Rewrite extract_item and extract_items functions

This commit completely rewrites the extract_item and extract_items
function. Before this commit these two function were an unreadable
mess. The extract_item function was a lengthy if-elsif chain
while the extract_items function contained an incomprehensible
mess of .try, else and ||.

With this commit both of these functions have been pulled into a
separate file with the internal logic being moved to a few classes.

This significantly reduces the size of these two methods, enhances
readability and makes adding new extraction/parse rules much simpler.

See diff for details.

--
This cherry-picked commit also removes the code for parsing featured
channels present on the original.

(cherry picked from commit a027fbf7af1f96dc26fe5a610525ae52bcc40c28)
This commit is contained in:
syeopite
2021-05-04 01:48:51 -07:00
parent 6d68fbc31d
commit 1323b94b7a
2 changed files with 320 additions and 159 deletions

View File

@@ -251,165 +251,9 @@ def extract_videos(initial_data : Hash(String, JSON::Any), author_fallback : Str
extract_items(initial_data, author_fallback, author_id_fallback).select(&.is_a?(SearchVideo)).map(&.as(SearchVideo))
end
def extract_item(item : JSON::Any, author_fallback : String? = nil, author_id_fallback : String? = nil)
if i = (item["videoRenderer"]? || item["gridVideoRenderer"]?)
video_id = i["videoId"].as_s
title = i["title"].try { |t| t["simpleText"]?.try &.as_s || t["runs"]?.try &.as_a.map(&.["text"].as_s).join("") } || ""
author_info = i["ownerText"]?.try &.["runs"]?.try &.as_a?.try &.[0]?
author = author_info.try &.["text"].as_s || author_fallback || ""
author_id = author_info.try &.["navigationEndpoint"]?.try &.["browseEndpoint"]["browseId"].as_s || author_id_fallback || ""
published = i["publishedTimeText"]?.try &.["simpleText"]?.try { |t| decode_date(t.as_s) } || Time.local
view_count = i["viewCountText"]?.try &.["simpleText"]?.try &.as_s.gsub(/\D+/, "").to_i64? || 0_i64
description_html = i["descriptionSnippet"]?.try { |t| parse_content(t) } || ""
length_seconds = i["lengthText"]?.try &.["simpleText"]?.try &.as_s.try { |t| decode_length_seconds(t) } ||
i["thumbnailOverlays"]?.try &.as_a.find(&.["thumbnailOverlayTimeStatusRenderer"]?).try &.["thumbnailOverlayTimeStatusRenderer"]?
.try &.["text"]?.try &.["simpleText"]?.try &.as_s.try { |t| decode_length_seconds(t) } || 0
live_now = false
premium = false
premiere_timestamp = i["upcomingEventData"]?.try &.["startTime"]?.try { |t| Time.unix(t.as_s.to_i64) }
i["badges"]?.try &.as_a.each do |badge|
b = badge["metadataBadgeRenderer"]
case b["label"].as_s
when "LIVE NOW"
live_now = true
when "New", "4K", "CC"
# TODO
when "Premium"
# TODO: Potentially available as i["topStandaloneBadge"]["metadataBadgeRenderer"]
premium = true
else nil # Ignore
end
end
SearchVideo.new({
title: title,
id: video_id,
author: author,
ucid: author_id,
published: published,
views: view_count,
description_html: description_html,
length_seconds: length_seconds,
live_now: live_now,
premium: premium,
premiere_timestamp: premiere_timestamp,
})
elsif i = item["channelRenderer"]?
author = i["title"]["simpleText"]?.try &.as_s || author_fallback || ""
author_id = i["channelId"]?.try &.as_s || author_id_fallback || ""
author_thumbnail = i["thumbnail"]["thumbnails"]?.try &.as_a[0]?.try &.["url"]?.try &.as_s || ""
subscriber_count = i["subscriberCountText"]?.try &.["simpleText"]?.try &.as_s.try { |s| short_text_to_number(s.split(" ")[0]) } || 0
auto_generated = false
auto_generated = true if !i["videoCountText"]?
video_count = i["videoCountText"]?.try &.["runs"].as_a[0]?.try &.["text"].as_s.gsub(/\D/, "").to_i || 0
