Merge branch 'subtitles-rework'

[youtube-dl] / youtube_dl / YoutubeDL.py

diff --git a/youtube_dl/YoutubeDL.py b/youtube_dl/YoutubeDL.py
index e665e3d53bb4ca1f557074f4c46179bfaf65de5c..76fc394bcff44f30ae6fa383ea54621a654a0864 100755 (executable)
--- a/youtube_dl/YoutubeDL.py
+++ b/youtube_dl/YoutubeDL.py
@@ -199,18 +199,25 @@ class YoutubeDL(object):
                        postprocessor.
     progress_hooks:    A list of functions that get called on download
                        progress, with a dictionary with the entries
                        postprocessor.
     progress_hooks:    A list of functions that get called on download
                        progress, with a dictionary with the entries

-                       * status: One of "downloading" and "finished".
+                       * status: One of "downloading", "error", or "finished".

                                  Check this first and ignore unknown values.
 
                                  Check this first and ignore unknown values.
 

-                       If status is one of "downloading" or "finished", the
+                       If status is one of "downloading", or "finished", the

                        following properties may also be present:
                        * filename: The final filename (always present)
                        following properties may also be present:
                        * filename: The final filename (always present)

+                       * tmpfilename: The filename we're currently writing to

                        * downloaded_bytes: Bytes on disk
                        * total_bytes: Size of the whole file, None if unknown
                        * downloaded_bytes: Bytes on disk
                        * total_bytes: Size of the whole file, None if unknown

-                       * tmpfilename: The filename we're currently writing to
+                       * total_bytes_estimate: Guess of the eventual file size,
+                                               None if unavailable.
+                       * elapsed: The number of seconds since download started.

                        * eta: The estimated time in seconds, None if unknown
                        * speed: The download speed in bytes/second, None if
                                 unknown
                        * eta: The estimated time in seconds, None if unknown
                        * speed: The download speed in bytes/second, None if
                                 unknown

+                       * fragment_index: The counter of the currently
+                                         downloaded video fragment.
+                       * fragment_count: The number of fragments (= individual
+                                         files that will be merged)

 
                        Progress hooks are guaranteed to be called at least once
                        (with status "finished") if the download is successful.
 
                        Progress hooks are guaranteed to be called at least once
                        (with status "finished") if the download is successful.


@@ -225,7 +232,6 @@ class YoutubeDL(object):
     call_home:         Boolean, true iff we are allowed to contact the
                        youtube-dl servers for debugging.
     sleep_interval:    Number of seconds to sleep before each download.
     call_home:         Boolean, true iff we are allowed to contact the
                        youtube-dl servers for debugging.
     sleep_interval:    Number of seconds to sleep before each download.

-    external_downloader:  Executable of the external downloader to call.

     listformats:       Print an overview of available video formats and exit.
     list_thumbnails:   Print a table of all thumbnails and exit.
     match_filter:      A function that gets called with the info_dict of
     listformats:       Print an overview of available video formats and exit.
     list_thumbnails:   Print a table of all thumbnails and exit.
     match_filter:      A function that gets called with the info_dict of


@@ -235,6 +241,10 @@ class YoutubeDL(object):
                        match_filter_func in utils.py is one example for this.
     no_color:          Do not emit color codes in output.
 
                        match_filter_func in utils.py is one example for this.
     no_color:          Do not emit color codes in output.
 

+    The following options determine which downloader is picked:
+    external_downloader: Executable of the external downloader to call.
+                       None or unset for standard (built-in) downloader.
+    hls_prefer_native: Use the native HLS downloader instead of ffmpeg/avconv.

 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:


@@ -298,8 +308,8 @@ class YoutubeDL(object):
                     raise
 
         if (sys.version_info >= (3,) and sys.platform != 'win32' and
                     raise
 
         if (sys.version_info >= (3,) and sys.platform != 'win32' and

-                sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968']
-                and not params.get('restrictfilenames', False)):
+                sys.getfilesystemencoding() in ['ascii', 'ANSI_X3.4-1968'] and
+                not params.get('restrictfilenames', False)):

             # On Python 3, the Unicode filesystem API will throw errors (#1474)
             self.report_warning(
                 'Assuming --restrict-filenames since file system encoding '
             # On Python 3, the Unicode filesystem API will throw errors (#1474)
             self.report_warning(
                 'Assuming --restrict-filenames since file system encoding '


@@ -951,30 +961,9 @@ class YoutubeDL(object):
         return res
 
     def _calc_cookies(self, info_dict):
         return res
 
     def _calc_cookies(self, info_dict):

