yt-local/youtube/search.py
2018-06-30 23:34:46 -07:00

231 lines
9.8 KiB
Python

import json
import urllib
import html
from string import Template
import base64
from math import ceil
from youtube.common import default_multi_get, get_thumbnail_url, URL_ORIGIN
import youtube.common as common
with open("yt_search_results_template.html", "r") as file:
yt_search_results_template = file.read()
with open("yt_search_template.html", "r") as file:
yt_search_template = file.read()
page_button_template = Template('''<a class="page-button" href="$href">$page</a>''')
current_page_button_template = Template('''<div class="page-button">$page</div>''')
video_result_template = '''
<div class="medium-item">
<a class="video-thumbnail-box" href="$video_url" title="$video_title">
<img class="video-thumbnail-img" src="$thumbnail_url">
<span class="video-duration">$length</span>
</a>
<a class="title" href="$video_url">$video_title</a>
<address>Uploaded by <a href="$uploader_channel_url">$uploader</a></address>
<span class="views">$views</span>
<time datetime="$datetime">Uploaded $upload_date</time>
<span class="description">$description</span>
</div>
'''
# Sort: 1
# Upload date: 2
# View count: 3
# Rating: 1
# Offset: 9
# Filters: 2
# Upload date: 1
# Type: 2
# Duration: 3
features = {
'4k': 14,
'hd': 4,
'hdr': 25,
'subtitles': 5,
'creative_commons': 6,
'3d': 7,
'live': 8,
'purchased': 9,
'360': 15,
'location': 23,
}
def page_number_to_sp_parameter(page):
offset = (int(page) - 1)*20 # 20 results per page
first_byte = 255 & offset
second_byte = 255 & (offset >> 7)
second_byte = second_byte | 1
# 0b01001000 is required, and is always the same.
# The next 2 bytes encode the offset in little endian order,
# BUT, it's done in a strange way. The least significant bit (LSB) of the second byte is not part
# of the offset. Instead, to get the number which the two bytes encode, that LSB
# of the second byte is combined with the most significant bit (MSB) of the first byte
# in a logical AND. Replace the two bits with the result of the AND to get the two little endian
# bytes that represent the offset.
# I figured this out by trial and error on the sp parameter. I don't know why it's done like this;
# perhaps it's just obfuscation.
param_bytes = bytes((0b01001000, first_byte, second_byte))
param_encoded = urllib.parse.quote(base64.urlsafe_b64encode(param_bytes))
return param_encoded
def get_search_json(query, page):
url = "https://www.youtube.com/results?search_query=" + urllib.parse.quote_plus(query)
headers = {
'Host': 'www.youtube.com',
'User-Agent': 'Mozilla/5.0 (Windows NT 6.1; Win64; x64)',
'Accept': '*/*',
'Accept-Language': 'en-US,en;q=0.5',
'X-YouTube-Client-Name': '1',
'X-YouTube-Client-Version': '2.20180418',
}
url += "&pbj=1&sp=" + page_number_to_sp_parameter(page)
content = common.fetch_url(url, headers=headers)
info = json.loads(content)
return info
"""def get_search_info(query, page):
result_info = dict()
info = get_bloated_search_info(query, page)
estimated_results = int(info[1]['response']['estimatedResults'])
estimated_pages = ceil(estimated_results/20)
result_info['estimated_results'] = estimated_results
result_info['estimated_pages'] = estimated_pages
result_info['results'] = []
# this is what you get when you hire H-1B's
video_list = info[1]['response']['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'][0]['itemSectionRenderer']['contents']
for video_json_crap in video_list:
