implement ThisVid extractor

deobfuscates the video URL using a reverse engineered version of KVS
player's algorithm. This was tested against version 4.0.4, 5.0.1,
5.1.1.4 and 5.2.0.4 of the player and a warning will be issued if the
major version changes.

diff --git a/youtube_dlc/extractor/extractors.py b/youtube_dlc/extractor/extractors.py
index 666134d868956a876d611fc093e134b283b8350a..ee404f78d07d4192e0ae89cc1f92d7beeeb33a7b 100644 (file)
--- a/youtube_dlc/extractor/extractors.py
+++ b/youtube_dlc/extractor/extractors.py
@@ -1175,6 +1175,7 @@
 from .thisamericanlife import ThisAmericanLifeIE
 from .thisav import ThisAVIE
 from .thisoldhouse import ThisOldHouseIE
+from .thisvid import ThisVidIE
 from .threeqsdn import ThreeQSDNIE
 from .tiktok import TikTokIE
 from .tinypic import TinyPicIE

diff --git a/youtube_dlc/extractor/thisvid.py b/youtube_dlc/extractor/thisvid.py
new file mode 100644 (file)
index 0000000..f507e1b
--- /dev/null
+++ b/youtube_dlc/extractor/thisvid.py
@@ -0,0 +1,97 @@
+# coding: utf-8
+from __future__ import unicode_literals
+import re
+
+from .common import InfoExtractor
+
+
+class ThisVidIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?thisvid\.com/(?P<type>videos|embed)/(?P<id>[A-Za-z0-9-]+/?)'
+    _TESTS = [{
+        'url': 'https://thisvid.com/videos/french-boy-pantsed/',
+        'md5': '3397979512c682f6b85b3b04989df224',
+        'info_dict': {
+            'id': '2400174',
+            'ext': 'mp4',
+            'title': 'French Boy Pantsed',
+            'thumbnail': 'https://media.thisvid.com/contents/videos_screenshots/2400000/2400174/preview.mp4.jpg',
+            'age_limit': 18,
+        }
+    }, {
+        'url': 'https://thisvid.com/embed/2400174/',
+        'md5': '3397979512c682f6b85b3b04989df224',
+        'info_dict': {
+            'id': '2400174',
+            'ext': 'mp4',
+            'title': 'French Boy Pantsed',
+            'thumbnail': 'https://media.thisvid.com/contents/videos_screenshots/2400000/2400174/preview.mp4.jpg',
+            'age_limit': 18,
+        }
+    }]
+
+    def _real_extract(self, url):
+        main_id = self._match_id(url)
+        webpage = self._download_webpage(url, main_id)
+
+        # URL decryptor was reversed from version 4.0.4, later verified working with 5.2.0 and may change in the future.
+        kvs_version = self._html_search_regex(r'<script [^>]+?src="https://thisvid\.com/player/kt_player\.js\?v=(\d+(\.\d+)+)">', webpage, 'kvs_version', fatal=False)
+        if not kvs_version.startswith("5."):
+            self.report_warning("Major version change (" + kvs_version + ") in player engine--Download may fail.")
+
+        title = self._html_search_regex(r'<title>(?:Video: )?(.+?)(?: - (?:\w+ porn at )?ThisVid(?:.com| tube))?</title>', webpage, 'title')
+        # video_id, video_url and license_code from the 'flashvars' JSON object:
+        video_id = self._html_search_regex(r"video_id: '([0-9]+)',", webpage, 'video_id')
+        video_url = self._html_search_regex(r"video_url: '(function/0/.+?)',", webpage, 'video_url')
+        license_code = self._html_search_regex(r"license_code: '([0-9$]{16})',", webpage, 'license_code')
+        thumbnail = self._html_search_regex(r"preview_url: '((?:https?:)?//media.thisvid.com/.+?.jpg)',", webpage, 'thumbnail', fatal=False)
+        if thumbnail.startswith("//"):
+            thumbnail = "https:" + thumbnail
+        if (re.match(self._VALID_URL, url).group('type') == "videos"):
+            display_id = main_id
+        else:
+            display_id = self._search_regex(r'<link rel="canonical" href="' + self._VALID_URL + r'">', webpage, 'display_id', fatal=False),
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'url': getrealurl(video_url, license_code),
+            'thumbnail': thumbnail,
+            'age_limit': 18,
+        }
+
+
+def getrealurl(video_url, license_code):
+    urlparts = video_url.split('/')[2:]
+    license = getlicensetoken(license_code)
+    newmagic = urlparts[5][:32]
+
+    for o in range(len(newmagic) - 1, -1, -1):
+        new = ""
+        l = (o + sum([int(n) for n in license[o:]])) % 32
+
+        for i in range(0, len(newmagic)):
+            if i == o:
+                new += newmagic[l]
+            elif i == l:
+                new += newmagic[o]
+            else:
+                new += newmagic[i]
+        newmagic = new
+
+    urlparts[5] = newmagic + urlparts[5][32:]
+    return "/".join(urlparts)
+
+
+def getlicensetoken(license):
+    modlicense = license.replace("$", "").replace("0", "1")
+    center = int(len(modlicense) / 2)
+    fronthalf = int(modlicense[:center + 1])
+    backhalf = int(modlicense[center:])
+
+    modlicense = str(4 * abs(fronthalf - backhalf))
+    retval = ""
+    for o in range(0, center + 1):
+        for i in range(1, 5):
+            retval += str((int(license[o + i]) + int(modlicense[o])) % 10)
+    return retval