From 8cad77ad0d7e0a0d07629087e2ee1709688cb58d Mon Sep 17 00:00:00 2001 From: James Taylor Date: Fri, 19 Jul 2019 22:27:10 -0700 Subject: Convert comments to flask framework --- youtube/comments.py | 397 +++++++++++------------------------ youtube/templates/comments.html | 47 +++++ youtube/templates/comments_page.html | 83 ++++++++ youtube/templates/watch.html | 6 +- youtube/watch.py | 6 +- 5 files changed, 254 insertions(+), 285 deletions(-) create mode 100644 youtube/templates/comments.html create mode 100644 youtube/templates/comments_page.html (limited to 'youtube') diff --git a/youtube/comments.py b/youtube/comments.py index 94b086e..768bc13 100644 --- a/youtube/comments.py +++ b/youtube/comments.py @@ -1,57 +1,14 @@ -from youtube import proto, util, html_common, yt_data_extract, accounts +from youtube import proto, util, yt_data_extract, accounts +from youtube import yt_app import settings import json import base64 -from string import Template -import urllib.request import urllib -import html import re -comment_area_template = Template(''' -
-$video-metadata -$comment-links -$comment-box -$comments -$more-comments-button -
-''') -comment_template = Template(''' -
-
- -$avatar - -
- $author -
- - $text - - -
-$replies -$action_buttons -
-
- -
-''') -comment_avatar_template = Template(''' ''') - -reply_link_template = Template(''' - $view_replies_text -''') -with open("yt_comments_template.html", "r") as file: - yt_comments_template = Template(file.read()) - - -# $replies_link_text - +import flask +from flask import request # Here's what I know about the secret key (starting with ASJN_i) # *The secret key definitely contains the following information (or perhaps the information is stored at youtube's servers): @@ -102,6 +59,7 @@ def ctoken_metadata(ctoken): result['is_replies'] = False if (3 in offset_information) and (2 in proto.parse(offset_information[3])): result['is_replies'] = True + result['sort'] = None else: try: result['sort'] = proto.parse(offset_information[4])[6] @@ -109,12 +67,6 @@ def ctoken_metadata(ctoken): result['sort'] = 0 return result -def get_ids(ctoken): - params = proto.parse(proto.b64_to_bytes(ctoken)) - video_id = proto.parse(params[2])[2] - params = proto.parse(params[6]) - params = proto.parse(params[3]) - return params[2].decode('ascii'), video_id.decode('ascii') mobile_headers = { 'User-Agent': 'Mozilla/5.0 (iPhone; CPU iPhone OS 10_3_1 like Mac OS X) AppleWebKit/603.1.30 (KHTML, like Gecko) Version/10.0 Mobile/14E304 Safari/602.1', @@ -143,112 +95,65 @@ def request_comments(ctoken, replies=False): f.write(content)''' return content + def single_comment_ctoken(video_id, comment_id): page_params = proto.string(2, video_id) + proto.string(6, proto.percent_b64encode(proto.string(15, comment_id))) result = proto.nested(2, page_params) + proto.uint(3,6) return base64.urlsafe_b64encode(result).decode('ascii') - -def parse_comments_ajax(content, replies=False): - try: - content = json.loads(util.uppercase_escape(content.decode('utf-8'))) - #print(content) - comments_raw = content['content']['continuation_contents']['contents'] - ctoken = util.default_multi_get(content, 'content', 'continuation_contents', 'continuations', 0, 'continuation', default='') - - comments = [] - for comment_raw in comments_raw: - replies_url = '' - if not replies: - if comment_raw['replies'] is not None: - reply_ctoken = comment_raw['replies']['continuations'][0]['continuation'] - comment_id, video_id = get_ids(reply_ctoken) - replies_url = util.URL_ORIGIN + '/comments?parent_id=' + comment_id + "&video_id=" + video_id - comment_raw = comment_raw['comment'] - comment = { - 'author': comment_raw['author']['runs'][0]['text'], - 'author_url': comment_raw['author_endpoint']['url'], - 'author_channel_id': '', - 'author_id': '', - 'author_avatar': comment_raw['author_thumbnail']['url'], - 'likes': comment_raw['like_count'], - 'published': comment_raw['published_time']['runs'][0]['text'], - 'text': comment_raw['content']['runs'], - 'reply_count': '', - 'replies_url': replies_url, - } - comments.append(comment) - except Exception as e: - print('Error parsing comments: ' + str(e)) - comments = () - ctoken = '' - return {'ctoken': ctoken, 'comments': comments} -reply_count_regex = re.compile(r'(\d+)') -def parse_comments_polymer(content, replies=False): +def parse_comments_polymer(content): try: video_title = '' content = json.loads(util.uppercase_escape(content.decode('utf-8'))) url = content[1]['url'] ctoken = urllib.parse.parse_qs(url[url.find('?')