petrarca 0.3.0 → 0.4.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e588fc804e558ead549eeb33f9eeecf157c2d0652927a8ec8057f0057400f932
-  data.tar.gz: e25faf3cf900c1ced7f34884e347c21b3094db97e3ad49640ed8a721c07dc376
+  metadata.gz: cf5f21fb7dd10bb65123ae526db409663c9ef3c2967268fb4f4839996aa97cc9
+  data.tar.gz: 9e0abbe048c5b96d02bfa407981738cf3e9fe74704e8713263120725a97b86d3
 SHA512:
-  metadata.gz: 4c0add5276c9ab9f1c1babae1d7838771eaf693eda6acf6840fb5c94e78fc85ac70b2e659db2c37e1afa8757f4cba205120f7687385e3406c8b44127dfbd8ce7
-  data.tar.gz: 0efc1a6cdda198cec9799242a976fc5403d9374ec54237027133d181507e79c933606fcba37d11214d712c7fc79748fd664d54bc385f5a44a0868cbc2fa7d136
+  metadata.gz: 1df099128e166619d892ed2700267f3255bb555ef0aaab62ae652d9f6e42258b79b80d9d81b9e8e6b1a8c8b09a4c6b83e6464e20022e231bbbf4b0e19f369f50
+  data.tar.gz: d3e276b38b978639f91be5bbd766279ca305e7d4fc822a96ae1559046c476c4efcee9636a8ce0038048f604bb62b05c5e65794f2c67262feb414140f9a227875

data/.gitignore

@@ -9,3 +9,4 @@
 /vendor/
 /Gemfile.lock
 *.txt
+*.xml

data/README.md

@@ -9,7 +9,8 @@ This library Petrarca provides some utility functions to manipulate ISBN numbers
 
 All functions support both ISBN-13 and ISBN-10.
 
-Note: Only registrant ranges for Japan are currently supported.
+All ranges of registration groups and registrants are supported.
+Those depends on 'RangeMessage.xml' file, downloaded from [International ISBN Agency](https://www.isbn-international.org/range_file_generation).
 
 ## Installation
 

data/Rakefile

@@ -1,2 +1,9 @@
 require "bundler/gem_tasks"
 task :default => :spec
+
+namespace :range do
+  desc "Generate registrant range files"
+  task :generate do
+    ruby "scripts/generate_range_files.rb --dir data"
+  end
+end

data/lib/petrarca/helpers.rb

@@ -22,7 +22,7 @@ module Petrarca
         f.each_line do |line|
           next if line.start_with?("#")
           g, r = line.chomp.split(":")
-          ranges[g] = r.split(",")
+          ranges[g] = r.split(",") unless r.nil?
         end
       end
       ranges

data/lib/petrarca/version.rb

@@ -1,3 +1,3 @@
 module Petrarca
-  VERSION = "0.3.0"
+  VERSION = "0.4.0"
 end

data/petrarca.gemspec

@@ -40,4 +40,5 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "bundler", "~> 2.0"
   spec.add_development_dependency "rake", "~> 10.0"
   spec.add_development_dependency "rspec", ">= 3.0.0"
+  spec.add_development_dependency "nokogiri"
 end

data/scripts/generate_range_files.rb

@@ -0,0 +1,82 @@
+# encoding: utf-8
+
+require "nokogiri"
+require "optparse"
+
+
+RANGE_MESSAGE_FILE = "RangeMessage.xml"
+REGISTRATION_GROUP_RANGES_FILE = "registration_group_ranges.txt"
+REGISTRANT_RANGES_FILE = "registrant_ranges.txt"
+
+
+def main
+  options = {
+    dir: "."
+  }
+  opts = OptionParser.new
+  opts.on("-d", "--dir=DIR", String, "Set directory, output range fiels into."){|v|
+    options[:dir] = v
+  }
+  opts.parse!
+
+  doc = File.open(RANGE_MESSAGE_FILE, "r"){|f| Nokogiri::XML(f) }
+
+  metadata = {
+    source: doc.xpath("//MessageSource").text,
+    date: doc.xpath("//MessageDate").text
+  }
+
+  registration_groups = extract_ranges(doc.xpath("//EAN.UCC"))
+  registrants = extract_ranges(doc.xpath("//Group"))
+
+  output_range_file(
+    registration_groups,
+    File.join(options[:dir], REGISTRATION_GROUP_RANGES_FILE),
+    metadata
+  )
+  output_range_file(
+    registrants,
+    File.join(options[:dir], REGISTRANT_RANGES_FILE),
+    metadata
+  )
+end
+
+def extract_ranges(nodes)
+  nodes.map do |node|
+    prefix = node.xpath("Prefix").text
+    agency = node.xpath("Agency").text
+    ranges = node.xpath("Rules/Rule").map do |rule|
+      length = rule.xpath("Length").text.to_i
+      range = if length > 0
+        rule.xpath("Range").text.split("-").map{|s| s[0, length] }.join("-")
+      else
+        nil
+      end
+      range
+    end.compact
+    {
+      "prefix" => prefix,
+      "agency" => agency,
+      "ranges" => ranges
+    }
+  end
+end
+
+def output_range_file(ranges, range_file, metadata)
+  File.open(range_file, "w") do |f|
+    f.puts "# Generated from RangeMessage.xml file, downloaded from https://www.isbn-international.org/range_file_generation."
+    f.puts "#"
+    f.puts "# " + metadata[:source]
+    f.puts "# " + metadata[:date]
+    f.puts "#"
+    ranges.each do |range|
+      f.puts "# " + range["agency"]
+      f.puts range["prefix"] + ":" + range["ranges"].join(",")
+    end
+  end
+  $stderr.puts "Generated: #{range_file}"
+end
+
+
+
+main

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: petrarca
 version: !ruby/object:Gem::Version
-  version: 0.3.0
+  version: 0.4.0
 platform: ruby
 authors:
 - takatoh
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2020-09-04 00:00:00.000000000 Z
+date: 2020-09-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -52,6 +52,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: 3.0.0
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: Library to manipulate ISBN numbers for Ruby.
 email:
 - takatoh.m@gmail.com
@@ -75,6 +89,7 @@ files:
 - lib/petrarca/isbn13.rb
 - lib/petrarca/version.rb
 - petrarca.gemspec
+- scripts/generate_range_files.rb
 homepage: https://github.com/takatoh/Petrarca
 licenses:
 - MIT