Tidy the changelog.

[youtubedl] / youtube_dl / downloader / f4m.py

diff --git a/youtube_dl/downloader/f4m.py b/youtube_dl/downloader/f4m.py
index b607f6485ae7b0a064e55634ef8ca33937cd857d..0e7a1c20075499e58b977da4154ce287b144f958 100644 (file)
--- a/youtube_dl/downloader/f4m.py
+++ b/youtube_dl/downloader/f4m.py
@@ -9,10 +9,12 @@ import xml.etree.ElementTree as etree
 
 from .common import FileDownloader
 from .http import HttpFD
 
 from .common import FileDownloader
 from .http import HttpFD

+from ..compat import (
+    compat_urlparse,
+)

 from ..utils import (
     struct_pack,
     struct_unpack,
 from ..utils import (
     struct_pack,
     struct_unpack,

-    compat_urlparse,

     format_bytes,
     encodeFilename,
     sanitize_open,
     format_bytes,
     encodeFilename,
     sanitize_open,


@@ -55,7 +57,7 @@ class FlvReader(io.BytesIO):
         if size == 1:
             real_size = self.read_unsigned_long_long()
             header_end = 16
         if size == 1:
             real_size = self.read_unsigned_long_long()
             header_end = 16

-        return real_size, box_type, self.read(real_size-header_end)
+        return real_size, box_type, self.read(real_size - header_end)

 
     def read_asrt(self):
         # version
 
     def read_asrt(self):
         # version


@@ -175,34 +177,43 @@ def build_fragments_list(boot_info):
     """ Return a list of (segment, fragment) for each fragment in the video """
     res = []
     segment_run_table = boot_info['segments'][0]
     """ Return a list of (segment, fragment) for each fragment in the video """
     res = []
     segment_run_table = boot_info['segments'][0]

-    # I've only found videos with one segment
-    segment_run_entry = segment_run_table['segment_run'][0]
-    n_frags = segment_run_entry[1]

     fragment_run_entry_table = boot_info['fragments'][0]['fragments']
     first_frag_number = fragment_run_entry_table[0]['first']
     fragment_run_entry_table = boot_info['fragments'][0]['fragments']
     first_frag_number = fragment_run_entry_table[0]['first']

-    for (i, frag_number) in zip(range(1, n_frags+1), itertools.count(first_frag_number)):
-        res.append((1, frag_number))
+    fragments_counter = itertools.count(first_frag_number)
+    for segment, fragments_count in segment_run_table['segment_run']:
+        for _ in range(fragments_count):
+            res.append((segment, next(fragments_counter)))

     return res
 
 
     return res
 
 

-def write_flv_header(stream, metadata):
-    """Writes the FLV header and the metadata to stream"""
+def write_unsigned_int(stream, val):
+    stream.write(struct_pack('!I', val))
+
+
+def write_unsigned_int_24(stream, val):
+    stream.write(struct_pack('!I', val)[1:])
+
+
+def write_flv_header(stream):
+    """Writes the FLV header to stream"""

     # FLV header
     stream.write(b'FLV\x01')
     stream.write(b'\x05')
     stream.write(b'\x00\x00\x00\x09')
     # FLV header
     stream.write(b'FLV\x01')
     stream.write(b'\x05')
     stream.write(b'\x00\x00\x00\x09')

-    # FLV File body

     stream.write(b'\x00\x00\x00\x00')
     stream.write(b'\x00\x00\x00\x00')

-    # FLVTAG
-    # Script data
-    stream.write(b'\x12')
-    # Size of the metadata with 3 bytes
-    stream.write(struct_pack('!L', len(metadata))[1:])
-    stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
-    stream.write(metadata)
-    # Magic numbers extracted from the output files produced by AdobeHDS.php
-    #(https://github.com/K-S-V/Scripts)
-    stream.write(b'\x00\x00\x01\x73')
+
+
+def write_metadata_tag(stream, metadata):
+    """Writes optional metadata tag to stream"""
+    SCRIPT_TAG = b'\x12'
+    FLV_TAG_HEADER_LEN = 11
+
+    if metadata:
+        stream.write(SCRIPT_TAG)
+        write_unsigned_int_24(stream, len(metadata))
+        stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
+        stream.write(metadata)
+        write_unsigned_int(stream, FLV_TAG_HEADER_LEN + len(metadata))

 
 
 def _add_ns(prop):
 
 
 def _add_ns(prop):


@@ -219,22 +230,43 @@ class F4mFD(FileDownloader):
     A downloader for f4m manifests or AdobeHDS.
     """
 
     A downloader for f4m manifests or AdobeHDS.
     """
 

+    def _get_unencrypted_media(self, doc):
+        media = doc.findall(_add_ns('media'))
+        if not media:
+            self.report_error('No media found')
+        for e in (doc.findall(_add_ns('drmAdditionalHeader')) +
+                  doc.findall(_add_ns('drmAdditionalHeaderSet'))):
+            # If id attribute is missing it's valid for all media nodes
+            # without drmAdditionalHeaderId or drmAdditionalHeaderSetId attribute
+            if 'id' not in e.attrib:
+                self.report_error('Missing ID in f4m DRM')
+        media = list(filter(lambda e: 'drmAdditionalHeaderId' not in e.attrib and
+                                      'drmAdditionalHeaderSetId' not in e.attrib,
+                            media))
+        if not media:
+            self.report_error('Unsupported DRM')
+        return media
+