# they have a dictionary whose only content is another dictionary...
try:
type = list(video_json_crap.keys())[0]
except KeyError:
continue #channelRenderer or playlistRenderer
'''description = ""
for text_run in video_json_crap["descriptionSnippet"]["runs"]:
if text_run.get("bold", False):
description += "<b>" + html.escape'''
try:
result_info['results'].append({
"title": video_json_crap["title"]["simpleText"],
"video_id": video_json_crap["videoId"],
"description": video_json_crap.get("descriptionSnippet",dict()).get('runs',[]), # a list of text runs (formmated), rather than plain text
"thumbnail": get_thumbnail_url(video_json_crap["videoId"]),
"views_text": video_json_crap['viewCountText'].get('simpleText', None) or video_json_crap['viewCountText']['runs'][0]['text'],
"length_text": default_multi_get(video_json_crap, 'lengthText', 'simpleText', default=''), # livestreams dont have a length
"uploader": video_json_crap['longBylineText']['runs'][0]['text'],
"uploader_url": URL_ORIGIN + video_json_crap['longBylineText']['runs'][0]['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'],
"published_time_text": default_multi_get(video_json_crap, 'publishedTimeText', 'simpleText', default=''),
})
except KeyError:
print(video_json_crap)
raise
return result_info"""
def page_buttons_html(page_start, page_end, current_page, query):
result = ""
for page in range(page_start, page_end+1):
if page == current_page:
template = current_page_button_template
else:
template = page_button_template
result += template.substitute(page=page, href=URL_ORIGIN + "/search?query=" + urllib.parse.quote_plus(query) + "&page=" + str(page))
return result
showing_results_for = Template('''
<div>Showing results for <a>$corrected_query</a></div>
<div>Search instead for <a href="$original_query_url">$original_query</a></div>
''')
did_you_mean = Template('''
<div>Did you mean <a href="$corrected_query_url">$corrected_query</a></div>
''')
def get_search_page(query_string, parameters=()):
qs_query = urllib.parse.parse_qs(query_string)
if len(qs_query) == 0:
return yt_search_template
query = qs_query["query"][0]
page = qs_query.get("page", "1")[0]
info = get_search_json(query, page)
estimated_results = int(info[1]['response']['estimatedResults'])
estimated_pages = ceil(estimated_results/20)
results = info[1]['response']['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'][0]['itemSectionRenderer']['contents']
corrections = ''
result_list_html = ""
for renderer in results:
type = list(renderer.keys())[0]
if type == 'shelfRenderer':
continue
if type == 'didYouMeanRenderer':
renderer = renderer[type]
corrected_query_string = urllib.parse.parse_qs(query_string)
corrected_query_string['query'] = [renderer['correctedQueryEndpoint']['searchEndpoint']['query']]
corrected_query_url = URL_ORIGIN + '/search?' + common.make_query_string(corrected_query_string)
corrections = did_you_mean.substitute(
corrected_query_url = corrected_query_url,
corrected_query = common.format_text_runs(renderer['correctedQuery']['runs']),
)
continue
if type == 'showingResultsForRenderer':
renderer = renderer[type]
no_autocorrect_query_string = urllib.parse.parse_qs(query_string)
no_autocorrect_query_string['autocorrect'] = ['0']
no_autocorrect_query_url = URL_ORIGIN + '/search?' + common.make_query_string(no_autocorrect_query_string)
corrections = showing_results_for.substitute(
corrected_query = common.format_text_runs(renderer['correctedQuery']['runs']),
original_query_url = no_autocorrect_query_url,
original_query = html.escape(renderer['originalQuery']['simpleText']),
)
continue
result_list_html += common.renderer_html(renderer, current_query_string=query_string)
'''type = list(result.keys())[0]
result = result[type]
if type == "showingResultsForRenderer":
url = URL_ORIGIN + "/search"
if len(parameters) > 0:
url += ';' + ';'.join(parameters)
url += '?' + '&'.join(key + '=' + ','.join(values) for key,values in qs_query.items())
result_list_html += showing_results_for_template.substitute(
corrected_query=common.format_text_runs(result['correctedQuery']['runs']),
)
else:
result_list_html += common.html_functions[type](result)'''
page = int(page)
if page <= 5:
page_start = 1
page_end = min(9, estimated_pages)
else:
page_start = page - 4
page_end = min(page + 4, estimated_pages)
result = Template(yt_search_results_template).substitute(
results = result_list_html,
page_title = query + " - Search",
search_box_value = html.escape(query),
number_of_results = '{:,}'.format(estimated_results),
number_of_pages = '{:,}'.format(estimated_pages),
page_buttons = page_buttons_html(page_start, page_end, page, query),
corrections = corrections
)
return result