description_html = i["descriptionSnippet"]?.try { |t| parse_content(t) } || ""
SearchChannel.new({
author: author,
ucid: author_id,
author_thumbnail: author_thumbnail,
subscriber_count: subscriber_count,
video_count: video_count,
description_html: description_html,
auto_generated: auto_generated,
})
elsif i = item["gridPlaylistRenderer"]?
title = i["title"]["runs"].as_a[0]?.try &.["text"].as_s || ""
plid = i["playlistId"]?.try &.as_s || ""
video_count = i["videoCountText"]["runs"].as_a[0]?.try &.["text"].as_s.gsub(/\D/, "").to_i || 0
playlist_thumbnail = i["thumbnail"]["thumbnails"][0]?.try &.["url"]?.try &.as_s || ""
SearchPlaylist.new({
title: title,
id: plid,
author: author_fallback || "",
ucid: author_id_fallback || "",
video_count: video_count,
videos: [] of SearchPlaylistVideo,
thumbnail: playlist_thumbnail,
})
elsif i = item["playlistRenderer"]?
title = i["title"]["simpleText"]?.try &.as_s || ""
plid = i["playlistId"]?.try &.as_s || ""
video_count = i["videoCount"]?.try &.as_s.to_i || 0
playlist_thumbnail = i["thumbnails"].as_a[0]?.try &.["thumbnails"]?.try &.as_a[0]?.try &.["url"].as_s || ""
author_info = i["shortBylineText"]?.try &.["runs"]?.try &.as_a?.try &.[0]?
author = author_info.try &.["text"].as_s || author_fallback || ""
author_id = author_info.try &.["navigationEndpoint"]?.try &.["browseEndpoint"]["browseId"].as_s || author_id_fallback || ""
videos = i["videos"]?.try &.as_a.map do |v|
v = v["childVideoRenderer"]
v_title = v["title"]["simpleText"]?.try &.as_s || ""
v_id = v["videoId"]?.try &.as_s || ""
v_length_seconds = v["lengthText"]?.try &.["simpleText"]?.try { |t| decode_length_seconds(t.as_s) } || 0
SearchPlaylistVideo.new({
title: v_title,
id: v_id,
length_seconds: v_length_seconds,
})
end || [] of SearchPlaylistVideo
# TODO: i["publishedTimeText"]?
SearchPlaylist.new({
title: title,
id: plid,
author: author,
ucid: author_id,
video_count: video_count,
videos: videos,
thumbnail: playlist_thumbnail,
})
elsif i = item["radioRenderer"]? # Mix
# TODO
elsif i = item["showRenderer"]? # Show
# TODO
elsif i = item["shelfRenderer"]?
elsif i = item["horizontalCardListRenderer"]?
elsif i = item["searchPyvRenderer"]? # Ad
end
end
def extract_items(initial_data : Hash(String, JSON::Any), author_fallback : String? = nil, author_id_fallback : String? = nil)
items = [] of SearchItem
channel_v2_response = initial_data
.try &.["continuationContents"]?
.try &.["gridContinuation"]?
.try &.["items"]?
if channel_v2_response
channel_v2_response.try &.as_a.each { |item|
extract_item(item, author_fallback, author_id_fallback)
.try { |t| items << t }
}
else
initial_data.try { |t| t["contents"]? || t["response"]? }
.try { |t| t["twoColumnBrowseResultsRenderer"]?.try &.["tabs"].as_a.select(&.["tabRenderer"]?.try &.["selected"].as_bool)[0]?.try &.["tabRenderer"]["content"] ||
t["twoColumnSearchResultsRenderer"]?.try &.["primaryContents"] ||
t["continuationContents"]? }
.try { |t| t["sectionListRenderer"]? || t["sectionListContinuation"]? }
.try &.["contents"].as_a
.each { |c| c.try &.["itemSectionRenderer"]?.try &.["contents"].as_a
.try { |t| t[0]?.try &.["shelfRenderer"]?.try &.["content"]["expandedShelfContentsRenderer"]?.try &.["items"].as_a ||
t[0]?.try &.["gridRenderer"]?.try &.["items"].as_a || t }
.each { |item|
extract_item(item, author_fallback, author_id_fallback)
.try { |t| items << t }
} }
end
items
def extract_selected_tab(tabs)
# Extract the selected tab from the array of tabs Youtube returns
return selected_target = tabs.as_a.select(&.["tabRenderer"]?.try &.["selected"].as_bool)[0]
end
def check_enum(db, enum_name, struct_type = nil)