bookshark 1.0.6 → 1.0.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 140903206c1308ba6eda3c1c1ee6a910be0a28b3
-  data.tar.gz: d1891f2c7ba96340201e5bd2936a07ca83032e4b
+SHA256:
+  metadata.gz: 78062355b9dd9c943d795914ad098c00c495079a6f5e8012901aa241d8a56277
+  data.tar.gz: 3413cdf39cef48e8e283a62cd4dd49b8aba30f8bf123459c9c3d05b4cfac5a52
 SHA512:
-  metadata.gz: 1a0fa064e9fc157ad837682e520d5f287546ebb63631887f0108892f0d12118ba2ec3f0cf20e7a0def18a02e8213958c87963a7cbbfb9f05548ef57d11d8fc12
-  data.tar.gz: a51c475919cb2c55c5be0b1bda1765edfb9cc20ca1b68ea75fa9eefbfcc576d141ba6c1f6bd2744da3e0cb63dbc8de23a303dfecd0587dbc8ab62e8c9ab4221f
+  metadata.gz: 5719b3f3c023d170d9ed69e7b011b11dac9f62f35d98ae58f062bd3a463c2c41d5fd53729f33e1194123f04018974aaccf4f647916f3f1d17418d92c586451b1
+  data.tar.gz: 01da1f219929ef74c0fc3d20ac658139ae6a076eb05bf12baf4bf14b2a59b097fd14ae05382d9fdbcce7c53bf6203524ff3a96f0fbafa4bb74dd1ab8d9a3be53

data/bookshark.gemspec

@@ -1,34 +1,35 @@
-# coding: utf-8
+# frozen_string_literal: true
+
 lib = File.expand_path('../lib', __FILE__)
 $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
 require 'bookshark/version'
 
 Gem::Specification.new do |spec|
-  spec.name          = "bookshark"
+  spec.name          = 'bookshark'
   spec.version       = Bookshark::VERSION
-  spec.authors       = ["Dimitris Klisiaris"]
-  spec.email         = ["dklisiaris@gmail.com"]
-  spec.summary       = %q{Book metadata extractor from biblionet.gr.}
-  spec.description   = %q{Extracts book, author, publisher and category metadata from biblionet.gr.}
-  spec.homepage      = "https://github.com/dklisiaris/bookshark"
-  spec.license       = "MIT"
+  spec.authors       = ['Dimitris Klisiaris']
+  spec.email         = ['dklisiaris@gmail.com']
+  spec.summary       = 'Book metadata extractor from biblionet.gr.'
+  spec.description   = 'Extracts book, author, publisher and category metadata from biblionet.gr.'
+  spec.homepage      = 'https://github.com/dklisiaris/bookshark'
+  spec.license       = 'MIT'
 
   spec.required_ruby_version = '>= 1.9.3'
 
   spec.files         = `git ls-files -z`.split("\x0")
   spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
   spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
-  spec.require_paths = ["lib"]
+  spec.require_paths = ['lib']
 
-  spec.add_dependency "nokogiri", "~> 1.6", ">= 1.6.6"
-  spec.add_dependency "sanitize", "~> 4.0"
-  spec.add_dependency "json", "~> 1.8"
-  spec.add_dependency "htmlentities", "~> 4.3"
-  spec.add_dependency "marc", "~> 1.0"
+  spec.add_dependency 'htmlentities', '~> 4.3'
+  spec.add_dependency 'json', '~> 2.3'
+  spec.add_dependency 'marc', '~> 1.0'
+  spec.add_dependency 'nokogiri', '~> 1.10', '>= 1.10.10'
+  spec.add_dependency 'sanitize', '~> 5.2'
 
