youtube downloader

#!/usr/bin/env python
#
# Copyright (c) 2006-2008 Ricardo Garcia Gonzalez
#
# Permission is hereby granted, free of charge, to any person obtaining a
# copy of this software and associated documentation files (the "Software"),
# to deal in the Software without restriction, including without limitation
# the rights to use, copy, modify, merge, publish, distribute, sublicense,
# and/or sell copies of the Software, and to permit persons to whom the
# Software is furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included
# in all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
# OTHER DEALINGS IN THE SOFTWARE.
#
# Except as contained in this notice, the name(s) of the above copyright
# holders shall not be used in advertising or otherwise to promote the
# sale, use or other dealings in this Software without prior written
# authorization.
#
import getpass
import httplib
import math
import netrc
import optparse
import os
import re
import socket
import string
import sys
import time
import urllib2

# Global constants
const_1k = 1024
const_initial_block_size = 10 * const_1k
const_epsilon = 0.0001
const_timeout = 120

const_video_url_str = 'http://www.youtube.com/watch?v=%s'
const_video_url_re = re.compile(r'^((?:http://)?(?:\w+\.)?youtube\.com/(?:v/|(?:watch(?:\.php)?)?\?(?:.+&)?v=))?([0-9A-Za-z_-]+)(?(1)[&/].*)?$')
const_login_url_str = 'http://www.youtube.com/login?next=/watch%%3Fv%%3D%s'
const_login_post_str = 'current_form=loginForm&next=%%2Fwatch%%3Fv%%3D%s&username=%s&password=%s&action_login=Log+In'
const_age_url_str = 'http://www.youtube.com/verify_age?next_url=/watch%%3Fv%%3D%s'
const_age_post_str = 'next_url=%%2Fwatch%%3Fv%%3D%s&action_confirm=Confirm'
const_url_t_param_re = re.compile(r', "t": "([^"]+)"')
const_video_url_real_str = 'http://www.youtube.com/get_video?video_id=%s&t=%s'
const_video_title_re = re.compile(r'<title>YouTube - ([^<]*)</title>', re.M | re.I)

# Print error message, followed by standard advice information, and then exit
def error_advice_exit(error_text):
    sys.stderr.write('Error: %s.\n' % error_text)
    sys.stderr.write('Try again several times. It may be a temporary problem.\n')
    sys.stderr.write('Other typical problems:\n\n')
    sys.stderr.write('* Video no longer exists.\n')
    sys.stderr.write('* Video requires age confirmation but you did not provide an account.\n')
    sys.stderr.write('* You provided the account data, but it is not valid.\n')
    sys.stderr.write('* The connection was cut suddenly for some reason.\n')
    sys.stderr.write('* YouTube changed their system, and the program no longer works.\n')
    sys.stderr.write('\nTry to confirm you are able to view the video using a web browser.\n')
    sys.stderr.write('Use the same video URL and account information, if needed, with this program.\n')
    sys.stderr.write('When using a proxy, make sure http_proxy has http://host:port format.\n')
    sys.stderr.write('Try again several times and contact me if the problem persists.\n')
    sys.exit('\n')

# Wrapper to create custom requests with typical headers
def request_create(url, data=None):
    retval = urllib2.Request(url)
    if data is not None:
        retval.add_data(data)
    # Try to mimic Firefox, at least a little bit
    retval.add_header('User-Agent', 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.8.1.11) Gecko/20071127 Firefox/2.0.0.11')
    retval.add_header('Accept-Charset', 'ISO-8859-1,utf-8;q=0.7,*;q=0.7')
    retval.add_header('Accept', 'text/xml,application/xml,application/xhtml+xml,text/html;q=0.9,text/plain;q=0.8,image/png,*/*;q=0.5')
    retval.add_header('Accept-Language', 'en-us,en;q=0.5')
    return retval

# Perform a request, process headers and return response
def perform_request(url, data=None):
    request = request_create(url, data)
    response = urllib2.urlopen(request)
    return response

# Conditional print
def cond_print(str):
    global cmdl_opts
    if not (cmdl_opts.quiet or cmdl_opts.get_url):
        sys.stdout.write(str)
        sys.stdout.flush()

# Title string normalization
def title_string_norm(title):
    title = ''.join((x in string.ascii_letters or x in string.digits) and x or ' ' for x in title)
    title = '_'.join(title.split())
    title = title.lower()
    return title

# Generic download step
def download_step(return_data_flag, step_title, step_error, url, post_data=None):
    try:
        cond_print('%s... ' % step_title)
        data = perform_request(url, post_data).read()
        cond_print('done.\n')
        if return_data_flag:
            return data
        return None

    except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError, socket.error):
        cond_print('failed.\n')
        error_advice_exit(step_error)

    except KeyboardInterrupt:
        sys.exit('\n')