-        class _PseudoRequest(object):
-            def __init__(self, url):
-                self.url = url
-                self.headers = {}
-                self.unverifiable = False
-
-            def add_unredirected_header(self, k, v):
-                self.headers[k] = v
-
-            def get_full_url(self):
-                return self.url
-
-            def is_unverifiable(self):
-                return self.unverifiable
-
-            def has_header(self, h):
-                return h in self.headers
-
-            def get_header(self, h, default=None):
-                return self.headers.get(h, default)
-
-        pr = _PseudoRequest(info_dict['url'])
+        pr = compat_urllib_request.Request(info_dict['url'])

         self.cookiejar.add_cookie_header(pr)
         self.cookiejar.add_cookie_header(pr)

-        return pr.headers.get('Cookie')
+        return pr.get_header('Cookie')

 
     def process_video_result(self, info_dict, download=True):
         assert info_dict.get('_type', 'video') == 'video'
 
     def process_video_result(self, info_dict, download=True):
         assert info_dict.get('_type', 'video') == 'video'


@@ -1020,9 +1009,13 @@ class YoutubeDL(object):
             info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
 
         if self.params.get('listsubtitles', False):
             info_dict['upload_date'] = upload_date.strftime('%Y%m%d')
 
         if self.params.get('listsubtitles', False):

-            self.list_subtitles(info_dict['id'], info_dict.get('subtitles'))
+            if 'automatic_captions' in info_dict:
+                self.list_subtitles(info_dict['id'], info_dict.get('automatic_captions'), 'automatic captions')
+            self.list_subtitles(info_dict['id'], info_dict.get('subtitles'), 'subtitles')

             return
             return

-        info_dict['subtitles'] = self.process_subtitles(info_dict['id'], info_dict.get('subtitles'))
+        info_dict['requested_subtitles'] = self.process_subtitles(
+            info_dict['id'], info_dict.get('subtitles'),
+            info_dict.get('automatic_captions'))

 
         # This extractors handle format selection themselves
         if info_dict['extractor'] in ['Youku']:
 
         # This extractors handle format selection themselves
         if info_dict['extractor'] in ['Youku']:


@@ -1152,10 +1145,20 @@ class YoutubeDL(object):
         info_dict.update(formats_to_download[-1])
         return info_dict
 
         info_dict.update(formats_to_download[-1])
         return info_dict
 

-    def process_subtitles(self, video_id, available_subs):
+    def process_subtitles(self, video_id, normal_subtitles, automatic_captions):

         """Select the requested subtitles and their format"""
         """Select the requested subtitles and their format"""

-        if not available_subs:
-            return available_subs
+        available_subs = {}
+        if normal_subtitles and self.params.get('writesubtitles'):
+            available_subs.update(normal_subtitles)
+        if automatic_captions and self.params.get('writeautomaticsub'):
+            for lang, cap_info in automatic_captions.items():
+                if lang not in available_subs:
+                    available_subs[lang] = cap_info
+
+        if (not self.params.get('writesubtitles') and not
+                self.params.get('writeautomaticsub') or not
+                available_subs):
+            return None

 
         if self.params.get('allsubtitles', False):
             requested_langs = available_subs.keys()
 
         if self.params.get('allsubtitles', False):
             requested_langs = available_subs.keys()


@@ -1175,14 +1178,6 @@ class YoutubeDL(object):
             if formats is None:
                 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
                 continue
             if formats is None:
                 self.report_warning('%s subtitles not available for %s' % (lang, video_id))
                 continue

-            if isinstance(formats, compat_str):
-                # TODO: convert all IE with subtitles support to the new format
-                # and remove this
-                subs[lang] = {
-                    'ext': formats_preference[0],
-                    'data': formats,
-                }
-                continue

             for ext in formats_preference:
                 if ext == 'best':
                     f = formats[-1]
             for ext in formats_preference:
                 if ext == 'best':
                     f = formats[-1]


@@ -1301,10 +1296,10 @@ class YoutubeDL(object):
         subtitles_are_requested = any([self.params.get('writesubtitles', False),
                                        self.params.get('writeautomaticsub')])
 
         subtitles_are_requested = any([self.params.get('writesubtitles', False),
                                        self.params.get('writeautomaticsub')])
 

-        if subtitles_are_requested and 'subtitles' in info_dict and info_dict['subtitles']:
+        if subtitles_are_requested and info_dict.get('requested_subtitles'):

             # subtitles download errors are already managed as troubles in relevant IE
             # that way it will silently go on when used with unsupporting IE
             # subtitles download errors are already managed as troubles in relevant IE
             # that way it will silently go on when used with unsupporting IE

-            subtitles = info_dict['subtitles']
+            subtitles = info_dict['requested_subtitles']

             for sub_lang, sub_info in subtitles.items():
                 sub_format = sub_info['ext']
                 if sub_info.get('data') is not None:
             for sub_lang, sub_info in subtitles.items():
                 sub_format = sub_info['ext']
                 if sub_info.get('data') is not None:


@@ -1357,7 +1352,7 @@ class YoutubeDL(object):
                     downloaded = []
                     success = True
                     merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))
                     downloaded = []
                     success = True
                     merger = FFmpegMergerPP(self, not self.params.get('keepvideo'))

-                    if not merger._executable:
+                    if not merger.available:

                         postprocessors = []
                         self.report_warning('You have requested multiple '
                                             'formats but ffmpeg or avconv are not installed.'
                         postprocessors = []
                         self.report_warning('You have requested multiple '
                                             'formats but ffmpeg or avconv are not installed.'


