panchira 1.3.2 → 1.3.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 4b6ab5f09a68a55d7fa54720b6ed5544605c5b4fa25ff839583058ac3af5611b
-  data.tar.gz: 22814f2437ac515c59d22a634c4709212eb639735a831157b7a64375559c1cf1
+  metadata.gz: 56602175dff2975d0d6f606c31da48b8cf904c61aa98d8f0e6d940ef4c1d230d
+  data.tar.gz: b1ceb998bb60dafbb5b992ad156192d6ea2f58fec8ee606aac3d1d88ebe9b1b0
 SHA512:
-  metadata.gz: a2a42f6023db6ce19d076079044c6fd88e49870478590d89ad25810dd3b72e72439627b416bd3d439d78ab8eaa72683746617fe565f441e69a758811e4b50023
-  data.tar.gz: 6f4573bf646ea21d9d5887ee67a1e95e275e6b1de6d152d4c096c7b8a4e70fa29409143a8a00af0c32f6a1e023f4b0a5c84a30d50fd789dc5f6822cb161cd0ed
+  metadata.gz: 5909d32a1231288cc8567ada0a58fb038d1eaf8424ac9ff3b4298dd81b3f95651217ddb3b5b39e24834174abbc7bfd454d083836a0889df2188df511600f7a49
+  data.tar.gz: 0f8d01c68ceb44f10ef06d274667e8acd254254987056e25624d2a3f58de0e05fc95a07002b3cb7d0d8a207461bf6f0a96c211714fed015eadc3e9673eb17699

data/.github/workflows/ruby.yml

@@ -16,7 +16,7 @@ on:
 jobs:
   test:
 
-    runs-on: ubuntu-latest
+    runs-on: ubuntu-18.04
 
     steps:
     - uses: actions/checkout@v2

data/CHANGELOG.md

@@ -4,10 +4,35 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](http://keepachangelog.com/)
 and this project adheres to [Semantic Versioning](http://semver.org/).
 
+## 1.3.6 - 2021-10-19
+### Added
+- Added support for Twitter.
+
+### Changed
+- Melonbooks Resolver now returns image URI without resizing queries.
+
+## 1.3.5 - 2021-09-01
+### Added
+- Added support for Pixiv Novel.
+
+### Fixed
+- Fixed an issue where Image Resolver missed image scales.
+
+## 1.3.4 - 2021-07-26
+### Fixed
+- Fixed an issue where Iwara Resolver failed when a description was present.
+
+## 1.3.3 - 2021-07-25
+### Added
+- Added support for Iwara.
+
+### Fixed
+- Fixed an issue where DLsite Resolver was retrieving wrong tags.
+
 ## 1.3.2 - 2021-05-23
 ### Fixed
-- Fixed an issue where Fanza Resolver was retrieving incorrect cannonical URLs from meta tags.
-- Fixed an issue where Narou Resolver retrieved wrong descriptions from meta tags.
+- Fixed an issue where Fanza Resolver was retrieving incorrect canonical URLs.
+- Fixed an issue where Narou Resolver was retrieving wrong descriptions.
 
 ### Changed
 - Updated dependencies.
@@ -79,6 +104,10 @@ and this project adheres to [Semantic Versioning](http://semver.org/).
 ### Added
 - Released Panchira gem. At this time we can parse only 5 websites.
 
+[1.3.5]: https://github.com/nuita/panchira/releases/tag/v1.3.5
+[1.3.4]: https://github.com/nuita/panchira/releases/tag/v1.3.4
+[1.3.3]: https://github.com/nuita/panchira/releases/tag/v1.3.3
+[1.3.2]: https://github.com/nuita/panchira/releases/tag/v1.3.2
 [1.3.1]: https://github.com/nuita/panchira/releases/tag/v1.3.1
 [1.3.0]: https://github.com/nuita/panchira/releases/tag/v1.3.0
 [1.2.0]: https://github.com/nuita/panchira/releases/tag/v1.2.0

data/Gemfile.lock

@@ -1,9 +1,9 @@
 PATH
   remote: .
   specs:
-    panchira (1.3.2)
+    panchira (1.3.6)
       fastimage (~> 2.1.7)
-      nokogiri (>= 1.10.9, < 1.12.0)
+      nokogiri (>= 1.10.9, < 1.13.0)
 
 GEM
   remote: https://rubygems.org/
@@ -11,7 +11,7 @@ GEM
     ast (2.4.2)
     fastimage (2.1.7)
     minitest (5.14.4)
-    nokogiri (1.11.5-x86_64-darwin)
+    nokogiri (1.11.7-x86_64-darwin)
       racc (~> 1.4)
     parallel (1.20.1)
     parser (3.0.1.1)

data/README.md

@@ -6,7 +6,7 @@
 
 Due to some legal or ethical issues, most hentai and NSFW platforms don't clarify their content on meta tags. As a result, most hentai platforms are rendered poorly on the card previews on social media.
 
