-#!/usr/bin/env python
+#!/usr/bin/env python3
#
# smooth-dl - download videos served using Smooth Streaming technology
#
-# Copyright (C) 2010 Antonio Ospite <ospite@studenti.unina.it>
+# Copyright (C) 2010-2016 Antonio Ospite <ao2@ao2.it>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
import re
import sys
import xml.etree.ElementTree as etree
-import urllib2
+import urllib.request
+import urllib.error
+import urllib.parse
import struct
import tempfile
from optparse import OptionParser
-from urlparse import urlparse, urlunparse
+from urllib.parse import urlparse, urlunparse
__description__ = "Download videos served using Smooth Streaming technology"
__version__ = "0.x"
def hexstring_to_bytes(hex_string):
- res = ""
- for i in range(0, len(hex_string), 2):
- res += chr(int(hex_string[i:i + 2], 16))
-
- return res
+ return bytearray.fromhex(hex_string)
def write_wav_header(out_file, fmt, codec_private_data, data_len):
fmt_len = 18 + fmt['cbSize']
wave_len = len("WAVEfmt ") + 4 + fmt_len + len('data') + 4
- out_file.write("RIFF")
+ out_file.write(bytes("RIFF", "ascii"))
out_file.write(struct.pack('<L', wave_len))
- out_file.write("WAVEfmt ")
+ out_file.write(bytes("WAVEfmt ", "ascii"))
out_file.write(struct.pack('<L', fmt_len))
out_file.write(struct.pack('<H', fmt['wFormatTag']))
out_file.write(struct.pack('<H', fmt['nChannels']))
out_file.write(struct.pack('<H', fmt['wBitsPerSample']))
out_file.write(struct.pack('<H', fmt['cbSize']))
out_file.write(extradata)
- out_file.write("data")
+ out_file.write(bytes("data", "ascii"))
out_file.write(struct.pack('<L', data_len))
def download_file(src_url, dest_file, mode):
- try:
- response = urllib2.urlopen(src_url)
- data = response.read()
- except urllib2.HTTPError:
- sys.stderr.write("Error while dowloading URL: %s" % src_url)
- raise
+
+ if os.path.exists(src_url):
+ f = open(src_url, "rb")
+ data = f.read()
+ f.close()
+ else:
+ try:
+ response = urllib.request.urlopen(src_url)
+ data = response.read()
+ except urllib.error.HTTPError:
+ sys.stderr.write("Error while dowloading URL: %s\n" % src_url)
+ raise
if dest_file:
f = open(dest_file, mode)
return data
-def get_manifest(url, dest_dir=tempfile.gettempdir()):
- """Returns the manifest and the new URL if this is changed"""
-
- if not os.path.exists(dest_dir):
- os.mkdir(dest_dir, 0755)
+def get_manifest(url, dest_dir):
+ """Returns the manifest element and the base content URL"""
# Remove the querystring if present
manifest_url = urlunparse(urlparse(url)._replace(query=''))
if not manifest_url.lower().endswith(('/manifest', '.ismc', '.csm')):
manifest_url += '/Manifest'
- if manifest_url.startswith('http://'):
- local_manifest_path = os.path.join(dest_dir, 'Manifest')
- download_file(manifest_url, local_manifest_path, "w")
- else:
- local_manifest_path = url
+ local_manifest_path = os.path.join(dest_dir, 'Manifest')
+ download_file(manifest_url, local_manifest_path, "wb")
manifest = etree.parse(local_manifest_path)
if version != "2":
raise Exception('Only Smooth Streaming version 2 supported')
- try:
- # if some intermediate client Manifest is used, like in Rai Replay
- clip = manifest.find("Clip")
- manifest_url = clip.attrib["Url"]
