Merge branch 'lecture2go' of https://github.com/nichdu/youtube-dl into nichdu-lecture2go

[youtube-dl] / youtube_dl / extractor / ceskatelevize.py

diff --git a/youtube_dl/extractor/ceskatelevize.py b/youtube_dl/extractor/ceskatelevize.py
index 345ac5e859d7e6b66af789f99beabdf9f1c36cc7..dda583680a03ba3cb420beb74a99af2ec60cbc83 100644 (file)
--- a/youtube_dl/extractor/ceskatelevize.py
+++ b/youtube_dl/extractor/ceskatelevize.py
@@ -3,20 +3,20 @@ from __future__ import unicode_literals
 
 import re
 
-from .subtitles import SubtitlesInfoExtractor
+from .common import InfoExtractor
 from ..compat import (
     compat_urllib_request,
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_parse_urlparse,
 )
 from ..utils import (
     ExtractorError,
     float_or_none,
-    HEADRequest,
 )
 
 
-class CeskaTelevizeIE(SubtitlesInfoExtractor):
+class CeskaTelevizeIE(InfoExtractor):
     _VALID_URL = r'https?://www\.ceskatelevize\.cz/(porady|ivysilani)/(.+/)?(?P<id>[^?#]+)'
 
     _TESTS = [
@@ -89,7 +89,7 @@ class CeskaTelevizeIE(SubtitlesInfoExtractor):
         if playlist_url == 'error_region':
             raise ExtractorError(NOT_AVAILABLE_STRING, expected=True)
 
-        req = compat_urllib_request.Request(compat_urllib_parse.unquote(playlist_url))
+        req = compat_urllib_request.Request(compat_urllib_parse_unquote(playlist_url))
         req.add_header('Referer', url)
 
         playlist = self._download_json(req, video_id)
@@ -105,14 +105,10 @@ class CeskaTelevizeIE(SubtitlesInfoExtractor):
         duration = float_or_none(item.get('duration'))
         thumbnail = item.get('previewImageUrl')
 
-        # subtitles
-        subtitles = self.extract_subtitles(video_id, webpage)
-
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, webpage)
-            return
-
-        subtitles = self._fix_subtitles(self.extract_subtitles(video_id, webpage))
+        subtitles = {}
+        subs = item.get('subtitles')
+        if subs:
+            subtitles = self.extract_subtitles(episode_id, subs)
 
         return {
             'id': episode_id,
@@ -124,13 +120,23 @@ class CeskaTelevizeIE(SubtitlesInfoExtractor):
             'subtitles': subtitles,
         }
 
-    def _fix_subtitles(self, subtitles):
-        """ Convert milisecond-based subtitles to SRT """
-        if subtitles is None:
-            return subtitles  # subtitles not requested
+    def _get_subtitles(self, episode_id, subs):
+        original_subtitles = self._download_webpage(
+            subs[0]['url'], episode_id, 'Downloading subtitles')
+        srt_subs = self._fix_subtitles(original_subtitles)
+        return {
+            'cs': [{
+                'ext': 'srt',
+                'data': srt_subs,
+            }]
+        }
+
+    @staticmethod
+    def _fix_subtitles(subtitles):
+        """ Convert millisecond-based subtitles to SRT """
 
         def _msectotimecode(msec):
-            """ Helper utility to convert miliseconds to timecode """
+            """ Helper utility to convert milliseconds to timecode """
             components = []
             for divider in [1000, 60, 60, 100]:
                 components.append(msec % divider)
@@ -139,27 +145,12 @@ class CeskaTelevizeIE(SubtitlesInfoExtractor):
 
         def _fix_subtitle(subtitle):
             for line in subtitle.splitlines():
-                m = re.match(r"^ *([0-9]+); *([0-9]+) +([0-9]+) *$", line)
+                m = re.match(r"^\s*([0-9]+);\s*([0-9]+)\s+([0-9]+)\s*$", line)
                 if m:
                     yield m.group(1)
                     start, stop = (_msectotimecode(int(t)) for t in m.groups()[1:])
-                    yield "{} --> {}".format(start, stop)
+                    yield "{0} --> {1}".format(start, stop)
                 else:
                     yield line
 
-        fixed_subtitles = {}
-        for k, v in subtitles.items():
-            fixed_subtitles[k] = "\r\n".join(_fix_subtitle(v))
-        return fixed_subtitles
-
-    def _get_available_subtitles(self, video_id, webpage):
-        video_id = video_id.partition('-')[0]
-        url = 'http://imgct.ceskatelevize.cz/cache/data/ivysilani/' \
-              'subtitles/{}/{}/sub.txt'.format(video_id[:3], video_id)
-        req = HEADRequest(url)
-        sub = self._request_webpage(
-            req, video_id,
-            note="Checking subtitles",
-            errnote="No subtitles found",
-            fatal=False)
-        return {'cs': url} if sub else {}
+        return "\r\n".join(_fix_subtitle(subtitles))