-  spec.add_development_dependency "bundler", ">= 1.6"
-  spec.add_development_dependency "rake", "~> 10.0"
-  spec.add_development_dependency 'rspec', "~> 3.2"
-  spec.add_development_dependency "webmock", "~> 1.2"
-  spec.add_development_dependency "pry-byebug", "~> 3.4"
+  spec.add_development_dependency 'bundler', '~> 2.1'
+  spec.add_development_dependency 'pry-byebug', '~> 3.9'
+  spec.add_development_dependency 'rake', '~> 13.0'
+  spec.add_development_dependency 'rspec', '~> 3.9'
+  spec.add_development_dependency 'webmock', '~> 3.8'
 end

data/lib/bookshark.rb

@@ -1,4 +1,4 @@
-require "bookshark/version"
+require 'bookshark/version'
 require 'bookshark/storage/file_manager'
 
 require 'bookshark/extractors/author_extractor'

data/lib/bookshark/extractors/book_extractor.rb

@@ -46,41 +46,41 @@ module Biblionet
         details_hash = Hash.new
 
         details.each do |detail|
-          date_regex = /(^\d{4}$)/
+          date_regex = /(^(\d{4})|(, \d{4})$)/
           status_regex = /^\[\p{Word}+(?:\s*[\'\-\+\s]\s*\p{Word}+)*\]$/
           detail = decode_text(detail)
 
           begin
             if detail =~ date_regex
-              #puts "Publication Year: #{detail}"
+              # puts "Publication Year: #{detail}"
               details_hash[:publication_year] = detail
-            elsif detail.end_with? "σελ."
+            elsif detail.end_with? 'σελ.'
               pages = detail.gsub(/[^\d]/, '')
-              #puts "Pages: #{pages}"
+              # puts "Pages: #{pages}"
               details_hash[:pages] = pages
-            elsif detail.start_with? "ISBN-13"
-              isbn_13 = detail.gsub(/ISBN-13 /, "").gsub("&Chi","X")
-              details_hash[:isbn_13] = isbn_13
-              #puts "ISBN: #{isbn_13}"
-            elsif detail.start_with? "ISBN"
-              isbn = detail.gsub(/ISBN /, "").gsub("&Chi","X")
-              #puts "ISBN: #{isbn}"
+            elsif detail.start_with? 'ISBN-13'
+              isbn13 = detail.gsub(/ISBN-13 /, '').gsub('&Chi', 'X').gsub("\u0081]", '-')
+              details_hash[:isbn_13] = isbn13
+              # puts "ISBN: #{isbn_13}"
+            elsif detail.start_with? 'ISBN'
+              isbn = detail.gsub(/ISBN /, '').gsub('&Chi', 'X').gsub("\u0081]", '-')
+              # puts "ISBN: #{isbn}"
               details_hash[:isbn] = isbn
             elsif detail =~ status_regex
               status = detail.gsub(/\[|\]/, '')
-              #puts "Status: #{status}"
+              # puts "Status: #{status}"
               details_hash[:status] = status
-            elsif detail.start_with? "Τιμή"
+            elsif detail.start_with? 'Τιμή'
               price = detail.gsub(/[^\d,\d]/, '')
-              #puts "Price: #{price}"
+              # puts "Price: #{price}"
               details_hash[:price] = price
             elsif detail.start_with? '<img src="/images/award.jpg" border="0" title="Βραβείο">'
               award = Sanitize.clean(detail).strip
               details_hash[:awards] = [] if details_hash[:awards].nil?
               details_hash[:awards] << award
-            elsif detail.start_with? "ISMN" #Special typo case
-              isbn = detail.gsub(/ISMN /, "")
-              #puts "ISBN: #{isbn}"
+            elsif detail.start_with? 'ISMN' # Special typo case
+              isbn = detail.gsub(/ISMN /, '')
+              # puts "ISBN: #{isbn}"
               details_hash[:isbn] = isbn
             else
               raise NoIdeaWhatThisIsError.new(@biblionet_id, detail)
@@ -321,12 +321,21 @@ module Biblionet
       end
 