-To solve this issue, Panchira is made to parse correct and uncensored metadata from such web platforms (at this time we cover **DLSite, Komiflo, Melonbooks, Nijie, Pixiv, Shousetsuka ni narou, Fanza and Twitter**).
+To solve this issue, Panchira is made to parse correct and uncensored metadata from such web platforms (at this time we cover **DLSite, Komiflo, Melonbooks, Nijie, Pixiv, Shousetsuka ni narou, Fanza, Iwara and Twitter**).
 
 If you need card previews of hentai on your web application, but can't get them with simply parsing metatags, then it is time for Panchira.
 

data/lib/panchira/resolvers/dlsite_resolver.rb

@@ -39,7 +39,13 @@ module Panchira
       end
 
       def parse_tags
-        @page.css('.main_genre').children.children.map(&:text)
+        @page.css('table[id*="work_"] tr').each do |tr|
+          next unless tr.css('th').text =~ /ジャンル/
+
+          return tr.css('td a').map do |node|
+            node.text.strip
+          end
+        end
       end
   end
 

data/lib/panchira/resolvers/image_resolver.rb

@@ -9,6 +9,8 @@ module Panchira
       result.canonical_url = @url
       result.image = PanchiraImage.new
       result.image.url = @url
+      result.image.width, result.image.height = FastImage.size(result.image.url)
+
       result.resolver = parse_resolver
       result
     end

data/lib/panchira/resolvers/iwara_resolver.rb

@@ -0,0 +1,39 @@
+# frozen_string_literal: true
+
+module Panchira
+  class IwaraResolver < Resolver
+    URL_REGEXP = /(www|ecchi)\.iwara\.tv\//.freeze
+
+    private
+      def parse_title
+        super.split(' | ')[0]
+      end
+
+      def parse_image_url
+        url = @page.at_css('#video-player')&.attributes['poster']&.value
+        'https:' + url if url
+      end
+
+      def parse_author
+        @page.at_css('.node-info .username')&.children&.[](0)&.text
+      end
+
+      def parse_tags
+        @page.css('.field-name-field-categories .field-item').map { |e| e.children&.text }
+      end
+
+      def parse_description
+        @page.at_css('.field-name-body')&.text
+      end
+
+      def cookie
+        'show_adult=1'
+      end
+
+      def parse_canonical_url
+        @url # canonical has relative path. ignore it
+      end
+  end
+
+  ::Panchira::Extensions.register(Panchira::IwaraResolver)
+end

data/lib/panchira/resolvers/melonbooks_resolver.rb

@@ -57,7 +57,10 @@ module Panchira
       end
 
       def parse_image_url
-        @page.css('//meta[property="og:image"]/@content').first.to_s.sub(/&c=1/, '')
+        url = @page.css('//meta[property="og:image"]/@content').first.to_s
+        image = url.match(/resize_image.php\?image=([^&]+)/)[1]
+
+        "https://melonbooks.akamaized.net/user_data/packages/resize_image.php?image=#{image}"
       end
 
       def parse_tags

data/lib/panchira/resolvers/narou_resolver.rb

@@ -37,6 +37,11 @@ module Panchira
         # つらい。
         @desc&.xpath('//*[@id="noveltable1"]/tr[3]')&.text&.split("\n\n\n")&.dig(1)&.split(' ')
       end
+
+      # og:urlで指定されたncode.syosetu.com/~~~にアクセスすると301で戻されるので何もしない
+      def parse_canonical_url
+        @url
+      end
     end
 
     class NcodeResolver < Resolver
@@ -63,6 +68,11 @@ module Panchira
         # めっちゃつらい。
         @desc&.xpath('//*[@id="noveltable1"]/tr[3]')&.text&.split("\n\n\n")&.dig(1)&.delete("\u00A0")&.split(' ')&.grep_v('')
       end
+
+      # og:urlで指定されたncode.syosetu.com/~~~にアクセスすると301で戻されるので何もしない
+      def parse_canonical_url
+        @url
+      end
     end
   end
 

