Merge pull request #5032 from nonamethanks/factor-out-weibo

sources: factor out Source::URL::Weibo
This commit is contained in:
evazion
2022-03-07 18:31:15 -06:00
committed by GitHub
4 changed files with 119 additions and 147 deletions

View File

@@ -31,6 +31,7 @@ module Source
Source::URL::Plurk,
Source::URL::Skeb,
Source::URL::TwitPic,
Source::URL::Weibo,
]
# Parse a URL into a subclass of Source::URL, or raise an exception if the URL is not a valid HTTP or HTTPS URL.

View File

@@ -0,0 +1,95 @@
# frozen_string_literal: true
class Source::URL::Weibo < Source::URL
attr_reader :full_image_url
def self.match?(url)
url.domain.in?(["weibo.com", "weibo.cn", "sinaimg.cn"])
end
def parse
case [host, *path_segments]
# http://ww1.sinaimg.cn/large/69917555gw1f6ggdghk28j20c87lbhdt.jpg
# https://wx1.sinaimg.cn/large/002NQ2vhly1gqzqfk1agfj62981aw4qr02.jpg
# http://ww4.sinaimg.cn/mw690/77a2d531gw1f4u411ws3aj20m816fagg.jpg (sample)
# https://wx4.sinaimg.cn/orj360/e3930166gy1g546bz86cij20u00u040y.jpg (sample)
# http://ww3.sinaimg.cn/mw1024/0065kjmOgw1fabcanrzx6j30f00lcjwv.jpg (sample)
in /\w+\.sinaimg\.cn/ => host, size, file
@full_image_url = "https://#{host}/large/#{file}"
# http://tw.weibo.com/1300957955/3786333853668537
in "tw.weibo.com", /\w+/, /\d+/ => illust_long_id
@illust_long_id = illust_long_id
# http://weibo.com/3357910224/EEHA1AyJP
# https://www.weibo.com/5501756072/IF9fugHzj?from=page_1005055501756072_profile&wvr=6&mod=weibotime
in /(\w+\.)?weibo\.(com|cn)/, /\d+/ => artist_short_id, /\w+/ => illust_base62_id
@artist_short_id = artist_short_id
@illust_base62_id = illust_base62_id
# http://photo.weibo.com/2125874520/wbphotos/large/mid/4194742441135220/pid/7eb64558gy1fnbryb5nzoj20dw10419t
# http://photo.weibo.com/5732523783/talbum/detail/photo_id/4029784374069389?prel=p6_3
in "photo.weibo.com", /\d+/ => artist_short_id, _, _, _, /\d+/ => illust_long_id, *rest
@artist_short_id = artist_short_id
@illust_long_id = illust_long_id
# https://m.weibo.cn/detail/4506950043618873
in "m.weibo.cn", "detail", /\d+/ => illust_long_id
@illust_base62_id = illust_base62_id
# https://m.weibo.cn/status/J33G4tH1B
in "m.weibo.cn", "status", /\w+/ => illust_base62_id
@illust_base62_id = illust_base62_id
# https://www.weibo.com/u/5501756072
# https://m.weibo.cn/profile/5501756072
# https://m.weibo.cn/u/5501756072
in _, ("u" | "profile"), /\d+/ => artist_short_id
@artist_short_id = artist_short_id
# https://www.weibo.com/5501756072
in _, /\d+/ => artist_short_id
@artist_short_id = artist_short_id
in _, "p", /\d+/ => artist_long_id
@artist_long_id = artist_long_id
else
end
end
def image_url?
full_image_url.present?
end
def profile_urls
[profile_short_url, profile_long_url].compact
end
def profile_short_url
return if @artist_short_id.blank?
"https://www.weibo.com/u/#{@artist_short_id}"
end
def profile_long_url
return if @artist_long_id.blank?
"https://www.weibo.com/p/#{@artist_long_id}"
end
def mobile_url
if @illust_long_id.present?
"https://m.weibo.cn/detail/#{@illust_long_id}"
elsif @illust_base62_id.present?
"https://m.weibo.cn/status/#{@illust_base62_id}"
end
end
def normalized_url
if @artist_short_id.present? && @illust_base62_id.present?
"https://www.weibo.com/#{@artist_short_id}/#{@illust_base62_id}"
elsif mobile_url.present?
mobile_url
end
end
end