       def details
-        details = @nodeset.css('.book_details')[0].inner_html.gsub(/(^\d,\d)|(\D,|,\D)(?=[^\]]*(?:\[|$))/, "<br>").split("<br>").map(&:strip).reject!(&:empty?)
+        details = @nodeset.css('.book_details')[0]
+                          .inner_html
+                          .gsub(/(^\d,\d)|(\D,|,\D)(?=[^\]]*(?:\[|$))/, '<br>')
+                          .split('<br>').map(&:strip)
+                          .reject(&:empty?)
         if details.nil?
-          details = @nodeset.css('.book_details')[1].inner_html.gsub(/(^\d,\d)|(\D,|,\D)(?=[^\]]*(?:\[|$))/, "<br>").split("<br>").map(&:strip).reject!(&:empty?)
+          details = @nodeset.css('.book_details')[1]
+                            .inner_html
+                            .gsub(/(^\d,\d)|(\D,|,\D)(?=[^\]]*(?:\[|$))/, '<br>')
+                            .split('<br>')
+                            .map(&:strip)
+                            .reject(&:empty?)
         end
 
-        return details
+        details
       end
 
       def description

data/lib/bookshark/version.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Bookshark
-  VERSION = "1.0.6"
+  VERSION = '1.0.7'
 end

data/tasks/rspec.rake

@@ -1,3 +1,3 @@
 require 'rspec/core/rake_task'
 
-RSpec::Core::RakeTask.new(:spec)
+RSpec::Core::RakeTask.new(:spec)