# Generic extract step
def extract_step(step_title, step_error, regexp, data):
    try:
        cond_print('%s... ' % step_title)
        match = regexp.search(data)
        
        if match is None:
            cond_print('failed.\n')
            error_advice_exit(step_error)
        
        extracted_data = match.group(1)
        cond_print('done.\n')
        return extracted_data
    
    except KeyboardInterrupt:
        sys.exit('\n')

# Calculate new block size based on previous block size
def new_block_size(before, after, bytes):
    new_min = max(bytes / 2.0, 1.0)
    new_max = max(bytes * 2.0, 1.0)
    dif = after - before
    if dif < const_epsilon:
        return int(new_max)
    rate = bytes / dif
    if rate > new_max:
        return int(new_max)
    if rate < new_min:
        return int(new_min)
    return int(rate)

# Get optimum 1k exponent to represent a number of bytes
def optimum_k_exp(num_bytes):
    global const_1k
    if num_bytes == 0:
        return 0
    return long(math.log(num_bytes, const_1k))

# Get optimum representation of number of bytes
def format_bytes(num_bytes):
    global const_1k
    try:
        exp = optimum_k_exp(num_bytes)
        suffix = 'bkMGTPEZY'[exp]
        if exp == 0:
            return '%s%s' % (num_bytes, suffix)
        converted = float(num_bytes) / float(const_1k**exp)
        return '%.2f%s' % (converted, suffix)
    except IndexError:
        sys.exit('Error: internal error formatting number of bytes.')

# Calculate ETA and return it in string format as MM:SS
def calc_eta(start, now, total, current):
    dif = now - start
    if current == 0 or dif < const_epsilon:
        return '--:--'
    rate = float(current) / dif
    eta = long((total - current) / rate)
    (eta_mins, eta_secs) = divmod(eta, 60)
    if eta_mins > 99:
        return '--:--'
    return '%02d:%02d' % (eta_mins, eta_secs)

# Calculate speed and return it in string format
def calc_speed(start, now, bytes):
    dif = now - start
    if bytes == 0 or dif < const_epsilon:
        return 'N/A b'
    return format_bytes(float(bytes) / dif)


# Title string minimal transformation
def title_string_touch(title):
    return title.replace(os.sep, '%')

# Create the command line options parser and parse command line
cmdl_usage = 'usage: %prog [options] video_url'
cmdl_version = '2008.01.24'
cmdl_parser = optparse.OptionParser(usage=cmdl_usage, version=cmdl_version, conflict_handler='resolve')
cmdl_parser.add_option('-h', '--help', action='help', help='print this help text and exit')
cmdl_parser.add_option('-v', '--version', action='version', help='print program version and exit')
cmdl_parser.add_option('-u', '--username', dest='username', metavar='USERNAME', help='account username')
cmdl_parser.add_option('-p', '--password', dest='password', metavar='PASSWORD', help='account password')
cmdl_parser.add_option('-o', '--output', dest='outfile', metavar='FILE', help='output video file name')
cmdl_parser.add_option('-q', '--quiet', action='store_true', dest='quiet', help='activates quiet mode')
cmdl_parser.add_option('-s', '--simulate', action='store_true', dest='simulate', help='do not download video')
cmdl_parser.add_option('-t', '--title', action='store_true', dest='use_title', help='use title in file name')
cmdl_parser.add_option('-l', '--literal', action='store_true', dest='use_literal', help='use literal title in file name')
cmdl_parser.add_option('-n', '--netrc', action='store_true', dest='use_netrc', help='use .netrc authentication data')
cmdl_parser.add_option('-g', '--get-url', action='store_true', dest='get_url', help='print final video URL only')
cmdl_parser.add_option('-2', '--title-too', action='store_true', dest='get_title', help='used with -g, print title too')
(cmdl_opts, cmdl_args) = cmdl_parser.parse_args()

# Set socket timeout
socket.setdefaulttimeout(const_timeout)

# Get video URL
if len(cmdl_args) != 1:
    cmdl_parser.print_help()
    sys.exit('\n')
video_url_cmdl = cmdl_args[0]

# Verify video URL format and convert to "standard" format
video_url_mo = const_video_url_re.match(video_url_cmdl)
if video_url_mo is None:
    sys.exit('Error: URL does not seem to be a youtube video URL. If it is, report a bug.')
video_url_id = video_url_mo.group(2)
video_url = const_video_url_str % video_url_id

# Check conflicting options
if cmdl_opts.outfile is not None and (cmdl_opts.simulate or cmdl_opts.get_url):
    sys.stderr.write('Warning: video file name given but will not be used.\n')

if cmdl_opts.outfile is not None and (cmdl_opts.use_title or cmdl_opts.use_literal):
    sys.exit('Error: using the video title conflicts with using a given file name.')

if cmdl_opts.use_title and cmdl_opts.use_literal:
    sys.exit('Error: cannot use title and literal title at the same time.')

if cmdl_opts.quiet and cmdl_opts.get_url:
    sys.exit('Error: cannot be quiet and print final URL at the same time.')

