From 9d334aedc8e7e0497991cf13f40df9625efd8497 Mon Sep 17 00:00:00 2001
From: Mathias Rav <rav@cs.au.dk>
Date: Sat, 3 Jan 2015 09:05:34 +0100
Subject: [PATCH] [WatchTvSeries] Add new extractor for watch-tv-series.to

---
 youtube_dl/extractor/__init__.py        |   4 +
 youtube_dl/extractor/watch_tv_series.py | 108 ++++++++++++++++++++++++
 2 files changed, 112 insertions(+)
 create mode 100644 youtube_dl/extractor/watch_tv_series.py

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index 9ccd1b32e..a2ffdb613 100644
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -504,6 +504,10 @@ from .vulture import VultureIE
 from .walla import WallaIE
 from .washingtonpost import WashingtonPostIE
 from .wat import WatIE
+from .watch_tv_series import (
+    WatchTvSeriesSeasonIE,
+    WatchTvSeriesEpisodeIE,
+)
 from .wayofthemaster import WayOfTheMasterIE
 from .wdr import (
     WDRIE,
diff --git a/youtube_dl/extractor/watch_tv_series.py b/youtube_dl/extractor/watch_tv_series.py
new file mode 100644
index 000000000..3d18fc2dd
--- /dev/null
+++ b/youtube_dl/extractor/watch_tv_series.py
@@ -0,0 +1,108 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import clean_html, ExtractorError
+
+try:
+    import microdata
+except ImportError:
+    microdata = None
+
+
+class WatchTvSeriesSeasonIE(InfoExtractor):
+    domain_base = 'http://watch-tv-series.to/'
+    _VALID_URL = domain_base + r'(?P<path>season-(?P<season>\d+)/[a-z]+)'
+
+    def extract_microdata(self):
+        item, = microdata.get_items(self.content)
+        name = item.name
+        playlist_title = '%s - Season %s' % (name, self.season)
+        playlist_description = item.description
+
+        season = item.season
+        episodes = sorted(
+            season.get_all('episode'), key=lambda ep: ep.datepublished)
+
+        entries = [
+            self.url_result(self.domain_base + ep.url.lstrip('/'),
+                            'WatchTvSeriesEpisode')
+            for ep in episodes
+        ]
+
+        return self.playlist_result(
+            entries, playlist_id=item.url, playlist_title=playlist_title,
+            playlist_description=playlist_description)
+
+    def extract_strings(self):
+        name = self._search_regex(
+            r'<span itemprop="name">([^<]+)</span>',
+            self.content, 'show name')
+
+        playlist_title = '%s - Season %s' % (name, self.season)
+
+        data = {}
+        entries = []
+
+        # This parsing is sensitive to the order of HTML attributes in the
+        # content of the page, but it works for now.
+        itemprop_regex = r'itemprop="([a-z]+)"(?: content="([^"]+)"|>([^<]+))'
+
+        for mobj in re.finditer(itemprop_regex, self.content):
+            key = mobj.group(1)
+            value = mobj.group(2) or mobj.group(3)
+            value = clean_html(value).strip()
+            data[key] = value
+            if key == 'name' and data.get('url', '').startswith('/episode'):
+                url = self.domain_base + data['url'].lstrip('/')
+                entries.append(self.url_result(url, 'WatchTvSeriesEpisode'))
+
+        # Episodes are displayed with the latest first,
+        # but we want to retrieve the earliest first.
+        entries.reverse()
+
+        return self.playlist_result(
+            entries, playlist_id=self.path, playlist_title=playlist_title)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        self.path = mobj.group('path')
+        self.season = mobj.group('season')
+        self.content = self._download_webpage(url, self.path)
+        if microdata:
+            return self.extract_microdata()
+        else:
+            return self.extract_strings()
+
+
+class WatchTvSeriesEpisodeIE(InfoExtractor):
+    domain_base = 'http://watch-tv-series.to/'
+    _VALID_URL = 'http://watch-tv-series.to/(?P<path>episode/[a-z_0-9]+.html)'
+
+    def _real_extract(self, url):
+        # There is microdata on this page, but we don't need it,
+        # since we will return a url_result anyway.
+
+        mobj = re.match(self._VALID_URL, url)
+        path = mobj.group('path')
+
+        content = self._download_webpage(url, path)
+
+        movshare_regex = (
+            r'href="/([^"]+)" class="buttonlink" title="movshare.net"')
+
+        mobj = re.search(movshare_regex, content)
+        if mobj is None:
+            raise ExtractorError(
+                'Unable to extract movshare.net link. ' +
+                'No other video sites are supported.',
+                expected=True)
+
+        external_link = self._download_webpage(
+            self.domain_base + mobj.group(1), path + '[external]')
+
+        mobj = re.search(
+            r'http://www.movshare.net/video/[a-z0-9]+', external_link)
+
+        return self.url_result(mobj.group(0))