View File

@@ -1,91 +1,35 @@
# frozen_string_literal: true
# Image URLS
# * http://ww1.sinaimg.cn/large/69917555gw1f6ggdghk28j20c87lbhdt.jpg
# * https://wx1.sinaimg.cn/large/002NQ2vhly1gqzqfk1agfj62981aw4qr02.jpg (more than 32 characters in hash)
#
# Image Samples
# * http://ww4.sinaimg.cn/mw690/77a2d531gw1f4u411ws3aj20m816fagg.jpg
# * https://wx4.sinaimg.cn/orj360/e3930166gy1g546bz86cij20u00u040y.jpg
# * http://ww3.sinaimg.cn/mw1024/0065kjmOgw1fabcanrzx6j30f00lcjwv.jpg
#
# Page URLS
# * http://weibo.com/3357910224/EEHA1AyJP
# * https://www.weibo.com/5501756072/IF9fugHzj?from=page_1005055501756072_profile&wvr=6&mod=weibotime
#
# * http://photo.weibo.com/5732523783/talbum/detail/photo_id/4029784374069389?prel=p6_3
# * http://photo.weibo.com/2125874520/wbphotos/large/mid/4194742441135220/pid/7eb64558gy1fnbryb5nzoj20dw10419t
# * http://tw.weibo.com/1300957955/3786333853668537
#
# * https://m.weibo.cn/detail/4506950043618873
# * https://m.weibo.cn/status/J33G4tH1B
#
# Video
# * https://www.weibo.com/5501756072/IF9fugHzj
#
# Profile URLS
# ### Short ID
# * https://www.weibo.com/5501756072
# * https://www.weibo.com/u/5501756072
# * https://m.weibo.cn/profile/5501756072
# * https://m.weibo.cn/u/5501756072
# ### Long ID
# * https://www.weibo.com/p/1005055501756072
# @see Source::URL::Weibo
module Sources
module Strategies
class Weibo < Base
PROFILE_URL_1 = %r{\Ahttps?://(?:(?:www|m)\.)?weibo\.c(?:om|n)/(?:(?:u|profile)/)?(?<artist_short_id>\d+)\z}i
PROFILE_URL_2 = %r{\Ahttps?://photo\.weibo\.com/(?<artist_short_id>\d+)}i
PROFILE_URL_3 = %r{\Ahttps?://(?:www\.)?weibo\.com/p/(?<artist_long_id>\d+)}i
PAGE_URL_1 = %r{\Ahttps?://(?:www\.)?weibo\.com/(?<artist_short_id>\d+)/(?<illust_base62_id>\w+)(?:\?.*)?\z}i
PAGE_URL_2 = %r{#{PROFILE_URL_2}/(?:wbphotos/large/mid|talbum/detail/photo_id)/(?<illust_long_id>\d+)(?:/pid/(?<image_id>\w{32}))?}i
PAGE_URL_3 = %r{\Ahttps?://m\.weibo\.cn/(?:detail/(?<illust_long_id>\d+)|status/(?<illust_base62_id>\w+))}i
PAGE_URL_4 = %r{\Ahttps?://tw\.weibo\.com/(?:(?<artist_short_id>\d+)|\w+)/(?<illust_long_id>\d+)}i
IMAGE_URL = %r{\Ahttps?://\w+\.sinaimg\.cn/\w+/(?<image_id>\w+)\.}i
def domains
["weibo.com", "weibo.cn", "weibocdn.com", "sinaimg.cn"]
def match?
Source::URL::Weibo === parsed_url
end
def site_name
"Weibo"
parsed_url.site_name
end
def image_urls
urls = []
if url =~ IMAGE_URL
urls << self.class.convert_image_to_large(url)
if parsed_url.image_url?
[parsed_url.full_image_url]
elsif api_response.present?
if api_response["pics"].present?
urls += api_response["pics"].to_a.map { |pic| self.class.convert_image_to_large(pic["url"]) }
api_response["pics"].pluck("url").map { |url| Source::URL.parse(url).full_image_url }
elsif api_response.dig("page_info", "type") == "video"
variants = api_response["page_info"]["media_info"].to_h.values + api_response["page_info"]["urls"].to_h.values
urls << variants.max_by do |variant|
largest_video = variants.max_by do |variant|
if /template=(?<width>\d+)x(?<height>\d+)/ =~ variant.to_s
width.to_i * height.to_i
else
0
end
end
[largest_video]
end
else
urls << url
end
urls
end
def image_url
image_id = url[PAGE_URL_2, :image_id] if url =~ PAGE_URL_2
if image_id.present?
image_urls.select { |i| i[IMAGE_URL, :image_id] == image_id }.compact.first
else
image_urls.first
end
end
@@ -94,47 +38,30 @@ module Sources
end
def page_url
if api_response.present?
artist_id = api_response["user"]["id"]
illust_id = api_response["bid"]
"https://www.weibo.com/#{artist_id}/#{illust_id}"
elsif url =~ IMAGE_URL
self.class.convert_image_to_large(url)
else
url
end
return nil unless api_response.present?
artist_id = api_response["user"]["id"]
illust_base62_id = api_response["bid"]
"https://www.weibo.com/#{artist_id}/#{illust_base62_id}"
end
def tags
return [] if api_response.blank?
matches = api_response["text"]&.scan(/surl-text">#(.*?)#</).to_a.map { |m| m[0] }
matches.map do |match|
[match, "https://s.weibo.com/weibo/#{match}"]
end
end
def profile_urls
[profile_short_url, profile_long_url].compact
(parsed_url.profile_urls + parsed_referer&.profile_urls.to_a).uniq
end
def profile_url
profile_urls.first
end
def profile_short_url
return if artist_short_id.blank?
"https://www.weibo.com/u/#{artist_short_id}"
end
def profile_long_url
return if artist_long_id.blank?
"https://www.weibo.com/p/#{artist_long_id}"
end
def artist_name
api_response&.dig("user", "screen_name")
end
@@ -163,54 +90,11 @@ module Sources
end
def normalize_for_source
return url if url =~ PAGE_URL_2
artist_id = artist_short_id_from_url
if artist_id.present?
if illust_base62_id.present?
"https://www.weibo.com/#{artist_id}/#{illust_base62_id}"
elsif illust_long_id.present?
"https://photo.weibo.com/#{artist_id}/talbum/detail/photo_id/#{illust_long_id}"
end
elsif mobile_url.present?
mobile_url
end
end
def self.convert_image_to_large(url)
url.gsub(%r{.cn/\w+/(\w+)}, '.cn/large/\1')
end
def illust_long_id
urls.map { |x| x[PAGE_URL_2, :illust_long_id] || x[PAGE_URL_3, :illust_long_id] || x[PAGE_URL_4, :illust_long_id] }.compact.first
end
def illust_base62_id
urls.map { |x| x[PAGE_URL_1, :illust_base62_id] || x[PAGE_URL_3, :illust_base62_id] }.compact.first
end
def artist_short_id_from_url
urls.map { |x| x[PROFILE_URL_1, :artist_short_id] || x[PROFILE_URL_2, :artist_short_id] || x[PAGE_URL_1, :artist_short_id] || x[PAGE_URL_4, :artist_short_id] }.compact.first
end
def artist_short_id
artist_short_id_from_url || api_response&.dig("user", "id")
end
def artist_long_id
urls.map { |x| x[PROFILE_URL_3, :artist_long_id] }.compact.first
end
def mobile_url
if illust_long_id.present?
"https://m.weibo.cn/detail/#{illust_long_id}"
elsif illust_base62_id.present?
"https://m.weibo.cn/status/#{illust_base62_id}"
end
parsed_url.normalized_url
end
def api_response
return {} if mobile_url.blank?
return {} if (mobile_url = parsed_url.mobile_url || parsed_referer&.mobile_url).blank?
resp = http.cache(1.minute).get(mobile_url)
json_string = resp.to_s[/var \$render_data = \[(.*)\]\[0\]/m, 1]