metadata

@@ -1,161 +1,161 @@
 --- !ruby/object:Gem::Specification
 name: bookshark
 version: !ruby/object:Gem::Version
-  version: 1.0.6
+  version: 1.0.7
 platform: ruby
 authors:
 - Dimitris Klisiaris
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-05-01 00:00:00.000000000 Z
+date: 2020-08-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
-  name: nokogiri
+  name: htmlentities
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.6'
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 1.6.6
+        version: '4.3'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.6'
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: 1.6.6
+        version: '4.3'
 - !ruby/object:Gem::Dependency
-  name: sanitize
+  name: json
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.0'
+        version: '2.3'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.0'
+        version: '2.3'
 - !ruby/object:Gem::Dependency
-  name: json
+  name: marc
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.8'
+        version: '1.0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.8'
+        version: '1.0'
 - !ruby/object:Gem::Dependency
-  name: htmlentities
+  name: nokogiri
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.3'
+        version: '1.10'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.10.10
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.3'
+        version: '1.10'
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.10.10
 - !ruby/object:Gem::Dependency
-  name: marc
+  name: sanitize
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.0'
+        version: '5.2'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.0'
+        version: '5.2'
 - !ruby/object:Gem::Dependency
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.6'
+        version: '2.1'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.6'
+        version: '2.1'
 - !ruby/object:Gem::Dependency
-  name: rake
+  name: pry-byebug
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '10.0'
+        version: '3.9'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '10.0'
+        version: '3.9'
 - !ruby/object:Gem::Dependency
-  name: rspec
+  name: rake
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.2'
+        version: '13.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.2'
+        version: '13.0'
 - !ruby/object:Gem::Dependency
-  name: webmock
+  name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.2'
+        version: '3.9'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.2'
+        version: '3.9'
 - !ruby/object:Gem::Dependency
-  name: pry-byebug
+  name: webmock
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.4'
+        version: '3.8'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.4'
+        version: '3.8'
 description: Extracts book, author, publisher and category metadata from biblionet.gr.
 email:
 - dklisiaris@gmail.com
@@ -188,69 +188,6 @@ files:
 - lib/bookshark/extractors/publisher_extractor.rb
 - lib/bookshark/extractors/search.rb
 - lib/bookshark/storage/file_manager.rb
-- lib/bookshark/storage/json_book_records/0/book_63.json
-- lib/bookshark/storage/json_book_records/0/book_67.json
-- lib/bookshark/storage/json_book_records/2/book_2110.json
-- lib/bookshark/storage/json_book_records/2/book_2111.json
-- lib/bookshark/storage/json_book_records/2/book_2112.json
-- lib/bookshark/storage/json_book_records/2/book_2113.json
-- lib/bookshark/storage/json_book_records/2/book_2114.json
-- lib/bookshark/storage/json_book_records/2/book_2115.json
-- lib/bookshark/storage/json_book_records/2/book_2116.json
-- lib/bookshark/storage/json_book_records/2/book_2117.json
-- lib/bookshark/storage/json_book_records/2/book_2118.json
-- lib/bookshark/storage/json_book_records/2/book_2119.json
-- lib/bookshark/storage/json_book_records/2/book_2120.json
-- lib/bookshark/storage/json_book_records/2/book_2121.json
-- lib/bookshark/storage/json_book_records/2/book_2122.json
-- lib/bookshark/storage/json_book_records/2/book_2123.json
-- lib/bookshark/storage/json_book_records/2/book_2124.json
-- lib/bookshark/storage/json_book_records/2/book_2125.json
-- lib/bookshark/storage/json_book_records/2/book_2126.json
-- lib/bookshark/storage/json_book_records/2/book_2127.json
-- lib/bookshark/storage/json_book_records/2/book_2128.json
-- lib/bookshark/storage/json_book_records/2/book_2129.json
-- lib/bookshark/storage/json_book_records/2/book_2130.json
-- lib/bookshark/storage/json_book_records/2/book_2131.json
-- lib/bookshark/storage/json_book_records/2/book_2132.json
-- lib/bookshark/storage/json_book_records/2/book_2133.json
-- lib/bookshark/storage/json_book_records/2/book_2134.json
-- lib/bookshark/storage/json_book_records/2/book_2135.json
-- lib/bookshark/storage/json_book_records/2/book_2136.json
-- lib/bookshark/storage/json_book_records/2/book_2137.json
-- lib/bookshark/storage/json_book_records/2/book_2138.json
-- lib/bookshark/storage/json_book_records/2/book_2139.json
-- lib/bookshark/storage/json_book_records/2/book_2140.json
-- lib/bookshark/storage/json_book_records/2/book_2141.json
-- lib/bookshark/storage/json_book_records/2/book_2142.json
-- lib/bookshark/storage/json_book_records/2/book_2143.json
-- lib/bookshark/storage/json_book_records/2/book_2144.json
-- lib/bookshark/storage/json_book_records/2/book_2145.json
-- lib/bookshark/storage/json_book_records/2/book_2146.json
-- lib/bookshark/storage/json_book_records/2/book_2147.json
-- lib/bookshark/storage/json_book_records/2/book_2148.json
-- lib/bookshark/storage/json_book_records/2/book_2149.json
-- lib/bookshark/storage/json_book_records/2/book_2150.json
-- lib/bookshark/storage/json_book_records/2/book_2151.json
-- lib/bookshark/storage/json_book_records/2/book_2152.json
-- lib/bookshark/storage/json_book_records/2/book_2153.json
-- lib/bookshark/storage/json_book_records/2/book_2154.json
-- lib/bookshark/storage/json_book_records/2/book_2155.json
-- lib/bookshark/storage/json_book_records/2/book_2156.json
-- lib/bookshark/storage/json_book_records/2/book_2157.json
-- lib/bookshark/storage/json_book_records/2/book_2158.json
-- lib/bookshark/storage/json_book_records/2/book_2159.json
-- lib/bookshark/storage/json_book_records/2/book_2160.json
-- lib/bookshark/storage/json_book_records/2/book_2161.json
-- lib/bookshark/storage/json_book_records/2/book_2162.json
-- lib/bookshark/storage/json_book_records/2/book_2163.json
-- lib/bookshark/storage/json_book_records/2/book_2164.json
-- lib/bookshark/storage/json_book_records/2/book_2165.json
-- lib/bookshark/storage/json_book_records/2/book_2166.json
-- lib/bookshark/storage/json_book_records/2/book_2167.json
-- lib/bookshark/storage/json_book_records/2/book_2168.json
-- lib/bookshark/storage/json_book_records/2/book_2169.json
-- lib/bookshark/storage/json_book_records/2/book_2170.json
 - lib/bookshark/version.rb
 - spec/bookshark_spec.rb
 - spec/spec_helper.rb
@@ -289,8 +226,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.5.2
+rubygems_version: 3.0.3
 signing_key: 
 specification_version: 4
 summary: Book metadata extractor from biblionet.gr.
@@ -311,4 +247,3 @@ test_files:
 - spec/test_data/publisher_20.json
 - spec/test_data/search_01.json
 - spec/test_data/search_ids_01.json
-has_rdoc: