isobib 0.1.6 → 0.1.7

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9070ef52f2495bc73b0b261792057e84857a9bd593a8c2a758fec1a019e92669
-  data.tar.gz: 4dd30b01b9c5669610062e92903eb1cd663fab2ef667a9d9b716cf8e112456b8
+  metadata.gz: a1b8d223cafdb296a3332ca76dabcaeeffbe8a4d5658b4cd821e2b6cbee6a8aa
+  data.tar.gz: 443f128cee77c73bd940d55c68050f47863199419e90425b44c0fe5fc16034bb
 SHA512:
-  metadata.gz: b844d479c9114a7a63811d1e99cf32035407ccfe2bfda809a07f967878b9ee3a291a98c258de3b27633265b3e2014c4381806a35713705ccf889ee2aeca395cd
-  data.tar.gz: 711101caaed30badac5152e7ed42de96c0f21933a4133c8155f1823a0ea60f98d58949e3d05298799ef0ee363bffcb98f3a290b30a5b7b7a54a9a147d4f4299c
+  metadata.gz: b47023fc174c5ad11a8499aa98b17b9eaa60d5ba163afd110f86555625461239972d8b711fc981d2b6953756c4a0b2ff96ed3b95d5fe3d732f55f943ef0165f3
+  data.tar.gz: 5c0cc7ec86313db6274793c14ea5493e2d16b44f9a6b13af14eca204b0f6e45706d1e8cb9874a2bc99470197abb1a28231579832f3c3b3fbef5ed207bbfe9417

data/Gemfile

@@ -1,3 +1,6 @@
+Encoding.default_external = Encoding::UTF_8
+Encoding.default_internal = Encoding::UTF_8
+
 source "https://rubygems.org"
 
 git_source(:github) {|repo_name| "https://github.com/#{repo_name}" }

data/Gemfile.lock

@@ -1,29 +1,31 @@
 PATH
   remote: .
   specs:
-    isobib (0.1.5)
+    isobib (0.1.7)
       algoliasearch
       iso-bib-item (~> 0.1.2)
 
 GEM
   remote: https://rubygems.org/
   specs:
-    algoliasearch (1.22.0)
+    algoliasearch (1.23.2)
       httpclient (~> 2.8, >= 2.8.3)
       json (>= 1.5.1)
     byebug (10.0.2)
     coderay (1.1.2)
     diff-lcs (1.3)
     docile (1.3.1)
+    duplicate (1.1.1)
     httpclient (2.8.3)
-    iso-bib-item (0.1.5)
+    iso-bib-item (0.1.8)
+      duplicate
       isoics (~> 0.1.6)
       nokogiri
     isoics (0.1.6)
     json (2.1.0)
     method_source (0.9.0)
     mini_portile2 (2.3.0)
-    nokogiri (1.8.2)
+    nokogiri (1.8.3)
       mini_portile2 (~> 2.3.0)
     pry (0.11.3)
       coderay (~> 1.1.0)
@@ -56,6 +58,7 @@ PLATFORMS
 
 DEPENDENCIES
   bundler (~> 1.16)
+  byebug
   isobib!
   pry-byebug
   rake (~> 10.0)

data/README.adoc

@@ -1,5 +1,9 @@
 = IsoBib: retrieve ISO Standards for bibliographic use using the BibliographicItem model
 
+image:https://img.shields.io/gem/v/isobib.svg["Gem Version", link="https://rubygems.org/gems/isobib"]
+image:https://img.shields.io/travis/riboseinc/isobib/master.svg["Build Status", link="https://travis-ci.org/riboseinc/isobib"]
+image:https://codeclimate.com/github/riboseinc/isobib/badges/gpa.svg["Code Climate", link="https://codeclimate.com/github/riboseinc/isobib"]
+
 IsoBib is a Ruby gem that implements the https://github.com/riboseinc/isodoc-models#iso-bibliographic-item[IsoBibliographicItem model].
 