@@ -1436,8 +1431,8 @@ class YoutubeDL(object):
         """Download a given list of URLs."""
         outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
         if (len(url_list) > 1 and
         """Download a given list of URLs."""
         outtmpl = self.params.get('outtmpl', DEFAULT_OUTTMPL)
         if (len(url_list) > 1 and

-                '%' not in outtmpl
-                and self.params.get('max_downloads') != 1):
+                '%' not in outtmpl and
+                self.params.get('max_downloads') != 1):

             raise SameFileError(outtmpl)
 
         for url in url_list:
             raise SameFileError(outtmpl)
 
         for url in url_list:


@@ -1604,29 +1599,18 @@ class YoutubeDL(object):
         return res
 
     def list_formats(self, info_dict):
         return res
 
     def list_formats(self, info_dict):

-        def line(format, idlen=20):
-            return (('%-' + compat_str(idlen + 1) + 's%-10s%-12s%s') % (
-                format['format_id'],
-                format['ext'],
-                self.format_resolution(format),
-                self._format_note(format),
-            ))
-

         formats = info_dict.get('formats', [info_dict])
         formats = info_dict.get('formats', [info_dict])

-        idlen = max(len('format code'),
-                    max(len(f['format_id']) for f in formats))
-        formats_s = [
-            line(f, idlen) for f in formats
+        table = [
+            [f['format_id'], f['ext'], self.format_resolution(f), self._format_note(f)]
+            for f in formats

             if f.get('preference') is None or f['preference'] >= -1000]
         if len(formats) > 1:
             if f.get('preference') is None or f['preference'] >= -1000]
         if len(formats) > 1:

-            formats_s[-1] += (' ' if self._format_note(formats[-1]) else '') + '(best)'
+            table[-1][-1] += (' ' if table[-1][-1] else '') + '(best)'

 
 

-        header_line = line({
-            'format_id': 'format code', 'ext': 'extension',
-            'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
+        header_line = ['format code', 'extension', 'resolution', 'note']

         self.to_screen(
         self.to_screen(

-            '[info] Available formats for %s:\n%s\n%s' %
-            (info_dict['id'], header_line, '\n'.join(formats_s)))
+            '[info] Available formats for %s:\n%s' %
+            (info_dict['id'], render_table(header_line, table)))

 
     def list_thumbnails(self, info_dict):
         thumbnails = info_dict.get('thumbnails')
 
     def list_thumbnails(self, info_dict):
         thumbnails = info_dict.get('thumbnails')


@@ -1645,17 +1629,16 @@ class YoutubeDL(object):
             ['ID', 'width', 'height', 'URL'],
             [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
 
             ['ID', 'width', 'height', 'URL'],
             [[t['id'], t.get('width', 'unknown'), t.get('height', 'unknown'), t['url']] for t in thumbnails]))
 

-    def list_subtitles(self, video_id, subtitles):
+    def list_subtitles(self, video_id, subtitles, name='subtitles'):

         if not subtitles:
         if not subtitles:

-            self.to_screen('%s has no subtitles' % video_id)
+            self.to_screen('%s has no %s' % (video_id, name))

             return
             return

-        header_line = 'Language    formats'
-        sub_lines = [
-            '%-12s%s' % (lang, ', '.join(f['ext'] for f in reversed(formats)))
-            for lang, formats in subtitles.items()]

         self.to_screen(
         self.to_screen(

-            'Available subtitles for %s:\n%s\n%s' %
-            (video_id, header_line, '\n'.join(sub_lines)))
+            'Available %s for %s:' % (name, video_id))
+        self.to_screen(render_table(
+            ['Language', 'formats'],
+            [[lang, ', '.join(f['ext'] for f in reversed(formats))]
+                for lang, formats in subtitles.items()]))

 
     def urlopen(self, req):
         """ Start an HTTP download """
 
     def urlopen(self, req):
         """ Start an HTTP download """


@@ -1718,7 +1701,7 @@ class YoutubeDL(object):
         self._write_string('[debug] Python version %s - %s\n' % (
             platform.python_version(), platform_name()))
 
         self._write_string('[debug] Python version %s - %s\n' % (
             platform.python_version(), platform_name()))
 

-        exe_versions = FFmpegPostProcessor.get_versions()
+        exe_versions = FFmpegPostProcessor.get_versions(self)

         exe_versions['rtmpdump'] = rtmpdump_version()
         exe_str = ', '.join(
             '%s %s' % (exe, v)
         exe_versions['rtmpdump'] = rtmpdump_version()
         exe_str = ', '.join(
             '%s %s' % (exe, v)