     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
         self.to_screen('[download] Downloading f4m manifest')
         manifest = self.ydl.urlopen(man_url).read()
         self.report_destination(filename)
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
         self.to_screen('[download] Downloading f4m manifest')
         manifest = self.ydl.urlopen(man_url).read()
         self.report_destination(filename)

-        http_dl = HttpQuietDownloader(self.ydl,
+        http_dl = HttpQuietDownloader(
+            self.ydl,

             {
                 'continuedl': True,
                 'quiet': True,
                 'noprogress': True,
             {
                 'continuedl': True,
                 'quiet': True,
                 'noprogress': True,

+                'ratelimit': self.params.get('ratelimit', None),

                 'test': self.params.get('test', False),
                 'test': self.params.get('test', False),

-            })
+            }
+        )

 
         doc = etree.fromstring(manifest)
 
         doc = etree.fromstring(manifest)

-        formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
+        formats = [(int(f.attrib.get('bitrate', -1)), f)
+                   for f in self._get_unencrypted_media(doc)]

         if requested_bitrate is None:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])
         if requested_bitrate is None:
             # get the best format
             formats = sorted(formats, key=lambda f: f[0])


@@ -251,7 +283,11 @@ class F4mFD(FileDownloader):
             bootstrap = self.ydl.urlopen(bootstrap_url).read()
         else:
             bootstrap = base64.b64decode(bootstrap_node.text)
             bootstrap = self.ydl.urlopen(bootstrap_url).read()
         else:
             bootstrap = base64.b64decode(bootstrap_node.text)

-        metadata = base64.b64decode(media.find(_add_ns('metadata')).text)
+        metadata_node = media.find(_add_ns('metadata'))
+        if metadata_node is not None:
+            metadata = base64.b64decode(metadata_node.text)
+        else:
+            metadata = None

         boot_info = read_bootstrap_info(bootstrap)
 
         fragments_list = build_fragments_list(boot_info)
         boot_info = read_bootstrap_info(bootstrap)
 
         fragments_list = build_fragments_list(boot_info)


@@ -264,7 +300,8 @@ class F4mFD(FileDownloader):
 
         tmpfilename = self.temp_name(filename)
         (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')
 
         tmpfilename = self.temp_name(filename)
         (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')

-        write_flv_header(dest_stream, metadata)
+        write_flv_header(dest_stream)
+        write_metadata_tag(dest_stream, metadata)

 
         # This dict stores the download progress, it's updated by the progress
         # hook
 
         # This dict stores the download progress, it's updated by the progress
         # hook


@@ -277,7 +314,7 @@ class F4mFD(FileDownloader):
         def frag_progress_hook(status):
             frag_total_bytes = status.get('total_bytes', 0)
             estimated_size = (state['downloaded_bytes'] +
         def frag_progress_hook(status):
             frag_total_bytes = status.get('total_bytes', 0)
             estimated_size = (state['downloaded_bytes'] +

-                (total_frags - state['frag_counter']) * frag_total_bytes)
+                              (total_frags - state['frag_counter']) * frag_total_bytes)

             if status['status'] == 'finished':
                 state['downloaded_bytes'] += frag_total_bytes
                 state['frag_counter'] += 1
             if status['status'] == 'finished':
                 state['downloaded_bytes'] += frag_total_bytes
                 state['frag_counter'] += 1


@@ -287,13 +324,13 @@ class F4mFD(FileDownloader):
                 frag_downloaded_bytes = status['downloaded_bytes']
                 byte_counter = state['downloaded_bytes'] + frag_downloaded_bytes
                 frag_progress = self.calc_percent(frag_downloaded_bytes,
                 frag_downloaded_bytes = status['downloaded_bytes']
                 byte_counter = state['downloaded_bytes'] + frag_downloaded_bytes
                 frag_progress = self.calc_percent(frag_downloaded_bytes,

-                    frag_total_bytes)
+                                                  frag_total_bytes)

                 progress = self.calc_percent(state['frag_counter'], total_frags)
                 progress += frag_progress / float(total_frags)
 
             eta = self.calc_eta(start, time.time(), estimated_size, byte_counter)
             self.report_progress(progress, format_bytes(estimated_size),
                 progress = self.calc_percent(state['frag_counter'], total_frags)
                 progress += frag_progress / float(total_frags)
 
             eta = self.calc_eta(start, time.time(), estimated_size, byte_counter)
             self.report_progress(progress, format_bytes(estimated_size),

-                status.get('speed'), eta)
+                                 status.get('speed'), eta)

         http_dl.add_progress_hook(frag_progress_hook)
 
         frags_filenames = []
         http_dl.add_progress_hook(frag_progress_hook)
 
         frags_filenames = []