+1:])['ctoken'][0] - video_id = ctoken_metadata(ctoken)['video_id'] - #print(content) + metadata = ctoken_metadata(ctoken) + try: comments_raw = content[1]['response']['continuationContents']['commentSectionContinuation']['items'] except KeyError: comments_raw = content[1]['response']['continuationContents']['commentRepliesContinuation']['contents'] - replies = True ctoken = util.default_multi_get(content, 1, 'response', 'continuationContents', 'commentSectionContinuation', 'continuations', 0, 'nextContinuationData', 'continuation', default='') - + comments = [] - for comment_raw in comments_raw: - replies_url = '' - view_replies_text = '' + for comment_json in comments_raw: + number_of_replies = 0 try: - comment_raw = comment_raw['commentThreadRenderer'] + comment_thread = comment_json['commentThreadRenderer'] except KeyError: - pass + comment_renderer = comment_json['commentRenderer'] else: - if 'commentTargetTitle' in comment_raw: - video_title = comment_raw['commentTargetTitle']['runs'][0]['text'] - - parent_id = comment_raw['comment']['commentRenderer']['commentId'] - # TODO: move this stuff into the comments_html function - if 'replies' in comment_raw: - #reply_ctoken = comment_raw['replies']['commentRepliesRenderer']['continuations'][0]['nextContinuationData']['continuation'] - #comment_id, video_id = get_ids(reply_ctoken) - replies_url = util.URL_ORIGIN + '/comments?parent_id=' + parent_id + "&video_id=" + video_id - view_replies_text = yt_data_extract.get_plain_text(comment_raw['replies']['commentRepliesRenderer']['moreText']) - match = reply_count_regex.search(view_replies_text) + if 'commentTargetTitle' in comment_thread: + video_title = comment_thread['commentTargetTitle']['runs'][0]['text'] + + if 'replies' in comment_thread: + view_replies_text = yt_data_extract.get_plain_text(comment_thread['replies']['commentRepliesRenderer']['moreText']) + view_replies_text = view_replies_text.replace(',', '') + match = re.search(r'(\d+)', view_replies_text) if match is None: - view_replies_text = '1 reply' + number_of_replies = 1 else: - view_replies_text = match.group(1) + " replies" - elif not replies: - view_replies_text = "Reply" - replies_url = util.URL_ORIGIN + '/post_comment?parent_id=' + parent_id + "&video_id=" + video_id - comment_raw = comment_raw['comment'] - - comment_raw = comment_raw['commentRenderer'] + number_of_replies = int(match.group(1)) + comment_renderer = comment_thread['comment']['commentRenderer'] + comment = { - 'author_id': comment_raw.get('authorId', ''), - 'author_avatar': comment_raw['authorThumbnail']['thumbnails'][0]['url'], - 'likes': comment_raw['likeCount'], - 'published': yt_data_extract.get_plain_text(comment_raw['publishedTimeText']), - 'text': comment_raw['contentText'].get('runs', ''), - 'view_replies_text': view_replies_text, - 'replies_url': replies_url, - 'video_id': video_id, - 'comment_id': comment_raw['commentId'], + 'author_id': comment_renderer.get('authorId', ''), + 'author_avatar': comment_renderer['authorThumbnail']['thumbnails'][0]['url'], + 'likes': comment_renderer['likeCount'], + 'published': yt_data_extract.get_plain_text(comment_renderer['publishedTimeText']), + 'text': comment_renderer['contentText'].get('runs', ''), + 'number_of_replies': number_of_replies, + 'comment_id': comment_renderer['commentId'], } - if 'authorText' in comment_raw: # deleted channels have no name or channel link - comment['author'] = yt_data_extract.get_plain_text(comment_raw['authorText']) - comment['author_url'] = comment_raw['authorEndpoint']['commandMetadata']['webCommandMetadata']['url'] - comment['author_channel_id'] = comment_raw['authorEndpoint']['browseEndpoint']['browseId'] + if 'authorText' in comment_renderer: # deleted channels have no name or channel link + comment['author'] = yt_data_extract.get_plain_text(comment_renderer['authorText']) + comment['author_url'] = comment_renderer['authorEndpoint']['commandMetadata']['webCommandMetadata']['url'] + comment['author_channel_id'] = comment_renderer['authorEndpoint']['browseEndpoint']['browseId'] else: comment['author'] = '' comment['author_url'] = '' @@ -260,172 +165,104 @@ def parse_comments_polymer(content, replies=False): comments = () ctoken = '' - return {'ctoken': ctoken, 'comments': comments, 'video_title': video_title} + return { + 'ctoken': ctoken, + 'comments': comments, + 'video_title': video_title, + 'video_id': metadata['video_id'], + 'offset': metadata['offset'], + 'is_replies': metadata['is_replies'], + 'sort': metadata['sort'], + } +def post_process_comments_info(comments_info): + for comment in comments_info['comments']: + comment['author_url'] = util.URL_ORIGIN + comment['author_url'] + comment['author_avatar'] = '/' + comment['author_avatar'] + comment['permalink'] = util.URL_ORIGIN + '/watch?v=' + comments_info['video_id'] + '&lc=' + comment['comment_id'] -def get_comments_html(comments): - html_result = '' - for comment in comments: - replies = '' - if comment['replies_url']: - replies = reply_link_template.substitute(url=comment['replies_url'], view_replies_text=html.escape(comment['view_replies_text'])) - if settings.enable_comment_avatars: - avatar = comment_avatar_template.substitute( - author_url = util.URL_ORIGIN + comment['author_url'], - author_avatar = '/' + comment['author_avatar'], - ) - else: - avatar = '' if comment['author_channel_id'] in accounts.accounts: - delete_url = (util.URL_ORIGIN + '/delete_comment?video_id=' - + comment['video_id'] + comment['delete_url'] = (util.URL_ORIGIN + '/delete_comment?video_id=' + + comments_info['video_id'] + '&channel_id='+ comment['author_channel_id'] + '&author_id=' + comment['author_id'] + '&comment_id=' + comment['comment_id']) - action_buttons = '''Delete''' + num_replies = comment['number_of_replies'] + if num_replies == 0: + comment['replies_url'] = util.URL_ORIGIN + '/post_comment?parent_id=' + comment['comment_id'] + "&video_id=" + comments_info['video_id'] else: - action_buttons = '' - - permalink = util.URL_ORIGIN + '/watch?v=' + comment['video_id'] + '&lc=' + comment['comment_id'] - html_result += comment_template.substitute( - author=comment['author'], - author_url = util.URL_ORIGIN + comment['author_url'], - avatar = avatar, - likes = str(comment['likes']) + ' likes' if str(comment['likes']) != '0' else '', - published = comment['published'], - text = yt_data_extract.format_text_runs(comment['text']), - datetime = '', #TODO - replies = replies, - action_buttons = action_buttons, - permalink = permalink, - ) - return html_result - + comment['replies_url'] = util.URL_ORIGIN + '/comments?parent_id=' + comment['comment_id'] + "&video_id=" + comments_info['video_id'] + + if num_replies == 0: + comment['view_replies_text'] = 'Reply' + elif num_replies == 1: + comment['view_replies_text'] = '1 reply' + else: + comment['view_replies_text'] = str(num_replies) + ' replies' + + + if comment['likes'] == 1: + comment['likes_text'] = '1 like' + else: + comment['likes_text'] = str(comment['likes']) + ' likes' + + comments_info['include_avatars'] = settings.enable_comment_avatars + if comments_info['ctoken'] != '': + comments_info['more_comments_url'] = util.URL_ORIGIN + '/comments?ctoken=' + comments_info['ctoken'] + + comments_info['page_number'] = page_number = str(int(comments_info['offset']/20) + 1) + + if not comments_info['is_replies']: + comments_info['sort_text'] = 'top' if comments_info['sort'] == 0 else 'newest' + + + comments_info['video_url'] = util.URL_ORIGIN + '/watch?v=' + comments_info['video_id'] + comments_info['video_thumbnail'] = '/i.ytimg.com/vi/'+ comments_info['video_id'] + '/mqdefault.jpg' + + def video_comments(video_id, sort=0, offset=0, lc='', secret_key=''): if settings.enable_comments: - post_comment_url = util.URL_ORIGIN + "/post_comment?video_id=" + video_id - post_comment_link = '''Post comment''' + comments_info = parse_comments_polymer(request_comments(make_comment_ctoken(video_id, sort, offset, lc, secret_key))) + post_process_comments_info(comments_info) + post_comment_url = util.URL_ORIGIN + "/post_comment?video_id=" + video_id other_sort_url = util.URL_ORIGIN + '/comments?ctoken=' + make_comment_ctoken(video_id, sort=1 - sort, lc=lc) - other_sort_name = 'newest' if sort == 0 else 'top' - other_sort_link = '''Sort by ''' + other_sort_name + '''''' - - comment_links = '''