- manifest = download_file(manifest_url, None, None)
- except AttributeError:
- pass
+ # if some intermediate client Manifest is used, like in Rai Replay
+ # then get the final manifest
+ clip = manifest.find("Clip")
+ if clip is not None and "Url" in clip.attrib:
+ tmp_manifest_url = clip.attrib["Url"]
+ try:
+ tmp_manifest = download_file(tmp_manifest_url, None, None)
+ # set the new values only if the dowload succeded
+ manifest_url = tmp_manifest_url
+ manifest = tmp_manifest
+ except urllib.error.HTTPError:
+ pass
manifest_pattern = re.compile("/manifest$", re.IGNORECASE)
base_url = manifest_pattern.sub("", manifest_url)
for i, s in enumerate(streams):
stream_type = s.attrib["Type"]
- print "Stream: %s Type: %s" % (i, stream_type)
+ print("Stream: %s Type: %s" % (i, stream_type))
- print "\tQuality Levels:"
+ print("\tQuality Levels:")
qualities = s.findall("QualityLevel")
for i, q in enumerate(qualities):
bitrate = q.attrib["Bitrate"]
if stream_type == "video":
size = "%sx%s" % (q.attrib["MaxWidth"], q.attrib["MaxHeight"])
- print "\t%2s: %4s %10s @ %7s bps" % (i, fourcc, size, bitrate)
+ print("\t%2s: %4s %10s @ %7s bps" % (i, fourcc, size, bitrate))
if stream_type == "audio":
channels = q.attrib["Channels"]
sampling_rate = q.attrib["SamplingRate"]
bits_per_sample = q.attrib["BitsPerSample"]
- print "\t%2s: %4s %sHz %sbits %sch @ %7s bps" % \
- (i, fourcc, sampling_rate, bits_per_sample, channels,
- bitrate)
+ print("\t%2s: %4s %sHz %sbits %sch @ %7s bps" %
+ (i, fourcc, sampling_rate, bits_per_sample, channels,
+ bitrate))
- print
+ print()
def get_chunk_quality_string(stream, quality_level):
return chunks_quality
-def get_chunk_name_string(stream, chunk):
- t = chunk.attrib["t"]
+def get_chunk_name_string(stream, chunk_time):
url = stream.attrib["Url"]
- chunk_name = url.split('/')[1].replace("{start time}", t)
+ chunk_name = url.split('/')[1].replace("{start time}", str(chunk_time))
return chunk_name
def download_chunks(base_url, manifest, stream_index, quality_level, dest_dir):
-
- if not os.path.exists(dest_dir):
- os.mkdir(dest_dir, 0755)
-
stream = manifest.findall('.//StreamIndex')[stream_index]
chunks_quality = get_chunk_quality_string(stream, quality_level)
chunks_dest_dir = os.path.join(dest_dir, chunks_quality)
if not os.path.exists(chunks_dest_dir):
- os.mkdir(chunks_dest_dir, 0755)
+ os.mkdir(chunks_dest_dir, 0o755)
chunks = stream.findall("c")
data_size = 0
- print "\nDownloading Stream %d" % stream_index
- print "\tChunks %10d/%-10d" % (0, len(chunks)), "\r",
+ print("\nDownloading Stream %d" % stream_index)
+ print("\tChunks %10d/%-10d" % (0, len(chunks)), "\r", end=' ')
sys.stdout.flush()
- for i, c in enumerate(chunks):
- chunk_name = get_chunk_name_string(stream, c)
+ stream_duration = 0
+ for i, chunk in enumerate(chunks):
+
+ if "t" in chunk.attrib:
+ chunk_time = chunk.attrib["t"]
+ elif "d" in chunk.attrib:
+ chunk_time = stream_duration
+ stream_duration = chunk_time + int(chunk.attrib["d"])
+
+ chunk_name = get_chunk_name_string(stream, chunk_time)
chunk_file = os.path.join(dest_dir, chunks_quality, chunk_name)
if not os.path.exists(chunk_file):
f.close()
data_size += len(data)