# Incorrect option formatting
if cmdl_opts.username is None and cmdl_opts.password is not None:
    sys.exit('Error: password give but username is missing.')

if cmdl_opts.use_netrc and (cmdl_opts.username is not None or cmdl_opts.password is not None):
    sys.exit('Error: cannot use netrc and username/password at the same time.')

if cmdl_opts.get_url is None and cmdl_opts.get_title is not None:
    sys.exit('Error: getting title requires getting URL.')

# Get account information if any
account_username = None
account_password = None

if cmdl_opts.use_netrc:
    try:
        info = netrc.netrc().authenticators('youtube')
        if info is None:
            sys.exit('Error: no authenticators for machine youtube.')
        account_username = info[0]
        account_password = info[2]
    except IOError:
        sys.exit('Error: unable to read .netrc file.')
    except netrc.NetrcParseError:
        sys.exit('Error: unable to parse .netrc file.')
else:
    account_username = cmdl_opts.username
    if account_username is not None:
        if cmdl_opts.password is None:
            account_password = getpass.getpass('Type YouTube password and press return: ')
        else:
            account_password = cmdl_opts.password

# Get output file name
if cmdl_opts.outfile is None:
    video_filename = '%s.flv' % video_url_id
else:
    video_filename = cmdl_opts.outfile

# Install cookie and proxy handlers
urllib2.install_opener(urllib2.build_opener(urllib2.ProxyHandler()))
urllib2.install_opener(urllib2.build_opener(urllib2.HTTPCookieProcessor()))

# Log in and confirm age if needed
if account_username is not None:
    url = const_login_url_str % video_url_id
    post = const_login_post_str % (video_url_id, account_username, account_password)
    download_step(False, 'Logging in', 'unable to log in', url, post)

    url = const_age_url_str % video_url_id
    post = const_age_post_str % video_url_id
    download_step(False, 'Confirming age', 'unable to confirm age', url, post)

# Retrieve video webpage
video_webpage = download_step(True, 'Retrieving video webpage', 'unable to retrieve video webpage', video_url)

# Extract video title if needed
if cmdl_opts.use_title or cmdl_opts.use_literal or cmdl_opts.get_title:
    video_title = extract_step('Extracting video title', 'unable to extract video title', const_video_title_re, video_webpage)

# Extract needed video URL parameters
video_url_t_param = extract_step('Extracting URL "t" parameter', 'unable to extract URL "t" parameter', const_url_t_param_re, video_webpage)
video_url_real = const_video_url_real_str % (video_url_id, video_url_t_param)

# Rebuild filename if needed
if cmdl_opts.use_title or cmdl_opts.use_literal:
    if cmdl_opts.use_title:
        prefix = title_string_norm(video_title)
    else:
        prefix = title_string_touch(video_title)
    video_filename = '%s-%s.flv' % (prefix, video_url_id)

# Check name
if not video_filename.lower().endswith('.flv'):
    sys.stderr.write('Warning: video file name does not end in .flv\n')

# Retrieve video data
try:
    cond_print('Requesting video file... ')
    video_data = perform_request(video_url_real)
    cond_print('done.\n')
    cond_print('Video data found at %s\n' % video_data.geturl())

    if cmdl_opts.get_title:
        print video_title

    if cmdl_opts.get_url:
        print video_data.geturl()

    if cmdl_opts.simulate or cmdl_opts.get_url:
        sys.exit()

    try:
        video_file = open(video_filename, 'wb')
    except (IOError, OSError):
        sys.exit('Error: unable to open "%s" for writing.' % video_filename)
    try:
        video_len = long(video_data.info()['Content-length'])
        video_len_str = format_bytes(video_len)
    except KeyError:
        video_len = None
        video_len_str = 'N/A'

    byte_counter = 0
    block_size = const_initial_block_size
    start_time = time.time()
    while True:
        if video_len is not None:
            percent = float(byte_counter) / float(video_len) * 100.0
            percent_str = '%.1f' % percent
            eta_str = calc_eta(start_time, time.time(), video_len, byte_counter)
        else:
            percent_str = '---.-'
            eta_str = '--:--'
        counter = format_bytes(byte_counter)
        speed_str = calc_speed(start_time, time.time(), byte_counter)
        cond_print('\rRetrieving video data: %5s%% (%8s of %s) at %8s/s ETA %s ' % (percent_str, counter, video_len_str, speed_str, eta_str))

        before = time.time()
        video_block = video_data.read(block_size)
        after = time.time()
        dl_bytes = len(video_block)
        if dl_bytes == 0:
            break
        byte_counter += dl_bytes
        video_file.write(video_block)
        block_size = new_block_size(before, after, dl_bytes)

    if video_len is not None and byte_counter != video_len:
        error_advice_exit('server did not send the expected amount of data')

    video_file.close()
    cond_print('done.\n')
    cond_print('Video data saved to %s\n' % video_filename)

except (urllib2.URLError, ValueError, httplib.HTTPException, TypeError, socket.error):
    cond_print('failed.\n')
    error_advice_exit('unable to download video data')

except KeyboardInterrupt:
    sys.exit('\n')

# Finish
sys.exit()