Channel: Render age restricted channels

This commit is contained in:
ChunkyProgrammer
2023-11-26 20:24:04 -05:00
parent e319c35f09
commit 96ade642fa
4 changed files with 221 additions and 119 deletions

View File

@@ -15,7 +15,8 @@ record AboutChannel,
allowed_regions : Array(String),
tabs : Array(String),
tags : Array(String),
verified : Bool
verified : Bool,
is_age_gated : Bool
def get_about_info(ucid, locale) : AboutChannel
begin
@@ -45,46 +46,102 @@ def get_about_info(ucid, locale) : AboutChannel
end
tags = [] of String
tab_names = [] of String
total_views = 0_i64
joined = Time.unix(0)
if auto_generated
author = initdata["header"]["interactiveTabbedHeaderRenderer"]["title"]["simpleText"].as_s
author_url = initdata["microformat"]["microformatDataRenderer"]["urlCanonical"].as_s
author_thumbnail = initdata["header"]["interactiveTabbedHeaderRenderer"]["boxArt"]["thumbnails"][0]["url"].as_s
# Raises a KeyError on failure.
banners = initdata["header"]["interactiveTabbedHeaderRenderer"]?.try &.["banner"]?.try &.["thumbnails"]?
banner = banners.try &.[-1]?.try &.["url"].as_s?
description_base_node = initdata["header"]["interactiveTabbedHeaderRenderer"]["description"]
# some channels have the description in a simpleText
# ex: https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/
description_node = description_base_node.dig?("simpleText") || description_base_node
tags = initdata.dig?("header", "interactiveTabbedHeaderRenderer", "badges")
.try &.as_a.map(&.["metadataBadgeRenderer"]["label"].as_s) || [] of String
if ageGate = initdata.dig?("contents", "twoColumnBrowseResultsRenderer", "tabs", 0, "tabRenderer", "content", "sectionListRenderer", "contents", 0, "channelAgeGateRenderer")
description_node = nil
author = ageGate["channelTitle"].as_s
ucid = initdata["responseContext"]["serviceTrackingParams"][0]["params"][0]["value"].as_s
author_url = "https://www.youtube.com/channel/" + ucid
author_thumbnail = ageGate["avatar"]["thumbnails"][0]["url"].as_s
banner = nil
is_family_friendly = false
is_age_gated = true
tab_names = ["videos", "shorts", "streams"]
auto_generated = false
else
author = initdata["metadata"]["channelMetadataRenderer"]["title"].as_s
author_url = initdata["metadata"]["channelMetadataRenderer"]["channelUrl"].as_s
author_thumbnail = initdata["metadata"]["channelMetadataRenderer"]["avatar"]["thumbnails"][0]["url"].as_s
author_verified = has_verified_badge?(initdata.dig?("header", "c4TabbedHeaderRenderer", "badges"))
if auto_generated
author = initdata["header"]["interactiveTabbedHeaderRenderer"]["title"]["simpleText"].as_s
author_url = initdata["microformat"]["microformatDataRenderer"]["urlCanonical"].as_s
author_thumbnail = initdata["header"]["interactiveTabbedHeaderRenderer"]["boxArt"]["thumbnails"][0]["url"].as_s
ucid = initdata["metadata"]["channelMetadataRenderer"]["externalId"].as_s
# Raises a KeyError on failure.
banners = initdata["header"]["interactiveTabbedHeaderRenderer"]?.try &.["banner"]?.try &.["thumbnails"]?
banner = banners.try &.[-1]?.try &.["url"].as_s?
# Raises a KeyError on failure.
banners = initdata["header"]["c4TabbedHeaderRenderer"]?.try &.["banner"]?.try &.["thumbnails"]?
banners ||= initdata.dig?("header", "pageHeaderRenderer", "content", "pageHeaderViewModel", "banner", "imageBannerViewModel", "image", "sources")
banner = banners.try &.[-1]?.try &.["url"].as_s?
description_base_node = initdata["header"]["interactiveTabbedHeaderRenderer"]["description"]
# some channels have the description in a simpleText
# ex: https://www.youtube.com/channel/UCQvWX73GQygcwXOTSf_VDVg/
description_node = description_base_node.dig?("simpleText") || description_base_node
# if banner.includes? "channels/c4/default_banner"
# banner = nil
# end
tags = initdata.dig?("header", "interactiveTabbedHeaderRenderer", "badges")
.try &.as_a.map(&.["metadataBadgeRenderer"]["label"].as_s) || [] of String
else
author = initdata["metadata"]["channelMetadataRenderer"]["title"].as_s
author_url = initdata["metadata"]["channelMetadataRenderer"]["channelUrl"].as_s
author_thumbnail = initdata["metadata"]["channelMetadataRenderer"]["avatar"]["thumbnails"][0]["url"].as_s
author_verified = has_verified_badge?(initdata.dig?("header", "c4TabbedHeaderRenderer", "badges"))
description_node = initdata["metadata"]["channelMetadataRenderer"]?.try &.["description"]?
tags = initdata.dig?("microformat", "microformatDataRenderer", "tags").try &.as_a.map(&.as_s) || [] of String
ucid = initdata["metadata"]["channelMetadataRenderer"]["externalId"].as_s
# Raises a KeyError on failure.
banners = initdata["header"]["c4TabbedHeaderRenderer"]?.try &.["banner"]?.try &.["thumbnails"]?
banners ||= initdata.dig?("header", "pageHeaderRenderer", "content", "pageHeaderViewModel", "banner", "imageBannerViewModel", "image", "sources")
banner = banners.try &.[-1]?.try &.["url"].as_s?
# if banner.includes? "channels/c4/default_banner"
# banner = nil
# end
description_node = initdata["metadata"]["channelMetadataRenderer"]?.try &.["description"]?
tags = initdata.dig?("microformat", "microformatDataRenderer", "tags").try &.as_a.map(&.as_s) || [] of String
end
is_family_friendly = initdata["microformat"]["microformatDataRenderer"]["familySafe"].as_bool
if tabs_json = initdata["contents"]["twoColumnBrowseResultsRenderer"]["tabs"]?
# Get the name of the tabs available on this channel
tab_names = tabs_json.as_a.compact_map do |entry|
name = entry.dig?("tabRenderer", "title").try &.as_s.downcase
# This is a small fix to not add extra code on the HTML side
# I.e, the URL for the "live" tab is .../streams, so use "streams"
# everywhere for the sake of simplicity
(name == "live") ? "streams" : name
end
# Get the currently active tab ("About")
about_tab = extract_selected_tab(tabs_json)
# Try to find the about metadata section
channel_about_meta = about_tab.dig?(
"content",
"sectionListRenderer", "contents", 0,
"itemSectionRenderer", "contents", 0,
"channelAboutFullMetadataRenderer"
)
if !channel_about_meta.nil?
total_views = channel_about_meta.dig?("viewCountText", "simpleText").try &.as_s.gsub(/\D/, "").to_i64? || 0_i64
# The joined text is split to several sub strings. The reduce joins those strings before parsing the date.
joined = extract_text(channel_about_meta["joinedDateText"]?)
.try { |text| Time.parse(text, "Joined %b %-d, %Y", Time::Location.local) } || Time.unix(0)
# Normal Auto-generated channels
# https://support.google.com/youtube/answer/2579942
# For auto-generated channels, channel_about_meta only has
# ["description"]["simpleText"] and ["primaryLinks"][0]["title"]["simpleText"]
auto_generated = (
(channel_about_meta["primaryLinks"]?.try &.size) == 1 && \
extract_text(channel_about_meta.dig?("primaryLinks", 0, "title")) == "Auto-generated by YouTube" ||
channel_about_meta.dig?("links", 0, "channelExternalLinkViewModel", "title", "content").try &.as_s == "Auto-generated by YouTube"
)
end
end
end
is_family_friendly = initdata["microformat"]["microformatDataRenderer"]["familySafe"].as_bool
allowed_regions = initdata
.dig?("microformat", "microformatDataRenderer", "availableCountries")
.try &.as_a.map(&.as_s) || [] of String
@@ -102,52 +159,6 @@ def get_about_info(ucid, locale) : AboutChannel
end
end
total_views = 0_i64
joined = Time.unix(0)
tab_names = [] of String
if tabs_json = initdata["contents"]["twoColumnBrowseResultsRenderer"]["tabs"]?
# Get the name of the tabs available on this channel
tab_names = tabs_json.as_a.compact_map do |entry|
name = entry.dig?("tabRenderer", "title").try &.as_s.downcase
# This is a small fix to not add extra code on the HTML side
# I.e, the URL for the "live" tab is .../streams, so use "streams"
# everywhere for the sake of simplicity
(name == "live") ? "streams" : name
end
# Get the currently active tab ("About")
about_tab = extract_selected_tab(tabs_json)
# Try to find the about metadata section
channel_about_meta = about_tab.dig?(
"content",
"sectionListRenderer", "contents", 0,
"itemSectionRenderer", "contents", 0,
"channelAboutFullMetadataRenderer"
)
if !channel_about_meta.nil?
total_views = channel_about_meta.dig?("viewCountText", "simpleText").try &.as_s.gsub(/\D/, "").to_i64? || 0_i64
# The joined text is split to several sub strings. The reduce joins those strings before parsing the date.
joined = extract_text(channel_about_meta["joinedDateText"]?)
.try { |text| Time.parse(text, "Joined %b %-d, %Y", Time::Location.local) } || Time.unix(0)
# Normal Auto-generated channels
# https://support.google.com/youtube/answer/2579942
# For auto-generated channels, channel_about_meta only has
# ["description"]["simpleText"] and ["primaryLinks"][0]["title"]["simpleText"]
auto_generated = (
(channel_about_meta["primaryLinks"]?.try &.size) == 1 && \
extract_text(channel_about_meta.dig?("primaryLinks", 0, "title")) == "Auto-generated by YouTube" ||
channel_about_meta.dig?("links", 0, "channelExternalLinkViewModel", "title", "content").try &.as_s == "Auto-generated by YouTube"
)
end
end
sub_count = 0
if (metadata_rows = initdata.dig?("header", "pageHeaderRenderer", "content", "pageHeaderViewModel", "metadata", "contentMetadataViewModel", "metadataRows").try &.as_a)
@@ -177,6 +188,7 @@ def get_about_info(ucid, locale) : AboutChannel
tabs: tab_names,
tags: tags,
verified: author_verified || false,
is_age_gated: is_age_gated || false,
)
end