- print "\tChunks %10d/%-10d" % (i + 1, len(chunks)), "\r",
+ print("\tChunks %10d/%-10d" % (i + 1, len(chunks)), "\r", end=' ')
sys.stdout.flush()
- print "\tDownloaded size:", data_size
+ print("\tDownloaded size:", data_size)
def rebuild_stream(manifest, stream_index, quality_level, src_dir,
chunks = stream.findall("c")
data_size = 0
- print "\nRebuilding Stream %d" % stream_index
- print "\tChunks %10d/%-10d" % (0, len(chunks)), "\r",
+ print("\nRebuilding Stream %d" % stream_index)
+ print("\tChunks %10d/%-10d" % (0, len(chunks)), "\r", end=' ')
sys.stdout.flush()
- for i, c in enumerate(chunks):
- chunk_name = get_chunk_name_string(stream, c)
+ stream_duration = 0
+ for i, chunk in enumerate(chunks):
+
+ if "t" in chunk.attrib:
+ chunk_time = chunk.attrib["t"]
+ elif "d" in chunk.attrib:
+ chunk_time = stream_duration
+ stream_duration = chunk_time + int(chunk.attrib["d"])
+
+ chunk_name = get_chunk_name_string(stream, chunk_time)
chunk_file = os.path.join(chunks_src_dir, chunk_name)
f = open(chunk_file, "rb")
f.close()
dest_file.write(data)
data_size += len(data)
- print "\tChunks %10d/%-10d" % (i + 1, len(chunks)), "\r",
+ print("\tChunks %10d/%-10d" % (i + 1, len(chunks)), "\r", end=' ')
sys.stdout.flush()
# Add a nice WAV header
fmt['wFormatTag'] = int(quality.attrib['AudioTag'])
fmt['nChannels'] = int(quality.attrib['Channels'])
fmt['nSamplesPerSec'] = int(quality.attrib['SamplingRate'])
- fmt['nAvgBytesPerSec'] = int(quality.attrib['Bitrate']) / 8
+ fmt['nAvgBytesPerSec'] = int(quality.attrib['Bitrate']) // 8
fmt['wBitsPerSample'] = int(quality.attrib['BitsPerSample'])
fmt['nBlockAlign'] = int(quality.attrib['PacketSize'])
fmt['cbSize'] = 0
f.close()
dest_file.close()
- print
- print "Stream %d, actual data size: %d\n" % (stream_index, data_size)
+ print()
+ print("Stream %d, actual data size: %d\n" % (stream_index, data_size))
def calc_tracks_delay(manifest, stream1_index, stream2_index):
s1 = streams[stream1_index]
s2 = streams[stream2_index]
+ if "TimeScale" not in s1 or "TimeScale" not in s2:
+ return 0
+
s1_start_chunk = s1.find("c")
s2_start_chunk = s2.find("c")
+ if "t" not in s1_start_chunk.attrib \
+ or "t" not in s2_start_chunk.attrib:
+ return 0
+
s1_start_time = int(s1_start_chunk.attrib['t'])
s2_start_time = int(s2_start_chunk.attrib['t'])
return float(duration) / 10000000 # here is the default timescale
-def smooth_download(url, manifest, dest_dir=tempfile.gettempdir(),
+def smooth_download(url, manifest, dest_dir,
video_stream_index=0, audio_stream_index=1,
video_quality_level=0, audio_quality_level=0,
chunks_dir=None, download=True,
" -vcodec copy -acodec copy ffout.mkv") % \
(dest_video, delay, dest_audio + '.wav')
- print mux_command
+ print(mux_command)
def options_parser():
parser.print_help()
parser.exit(1)
+ if not os.path.exists(options.dest_dir):
+ os.mkdir(options.dest_dir, 0o755)
+
url = args[0]
manifest, url = get_manifest(url, options.dest_dir)
parser.exit(0)
if options.sync_delay:
- print calc_tracks_delay(manifest,
+ print(calc_tracks_delay(manifest,
options.video_stream_index,
- options.audio_stream_index)
+ options.audio_stream_index))
parser.exit(0)
if options.info_only:
projects / smooth-dl.git / blobdiff