data/lib/panchira/resolvers/pixiv_resolver.rb

@@ -44,5 +44,32 @@ module Panchira
       end
   end
 
+  class PixivNovelResolver < Resolver
+    URL_REGEXP = %r{pixiv\.net/novel/show.php\?id=(\d+)}.freeze
+
+    def initialize(url)
+      super(url)
+      @novel_id = url.slice(URL_REGEXP, 1)
+
+      raw_json = URI.parse("https://www.pixiv.net/ajax/novel/#{@novel_id}").read('User-Agent' => user_agent)
+      @json = JSON.parse(raw_json)
+    end
+
+    private
+
+      def parse_title
+        @json['body']['title']
+      end
+
+      def parse_author
+        @json['body']['userName']
+      end
+
+      def parse_tags
+        @json['body']['tags']['tags'].map { |content| content['tag'] }
+      end
+  end
+
   ::Panchira::Extensions.register(Panchira::PixivResolver)
+  ::Panchira::Extensions.register(Panchira::PixivNovelResolver)
 end

data/lib/panchira/resolvers/resolver.rb

@@ -67,17 +67,20 @@ module Panchira
         # fetch page and refresh canonical_url until canonical_url converges.
         loop do
           url_in_res = @page.css('//link[rel="canonical"]/@href').to_s
+          if url_in_res.empty?
+            url_in_res = @page.css('//meta[property="og:url"]/@content').to_s
+          end
 
           if url_in_res.empty?
             return history.last || @url
-          else
-            if history.include?(url_in_res) || history.length > 5
-              return url_in_res
-            else
-              history.push(url_in_res)
-              @page = fetch_page(url_in_res)
-            end
           end
+
+          if history.include?(url_in_res) || history.length > 5
+            return url_in_res
+          end
+
+          history.push(url_in_res)
+          @page = fetch_page(url_in_res)
         end
       end
 

data/lib/panchira/resolvers/twitter_resolver.rb

@@ -0,0 +1,24 @@
+module Panchira
+  class TwitterResolver < Resolver
+    URL_REGEXP = /twitter.com\/\w+\/status\/\d+/.freeze
+
+    private
+      def parse_title
+        @title = super
+      end
+
+      def parse_author
+        @title.match(/\A(.+) on Twitter\z/)[1]
+      end
+
+      def parse_description
+        @description = super.gsub(/\A“|”\z/, '')
+      end
+
+      def parse_tags
+        @description.scan(/[#＃]([^#＃\s]+)/).map(&:first)
+      end
+  end
+
+  ::Panchira::Extensions.register(Panchira::TwitterResolver)
+end

data/lib/panchira/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Panchira
-  VERSION = '1.3.2'
+  VERSION = '1.3.6'
 end

data/panchira.gemspec

@@ -40,5 +40,5 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency 'rubocop-minitest', '~> 0.10'
 
   spec.add_dependency 'fastimage', '~> 2.1.7'
-  spec.add_dependency 'nokogiri', '>= 1.10.9', '< 1.12.0'
+  spec.add_dependency 'nokogiri', '>= 1.10.9', '< 1.13.0'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: panchira
 version: !ruby/object:Gem::Version
-  version: 1.3.2
+  version: 1.3.6
 platform: ruby
 authors:
 - kyp
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2021-05-23 00:00:00.000000000 Z
+date: 2021-10-19 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -103,7 +103,7 @@ dependencies:
         version: 1.10.9
     - - "<"
       - !ruby/object:Gem::Version
-        version: 1.12.0
+        version: 1.13.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -113,7 +113,7 @@ dependencies:
         version: 1.10.9
     - - "<"
       - !ruby/object:Gem::Version
-        version: 1.12.0
+        version: 1.13.0
 description: |2
       Panchira allows you to parse attributes of hentais on some web platforms, such as Pixiv and DLSite.
       If you need card previews on hentai but can't get it with simply parsing metatags, then it is time for Panchira.
@@ -141,12 +141,14 @@ files:
 - lib/panchira/resolvers/dlsite_resolver.rb
 - lib/panchira/resolvers/fanza_resolver.rb
 - lib/panchira/resolvers/image_resolver.rb
+- lib/panchira/resolvers/iwara_resolver.rb
 - lib/panchira/resolvers/komiflo_resolver.rb
 - lib/panchira/resolvers/melonbooks_resolver.rb
 - lib/panchira/resolvers/narou_resolver.rb
 - lib/panchira/resolvers/nijie_resolver.rb
 - lib/panchira/resolvers/pixiv_resolver.rb
 - lib/panchira/resolvers/resolver.rb
+- lib/panchira/resolvers/twitter_resolver.rb
 - lib/panchira/version.rb
 - panchira.gemspec
 homepage: https://github.com/nuita/panchira