\n''' - comment_links += other_sort_link + '\n' + post_comment_link + '\n' - comment_links += '''
''' - - comment_info = parse_comments_polymer(request_comments(make_comment_ctoken(video_id, sort, offset, lc, secret_key))) - ctoken = comment_info['ctoken'] - - if ctoken == '': - more_comments_button = '' - else: - more_comments_button = more_comments_template.substitute(url = util.URL_ORIGIN + '/comments?ctoken=' + ctoken) - - result = '''
\n''' - result += comment_links + '\n' - result += '
\n' - result += get_comments_html(comment_info['comments']) + '\n' - result += '
\n' - result += more_comments_button + '\n' - result += '''
''' - return result - return '' - -more_comments_template = Template('''More comments''') -video_metadata_template = Template('''
- - - - $title - -

Comments page $page_number

- Sorted by $sort -
-''') -account_option_template = Template(''' - ''') - -def comment_box_account_options(): - return ''.join(account_option_template.substitute(channel_id=channel_id, display_name=display_name) for channel_id, display_name in accounts.account_list_data()) - -comment_box_template = Template(''' -
-
- - - Add account -
- - $video_id_input - -
''') -def get_comments_page(env, start_response): - start_response('200 OK', [('Content-type','text/html'),] ) - parameters = env['parameters'] - ctoken = util.default_multi_get(parameters, 'ctoken', 0, default='') + other_sort_text = 'Sort by ' + ('newest' if sort == 0 else 'top') + comments_info['comment_links'] = [('Post comment', post_comment_url), (other_sort_text, other_sort_url)] + + return comments_info + + return {} + + + +@yt_app.route('/comments') +def get_comments_page(): + ctoken = request.args.get('ctoken', '') replies = False if not ctoken: - video_id = parameters['video_id'][0] - parent_id = parameters['parent_id'][0] + video_id = request.args['video_id'] + parent_id = request.args['parent_id'] ctoken = comment_replies_ctoken(video_id, parent_id) replies = True - comment_info = parse_comments_polymer(request_comments(ctoken, replies), replies) + comments_info = parse_comments_polymer(request_comments(ctoken, replies)) + post_process_comments_info(comments_info) + + if not replies: + other_sort_url = util.URL_ORIGIN + '/comments?ctoken=' + make_comment_ctoken(comments_info['video_id'], sort=1 - comments_info['sort']) + other_sort_text = 'Sort by ' + ('newest' if comments_info['sort'] == 0 else 'top') + comments_info['comment_links'] = [(other_sort_text, other_sort_url)] + + + return flask.render_template('comments_page.html', + comments_info = comments_info, + + form_action = '' if replies else util.URL_ORIGIN + '/post_comment', + include_video_id_input = not replies, + accounts = accounts.account_list_data(), + ) - metadata = ctoken_metadata(ctoken) - if replies: - page_title = 'Replies' - video_metadata = '' - comment_box = comment_box_template.substitute(form_action='', video_id_input='', post_text='Post reply', options=comment_box_account_options()) - comment_links = '' - else: - page_number = str(int(metadata['offset']/20) + 1) - page_title = 'Comments page ' + page_number - - video_metadata = video_metadata_template.substitute( - page_number = page_number, - sort = 'top' if metadata['sort'] == 0 else 'newest', - title = html.escape(comment_info['video_title']), - url = util.URL_ORIGIN + '/watch?v=' + metadata['video_id'], - thumbnail = '/i.ytimg.com/vi/'+ metadata['video_id'] + '/mqdefault.jpg', - ) - comment_box = comment_box_template.substitute( - form_action= util.URL_ORIGIN + '/post_comment', - video_id_input='''''', - post_text='Post comment', - options=comment_box_account_options(), - ) - - other_sort_url = util.URL_ORIGIN + '/comments?ctoken=' + make_comment_ctoken(metadata['video_id'], sort=1 - metadata['sort']) - other_sort_name = 'newest' if metadata['sort'] == 0 else 'top' - other_sort_link = '''Sort by ''' + other_sort_name + '''''' - - - comment_links = '''
\n''' - comment_links += other_sort_link + '\n' - comment_links += '''
''' - - comments_html = get_comments_html(comment_info['comments']) - ctoken = comment_info['ctoken'] - if ctoken == '': - more_comments_button = '' - else: - more_comments_button = more_comments_template.substitute(url = util.URL_ORIGIN + '/comments?ctoken=' + ctoken) - comments_area = '
\n' - comments_area += video_metadata + comment_box + comment_links + '\n' - comments_area += '
\n' - comments_area += comments_html + '\n' - comments_area += '
\n' - comments_area += more_comments_button + '\n' - comments_area += '
\n' - return yt_comments_template.substitute( - header = html_common.get_header(), - comments_area = comments_area, - page_title = page_title, - ).encode('utf-8') diff --git a/youtube/templates/comments.html b/youtube/templates/comments.html new file mode 100644 index 0000000..901190f --- /dev/null +++ b/youtube/templates/comments.html @@ -0,0 +1,47 @@ +{% import "common_elements.html" as common_elements %} + +{% macro render_comment(comment, include_avatar) %} +
+
+ + {% if include_avatar %} + + {% endif %} + +
+ {{ comment['author'] }} +
+ + {{ common_elements.text_runs(comment['text']) }} + + +
+ {{ comment['view_replies_text'] }} + {% if 'delete_url' is in comment %} + Delete + {% endif %} +
+
+ +
+{% endmacro %} + +{% macro video_comments(comments_info) %} +
+
+ {% for link_text, link_url in comments_info['comment_links'] %} + {{ link_text }} + {% endfor %} +
+
+ {% for comment in comments_info['comments'] %} + {{ render_comment(comment, comments_info['include_avatars']) }} + {% endfor %} +
+ {% if 'more_comments_url' is in comments_info %} + More comments + {% endif %} +
+{% endmacro %} diff --git a/youtube/templates/comments_page.html b/youtube/templates/comments_page.html new file mode 100644 index 0000000..a77d2b4 --- /dev/null +++ b/youtube/templates/comments_page.html @@ -0,0 +1,83 @@ +{% extends "base.html" %} +{% import "comments.html" as comments %} + +{% block page_title %}{{ 'Replies' if comments_info['is_replies'] else 'Comments page ' + comments_info['page_number'] }}{% endblock %} + + +{% block style %} + main{ + display:grid; + grid-template-columns: 3fr 2fr; + } + #left{ + background-color:#bcbcbc; + + display: grid; + grid-column: 1; + grid-row: 1; + grid-template-columns: 1fr 640px; + grid-template-rows: 0fr 0fr 0fr; + } + .comments-area{ + grid-column:2; + } + .comment{ + width:640px; + } +{% endblock style %} + + +{% block main %} +
+
+ {% if not comments_info['is_replies'] %} +
+ + + + {{ comments_info['video_title'] }} + +

Comments page {{ comments_info['page_number'] }}

+ Sorted by {{ comments_info['sort_text'] }} +
+ {% endif %} + + +
+
+ + + Add account +
+ + {% if include_video_id_input %} + + {% endif %} + +
+ + {% if not comments_info['is_replies'] %} +
+ {% for link_text, link_url in comments_info['comment_links'] %} + {{ link_text }} + {% endfor %} +
+ {% endif %} + +
+ {% for comment in comments_info['comments'] %} + {{ comments.render_comment(comment, comments_info['include_avatars']) }} + {% endfor %} +
+ {% if 'more_comments_url' is in comments_info %} + More comments + {% endif %} +
+
+{% endblock main %} + + diff --git a/youtube/templates/watch.html b/youtube/templates/watch.html index 85c87ae..d61997f 100644 --- a/youtube/templates/watch.html +++ b/youtube/templates/watch.html @@ -1,5 +1,6 @@ {% extends "base.html" %} {% import "common_elements.html" as common_elements %} +{% import "comments.html" as comments %} {% block page_title %}{{ title }}{% endblock %} {% block style %} main{ @@ -211,7 +212,10 @@ {% endif %} -{{ comments|safe }} + + {% if comments_info %} + {{ comments.video_comments(comments_info) }} + {% endif %} diff --git a/youtube/watch.py b/youtube/watch.py index 1972cb4..dc50b7b 100644 --- a/youtube/watch.py +++ b/youtube/watch.py @@ -153,7 +153,7 @@ def get_watch_page(): gevent.spawn(extract_info, yt_dl_downloader, "https://www.youtube.com/watch?v=" + video_id, download=False) ) gevent.joinall(tasks) - comments_html, info = tasks[0].value, tasks[1].value + comments_info, info = tasks[0].value, tasks[1].value if isinstance(info, str): # youtube error return flask.render_template('error.html', error_message = info) @@ -207,9 +207,7 @@ def get_watch_page(): related = related_videos, music_list = info['music_list'], music_attributes = get_ordered_music_list_attributes(info['music_list']), - - # TODO: refactor these - comments = comments_html, + comments_info = comments_info, title = info['title'], uploader = info['uploader'], -- cgit v1.2.3