo_bypass_country', None)
- # If there is no explicit country for geo bypass specified and
- # the extractor is known to be geo restricted let's fake IP
- # as X-Forwarded-For right away.
- if (not country_code and
- self._GEO_BYPASS and
- self._downloader.params.get('geo_bypass', True) and
- countries):
- country_code = random.choice(countries)
- if country_code:
- self._x_forwarded_for_ip = GeoUtils.random_ipv4(country_code)
+
+ # Geo bypass mechanism is explicitly disabled by user
+ if not self._downloader.params.get('geo_bypass', True):
+ return
+
+ if not geo_bypass_context:
+ geo_bypass_context = {}
+
+ # Backward compatibility: previously _initialize_geo_bypass
+ # expected a list of countries, some 3rd party code may still use
+ # it this way
+ if isinstance(geo_bypass_context, (list, tuple)):
+ geo_bypass_context = {
+ 'countries': geo_bypass_context,
+ }
+
+ # The whole point of geo bypass mechanism is to fake IP
+ # as X-Forwarded-For HTTP header based on some IP block or
+ # country code.
+
+ # Path 1: bypassing based on IP block in CIDR notation
+
+ # Explicit IP block specified by user, use it right away
+ # regardless of whether extractor is geo bypassable or not
+ ip_block = self._downloader.params.get('geo_bypass_ip_block', None)
+
+ # Otherwise use random IP block from geo bypass context but only
+ # if extractor is known as geo bypassable
+ if not ip_block:
+ ip_blocks = geo_bypass_context.get('ip_blocks')
+ if self._GEO_BYPASS and ip_blocks:
+ ip_block = random.choice(ip_blocks)
+
+ if ip_block:
+ self._x_forwarded_for_ip = GeoUtils.random_ipv4(ip_block)
+ if self._downloader.params.get('verbose', False):
+ self._downloader.to_screen(
+ '[debug] Using fake IP %s as X-Forwarded-For.'
+ % self._x_forwarded_for_ip)
+ return
+
+ # Path 2: bypassing based on country code
+
+ # Explicit country code specified by user, use it right away
+ # regardless of whether extractor is geo bypassable or not
+ country = self._downloader.params.get('geo_bypass_country', None)
+
+ # Otherwise use random country code from geo bypass context but
+ # only if extractor is known as geo bypassable
+ if not country:
+ countries = geo_bypass_context.get('countries')
+ if self._GEO_BYPASS and countries:
+ country = random.choice(countries)
+
+ if country:
+ self._x_forwarded_for_ip = GeoUtils.random_ipv4(country)
if self._downloader.params.get('verbose', False):
self._downloader.to_screen(
'[debug] Using fake IP %s (%s) as X-Forwarded-For.'
- % (self._x_forwarded_for_ip, country_code.upper()))
+ % (self._x_forwarded_for_ip, country.upper()))
def extract(self, url):
"""Extracts URL information and returns it in list of dicts."""
diff --git a/youtube_dl/extractor/dplay.py b/youtube_dl/extractor/dplay.py
index b73446773..8e0374320 100644
--- a/youtube_dl/extractor/dplay.py
+++ b/youtube_dl/extractor/dplay.py
@@ -102,7 +102,9 @@ class DPlayIE(InfoExtractor):
display_id = mobj.group('id')
domain = mobj.group('domain')
- self._initialize_geo_bypass([mobj.group('country').upper()])
+ self._initialize_geo_bypass({
+ 'countries': [mobj.group('country').upper()],
+ })
webpage = self._download_webpage(url, display_id)
diff --git a/youtube_dl/extractor/go.py b/youtube_dl/extractor/go.py
index 9c7b1bd37..e781405f2 100644
--- a/youtube_dl/extractor/go.py
+++ b/youtube_dl/extractor/go.py
@@ -123,7 +123,7 @@ class GoIE(AdobePassIE):
'adobe_requestor_id': requestor_id,
})
else:
- self._initialize_geo_bypass(['US'])
+ self._initialize_geo_bypass({'countries': ['US']})
entitlement = self._download_json(
'https://api.entitlement.watchabc.go.com/vp2/ws-secure/entitlement/2020/authorize.json',
video_id, data=urlencode_postdata(data))
diff --git a/youtube_dl/extractor/limelight.py b/youtube_dl/extractor/limelight.py
index 2803d7e8d..729d8de50 100644
--- a/youtube_dl/extractor/limelight.py
+++ b/youtube_dl/extractor/limelight.py
@@ -282,7 +282,9 @@ class LimelightMediaIE(LimelightBaseIE):
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
video_id = self._match_id(url)
- self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
+ self._initialize_geo_bypass({
+ 'countries': smuggled_data.get('geo_countries'),
+ })
pc, mobile, metadata = self._extract(
video_id, 'getPlaylistByMediaId',
diff --git a/youtube_dl/extractor/tvplay.py b/youtube_dl/extractor/tvplay.py
index 84597b55e..e09b5f804 100644
--- a/youtube_dl/extractor/tvplay.py
+++ b/youtube_dl/extractor/tvplay.py
@@ -227,14 +227,16 @@ class TVPlayIE(InfoExtractor):
def _real_extract(self, url):
url, smuggled_data = unsmuggle_url(url, {})
- self._initialize_geo_bypass(smuggled_data.get('geo_countries'))
+ self._initialize_geo_bypass({
+ 'countries': smuggled_data.get('geo_countries'),
+ })
video_id = self._match_id(url)
geo_country = self._search_regex(
r'https?://[^/]+\.([a-z]{2})', url,
'geo country', default=None)
if geo_country:
- self._initialize_geo_bypass([geo_country.upper()])
+ self._initialize_geo_bypass({'countries': [geo_country.upper()]})
video = self._download_json(
'http://playapi.mtgx.tv/v3/videos/%s' % video_id, video_id, 'Downloading video JSON')