pipeline.py

# encoding=utf8
import datetime
from distutils.version import StrictVersion
import hashlib
import os.path
import random
from seesaw.config import realize, NumberConfigValue
from seesaw.externalprocess import ExternalProcess
from seesaw.item import ItemInterpolation, ItemValue
from seesaw.task import SimpleTask, LimitConcurrent
from seesaw.tracker import GetItemFromTracker, PrepareStatsForTracker, \
    UploadWithTracker, SendDoneToTracker
import shutil
import socket
import subprocess
import sys
import time
import string
import json
import re

import seesaw
from seesaw.externalprocess import WgetDownload
from seesaw.pipeline import Pipeline
from seesaw.project import Project
from seesaw.util import find_executable

import requests
import zstandard

if StrictVersion(seesaw.__version__) < StrictVersion('0.8.5'):
    raise Exception('This pipeline needs seesaw version 0.8.5 or higher.')


###########################################################################
# Find a useful Wget+Lua executable.
#
# WGET_AT will be set to the first path that
# 1. does not crash with --version, and
# 2. prints the required version string

WGET_AT = find_executable(
    'Wget+AT',
    ['GNU Wget 1.20.3-at.20211001.01'],
    [
        './wget-at',
        '/home/warrior/data/wget-at'
    ]
)

if not WGET_AT:
    raise Exception('No usable Wget+At found.')


###########################################################################
# The version number of this pipeline definition.
#
# Update this each time you make a non-cosmetic change.
# It will be added to the WARC files and reported to the tracker.
VERSION = '20211211.02'
USER_AGENT = 'Archive Team'
TRACKER_ID = 'youtube-dislikes'
TRACKER_HOST = 'legacy-api.arpa.li'
MULTI_ITEM_SIZE = 500
KEEP_WARC_ON_ABORT = True
YOUTUBEI_DATA = {
    'context': {
        'client': {
            'hl': 'en',
            'gl': 'US',
            'clientName': 'WEB',
            'clientVersion':'2.20210101'
        }
    }
}
with open('popular-videos', 'r') as f:
    POPULAR_VIDEOS = [l.strip() for l in f]

###########################################################################
# This section defines project-specific tasks.
#
# Simple tasks (tasks that do not need any concurrency) are based on the
# SimpleTask class and have a process(item) method that is called for
# each item.
class CheckIP(SimpleTask):
    def __init__(self):
        SimpleTask.__init__(self, 'CheckIP')
        self._counter = 0

    def process(self, item):
        # NEW for 2014! Check if we are behind firewall/proxy

        if self._counter <= 0:
            item.log_output('Checking IP address.')
            ip_set = set()

            ip_set.add(socket.gethostbyname('twitter.com'))
            ip_set.add(socket.gethostbyname('facebook.com'))
            ip_set.add(socket.gethostbyname('youtube.com'))
            ip_set.add(socket.gethostbyname('microsoft.com'))
            ip_set.add(socket.gethostbyname('icanhas.cheezburger.com'))
            ip_set.add(socket.gethostbyname('archiveteam.org'))

            if len(ip_set) != 6:
                item.log_output('Got IP addresses: {0}'.format(ip_set))
                item.log_output(
                    'Are you behind a firewall/proxy? That is a big no-no!')
                raise Exception(
                    'Are you behind a firewall/proxy? That is a big no-no!')

        # Check only occasionally
        if self._counter <= 0:
            self._counter = 10
        else:
            self._counter -= 1


class PrepareDirectories(SimpleTask):
    def __init__(self, warc_prefix):
        SimpleTask.__init__(self, 'PrepareDirectories')
        self.warc_prefix = warc_prefix

    def process(self, item):
        item_name = item['item_name']
        item_name_hash = hashlib.sha1(item_name.encode('utf8')).hexdigest()
        escaped_item_name = item_name_hash
        dirname = '/'.join((item['data_dir'], escaped_item_name))

        if os.path.isdir(dirname):
            shutil.rmtree(dirname)

        os.makedirs(dirname)

        item['item_dir'] = dirname
        item['warc_file_base'] = '-'.join([
            self.warc_prefix,
            item_name_hash,
            time.strftime('%Y%m%d-%H%M%S')
        ])

        open('%(item_dir)s/%(warc_file_base)s.warc.zst' % item, 'w').close()
        open('%(item_dir)s/%(warc_file_base)s_bad-items.txt' % item, 'w').close()
        open('%(item_dir)s/%(warc_file_base)s_data.txt' % item, 'w').close()

class MoveFiles(SimpleTask):
    def __init__(self):
        SimpleTask.__init__(self, 'MoveFiles')

    def process(self, item):
        os.rename('%(item_dir)s/%(warc_file_base)s.warc.zst' % item,
              '%(data_dir)s/%(warc_file_base)s.%(dict_project)s.%(dict_id)s.warc.zst' % item)
        os.rename('%(item_dir)s/%(warc_file_base)s_data.txt' % item,
              '%(data_dir)s/%(warc_file_base)s_data.txt' % item)

        shutil.rmtree('%(item_dir)s' % item)


class SetBadUrls(SimpleTask):
    def __init__(self):
        SimpleTask.__init__(self, 'SetBadUrls')

    def process(self, item):
        item['item_name_original'] = item['item_name']
        items = item['item_name'].split('\0')
        items_lower = [s for s in items]
        with open('%(item_dir)s/%(warc_file_base)s_bad-items.txt' % item, 'r') as f:
            for aborted_item in f:
                aborted_item = aborted_item.strip()
                index = items_lower.index(aborted_item)
                item.log_output('Item {} is aborted.'.format(aborted_item))
                items.pop(index)
                items_lower.pop(index)
        item['item_name'] = '\0'.join(items)


class MaybeUploadWithTracker(UploadWithTracker):
    def enqueue(self, item):
        if len(item['item_name']) == 0 and not KEEP_WARC_ON_ABORT:
            item.log_output('Skipping UploadWithTracker.')
            return self.complete_item(item)
        return super(UploadWithTracker, self).enqueue(item)


class MaybeSendDoneToTracker(SendDoneToTracker):
    def enqueue(self, item):
        if len(item['item_name']) == 0:
            item.log_output('Skipping SendDoneToTracker.')
            return self.complete_item(item)
        return super(MaybeSendDoneToTracker, self).enqueue(item)


class ZstdDict(object):
    created = 0
    data = None

    @classmethod
    def get_dict(cls):
        if cls.data is not None and time.time() - cls.created < 1800:
            return cls.data
        response = requests.get(
            'https://legacy-api.arpa.li/dictionary',
            params={
                'project': TRACKER_ID
            }
        )
        response.raise_for_status()
        response = response.json()
        if cls.data is not None and response['id'] == cls.data['id']:
            cls.created = time.time()
            return cls.data
        print('Downloading latest dictionary.')
        response_dict = requests.get(response['url'])
        response_dict.raise_for_status()
        raw_data = response_dict.content
        if hashlib.sha256(raw_data).hexdigest() != response['sha256']:
            raise ValueError('Hash of downloaded dictionary does not match.')
        if raw_data[:4] == b'\x28\xB5\x2F\xFD':
            raw_data = zstandard.ZstdDecompressor().decompress(raw_data)
        cls.data = {
            'id': response['id'],
            'dict': raw_data
        }
        cls.created = time.time()
        return cls.data


def get_hash(filename):
    with open(filename, 'rb') as in_file:
        return hashlib.sha1(in_file.read()).hexdigest()

CWD = os.getcwd()
PIPELINE_SHA1 = get_hash(os.path.join(CWD, 'pipeline.py'))
LUA_SHA1 = get_hash(os.path.join(CWD, 'youtube-dislikes.lua'))

def stats_id_function(item):
    d = {
        'pipeline_hash': PIPELINE_SHA1,
        'lua_hash': LUA_SHA1,
        'python_version': sys.version,
    }

    return d


class WgetArgs(object):
    VISITOR_DATA = None

    @classmethod
    def visitor_data(cls, visitor_data=None):
        if visitor_data is None:
            if cls.VISITOR_DATA is None and os.path.isfile('visitor_data.txt'):
                with open('visitor_data.txt', 'r') as f:
                    cls.VISITOR_DATA = f.read().strip()
            return cls.VISITOR_DATA
        cls.VISITOR_DATA = visitor_data
        with open('visitor_data.txt', 'w') as f:
            f.write(visitor_data)
        return cls.VISITOR_DATA

    def realize(self, item):
        wget_args = [
            WGET_AT,
            '-U', USER_AGENT,
            '-nv',
            '--no-cookies',
            '--content-on-error',
            '--content-on-redirect',
            '--lua-script', 'youtube-dislikes.lua',
            '-o', ItemInterpolation('%(item_dir)s/wget.log'),
            '--no-check-certificate',
            '--output-document', ItemInterpolation('%(item_dir)s/wget.tmp'),
            '--truncate-output',
            '-e', 'robots=off',
            '--rotate-dns',
            '--recursive', '--level=inf',
            '--no-parent',
            '--page-requisites',
            '--timeout', '30',
            '--tries', 'inf',
            '--domains', 'youtube.com',
            '--span-hosts',
            '--waitretry', '30',
            '--warc-file', ItemInterpolation('%(item_dir)s/%(warc_file_base)s'),
            '--warc-header', 'operator: Archive Team',
            '--warc-header', 'x-wget-at-project-version: ' + VERSION,
            '--warc-header', 'x-wget-at-project-name: ' + TRACKER_ID,
            '--warc-dedup-url-agnostic',
            '--warc-compression-use-zstd',
            '--warc-zstd-dict-no-include',
        ]
        dict_data = ZstdDict.get_dict()
        with open(os.path.join(item['item_dir'], 'zstdict'), 'wb') as f:
            f.write(dict_data['dict'])
        item['dict_id'] = dict_data['id']
        item['dict_project'] = TRACKER_ID
        wget_args.extend([
            '--warc-zstd-dict', ItemInterpolation('%(item_dir)s/zstdict'),
        ])

        items = item['item_name'].split('\0')