View File

@@ -1,4 +1,4 @@
require 'test_helper'
require "test_helper"
module Sources
class WeiboTest < ActiveSupport::TestCase
@@ -40,7 +40,7 @@ module Sources
should "get the tags" do
tags = [
%w[fgo https://s.weibo.com/weibo/fgo],
%w[Alter组 https://s.weibo.com/weibo/Alter组]
%w[Alter组 https://s.weibo.com/weibo/Alter组],
]
assert_equal(tags, @site.tags)
end
@@ -76,18 +76,10 @@ module Sources
end
end
context "An album url for a post with multiple pictures" do
should "upload the right picture rather than just the first" do
site = Sources::Strategies.find("http://photo.weibo.com/2125874520/wbphotos/large/mid/4194742441135220/pid/7eb64558gy1fnbryb5nzoj20dw10419t")
assert_equal("https://wx4.sinaimg.cn/large/7eb64558gy1fnbryb5nzoj20dw10419t.jpg", site.image_url)
end
end
context "An upload from the batch bookmarklet" do
context "A multi-page upload" do
should "set the right source" do
url = "https://wx1.sinaimg.cn/large/7eb64558gy1fnbryriihwj20dw104wtu.jpg"
ref = "http://photo.weibo.com/2125874520/wbphotos/large/mid/4194742441135220/pid/7eb64558gy1fnbryb5nzoj20dw10419t"
ref = "https://photo.weibo.com/2125874520/wbphotos/large/mid/4194742441135220/pid/7eb64558gy1fnbryb5nzoj20dw10419t"
site = Sources::Strategies.find(url, ref)
assert_equal("https://www.weibo.com/2125874520/FDKGo4Lk0", site.canonical_url)
@@ -97,12 +89,12 @@ module Sources
context "normalizing for source" do
should "normalize correctly" do
source1 = "https://www.weibo.com/3150932560/H4cFbeKKA?from=page_1005053150932560_profile&wvr=6&mod=weibotime"
source2 = "http://photo.weibo.com/2125874520/wbphotos/large/mid/4242129997905387/pid/7eb64558ly1friyzhj44lj20dw2qxe81"
source2 = "https://photo.weibo.com/2125874520/wbphotos/large/mid/4242129997905387/pid/7eb64558ly1friyzhj44lj20dw2qxe81"
source3 = "https://m.weibo.cn/status/4173757483008088?luicode=20000061&lfid=4170879204256635"
source4 = "https://tw.weibo.com/SEINEN/4098035921690224"
assert_equal("https://www.weibo.com/3150932560/H4cFbeKKA", Sources::Strategies.normalize_source(source1))
assert_equal(source2, Sources::Strategies.normalize_source(source2))
assert_equal("https://m.weibo.cn/detail/4242129997905387", Sources::Strategies.normalize_source(source2))
assert_equal("https://m.weibo.cn/status/4173757483008088", Sources::Strategies.normalize_source(source3))
assert_equal("https://m.weibo.cn/detail/4098035921690224", Sources::Strategies.normalize_source(source4))
end