        visitor_data = self.visitor_data()
        found_existing = visitor_data is not None
        if found_existing:
            print('Found a previous visitorData key.')
        for i in range(300):
            if visitor_data is None:
                print('Attempting to get visitorData token')
                response = requests.post(
                    'https://www.youtubekids.com/youtubei/v1/visitor_id',
                    params={
                        'alt': 'json',
                        'key': 'AIzaSyBbZV_fZ3an51sF-mvs5w37OqqbsTOzwtU'
                    },
                    headers={
                        'Referer': 'https://www.youtubekids.com/',
                        'Content-Type': 'application/json'
                    },
                    data=json.dumps(YOUTUBEI_DATA)
                )
                visitor_data = response.json()['responseContext']['visitorData']
            for _ in range(1 if not found_existing else 10):
                video_id = random.choice(POPULAR_VIDEOS)
                print('Trying', visitor_data, 'on', video_id)
                response = requests.post(
                    'https://www.youtube.com/youtubei/v1/next',
                    params={
                        'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'
                    },
                    headers={
                        'Content-Type': 'application/json',
                        'X-Goog-Visitor-Id': visitor_data,
                        'Accept-Language': 'en-US;q=0.9, en;q=0.8'
                    },
                    data=json.dumps({
                        **YOUTUBEI_DATA,
                        'videoId': video_id
                    })
                )
                if not re.search('dislike this video along with [0-9,]+ other', response.text):
                    continue
                break
            else:
                print('Got bad visitorData token.')
                visitor_data = None
            found_existing = False
            if visitor_data is not None:
                print('Got good visitorData token.')
                break
        else:
            raise Exception('Could not get visitorData key.')

        assert visitor_data is not None

        print('Using visitorData key', visitor_data)

        self.visitor_data(visitor_data)

        item['visitor_data'] = visitor_data
        item['item_name_komma'] = item['item_name'].replace('\0', ',')

        for item_name in items:
            wget_args.extend(['--warc-header', 'x-wget-at-project-item-name: '+item_name])
            wget_args.extend(['--warc-header', 'video: '+item_name])
            #wget_args.append('item-name://'+item_name)
        wget_args.append('https://this-is-a.dummy-site/')

        if 'bind_address' in globals():
            wget_args.extend(['--bind-address', globals()['bind_address']])
            print('')
            print('*** Wget will bind address at {0} ***'.format(
                globals()['bind_address']))
            print('')

        return realize(wget_args, item)

###########################################################################
# Initialize the project.
#
# This will be shown in the warrior management panel. The logo should not
# be too big. The deadline is optional.
project = Project(
    title = 'YouTube dislikes',
    project_html = '''
    <img class="project-logo" alt="logo" src="https://wiki.archiveteam.org/images/4/4d/YouTube_logo_2017.png" height="50px"/>
    <h2>youtube.com <span class="links"><a href="https://youtube.com/">Website</a> &middot; <a href="http://tracker.archiveteam.org/youtube-dislikes/">Leaderboard</a></span></h2>
    '''
)

pipeline = Pipeline(
    CheckIP(),
    GetItemFromTracker('http://{}/{}/multi={}/'
        .format(TRACKER_HOST, TRACKER_ID, MULTI_ITEM_SIZE),
        downloader, VERSION),
    PrepareDirectories(warc_prefix='youtube-dislikes'),
    WgetDownload(
        WgetArgs(),
        max_tries=1,
        accept_on_exit_code=[0, 4, 8],
        env={
            'item_dir': ItemValue('item_dir'),
            'warc_file_base': ItemValue('warc_file_base'),
            'ids': ItemValue('item_name_komma'),
            'visitor_data': ItemValue('visitor_data')
        }
    ),
    SetBadUrls(),
    PrepareStatsForTracker(
        defaults={'downloader': downloader, 'version': VERSION},
        file_groups={
            'data': [
                ItemInterpolation('%(item_dir)s/%(warc_file_base)s.warc.zst')
            ]
        },
        id_function=stats_id_function,
    ),
    MoveFiles(),
    LimitConcurrent(NumberConfigValue(min=1, max=20, default='2',
        name='shared:rsync_threads', title='Rsync threads',
        description='The maximum number of concurrent uploads.'),
        MaybeUploadWithTracker(
            'http://%s/%s' % (TRACKER_HOST, TRACKER_ID),
            downloader=downloader,
            version=VERSION,
            files=[
                ItemInterpolation('%(data_dir)s/%(warc_file_base)s.%(dict_project)s.%(dict_id)s.warc.zst'),
                ItemInterpolation('%(data_dir)s/%(warc_file_base)s_data.txt')
            ],
            rsync_target_source_path=ItemInterpolation('%(data_dir)s/'),
            rsync_extra_args=[
                '--recursive',
                '--partial',
                '--partial-dir', '.rsync-tmp',
                '--min-size', '1',
                '--no-compress',
                '--compress-level', '0'
            ]
        ),
    ),
    MaybeSendDoneToTracker(
        tracker_url='http://%s/%s' % (TRACKER_HOST, TRACKER_ID),
        stats=ItemValue('stats')
    )
)