 You can use it to retrieve metadata of ISO Standards from https://www.iso.org, and access such metadata through the `IsoBibliographicItem` object.

data/isobib.gemspec

@@ -27,6 +27,7 @@ Gem::Specification.new do |spec|
 
   spec.add_development_dependency 'bundler', '~> 1.16'
   spec.add_development_dependency 'pry-byebug'
+  spec.add_development_dependency 'byebug'
   spec.add_development_dependency 'rake', '~> 10.0'
   spec.add_development_dependency 'rspec', '~> 3.0'
   spec.add_development_dependency 'simplecov'

data/lib/isobib/iso_bibliography.rb

@@ -19,6 +19,130 @@ module Isobib
       def search_and_fetch(text)
         Scrapper.get(text)
       end
+
+      # @param code [String] the ISO standard Code to look up (e..g "ISO 9000")
+      # @param year [String] the year the standard was published (optional)
+      # @param opts [Hash] options; restricted to :all_parts if all-parts reference is required
+      # @return [String] Relaton XML serialisation of reference
+      def isobib_get(code, year, opts)
+        return iev.to_xml if code.casecmp? "IEV"
+        code += "-1" if opts[:all_parts]
+        ret = isobib_get1(code, year, opts)
+        return nil if ret.nil?
+        ret.to_most_recent_reference if !year
+        ret.to_all_parts if opts[:all_parts]
+        ret.to_xml
+      end
+
+      private
+
+      def fetch_ref_err(code, year, missed_years)
+        id = year ? "#{code}:#{year}" : code
+        warn "WARNING: no match found on the ISO website for #{id}. "\
+          "The code must be exactly like it is on the website."
+        warn "(There was no match for #{year}, though there were matches "\
+          "found for #{missed_years.join(', ')}.)" unless missed_years.empty?
+        if /\d-\d/.match? code
+          warn "The provided document part may not exist, or the document "\
+            "may no longer be published in parts."
+        else
+          warn "If you wanted to cite all document parts for the reference, "\
+            "use \"#{code} (all parts)\".\nIf the document is not a standard, "\
+            "use its document type abbreviation (TS, TR, PAS, Guide)."
+        end
+        nil
+      end
+
+      def fetch_pages(s, n)
+        workers = WorkersPool.new n
+        workers.worker { |w| { i: w[:i], hit: w[:hit].fetch } }
+        s.each_with_index { |hit, i| workers << { i: i, hit: hit } }
+        workers.end
+        workers.result.sort { |x, y| x[:i] <=> y[:i] }.map { |x| x[:hit] }
+      end
+
+      def isobib_search_filter(code)
+        docidrx = %r{^(ISO|IEC)[^0-9]*\s[0-9-]+}
+        corrigrx = %r{^(ISO|IEC)[^0-9]*\s[0-9-]+:[0-9]+/}
+        warn "fetching #{code}..."
+        result = search(code)
+        result.each do |page|
+          ret = page.select do |i|
+            i.hit["title"] &&
+              i.hit["title"].match(docidrx).to_s == code &&
+              !corrigrx.match?(i.hit["title"])
+          end
+          return ret unless ret.empty?
+        end
+        []
+      end
+
+
+      def iev
+        Nokogiri::XML.fragment(<<~"END")
+          <bibitem type="international-standard" id="IEV">
+  <title format="text/plain" language="en" script="Latn">Electropedia: 
+  The World's Online Electrotechnical Vocabulary</title>
+  <source type="src">http://www.electropedia.org</source>
+  <docidentifier>IEV</docidentifier>
+  <date type="published"> <on>#{Date.today.year}</on> </date>
+  <contributor>
+    <role type="publisher"/>
+    <organization>
+      <name>International Electrotechnical Commission</name>
+      <abbreviation>IEC</abbreviation>
+      <uri>www.iec.ch</uri>
+    </organization>
+  </contributor>
+  <language>en</language> <language>fr</language>
+  <script>Latn</script>
+  <copyright>
+    <from>#{Date.today.year}</from>
+    <owner>
+      <organization>
+      <name>International Electrotechnical Commission</name>
+      <abbreviation>IEC</abbreviation>
+      <uri>www.iec.ch</uri>
+      </organization>
+    </owner>
+  </copyright>
+  <relation type="updates">
+    <bibitem>
+      <formattedref>IEC 60050</formattedref>
+    </bibitem>
+  </relation>
+</bibitem>
+        END
+      end
+
+
+      # Sort through the results from Isobib, fetching them three at a time,
+      # and return the first result that matches the code,
+      # matches the year (if provided), and which # has a title (amendments do not).
+      # Only expects the first page of results to be populated.
+      # Does not match corrigenda etc (e.g. ISO 3166-1:2006/Cor 1:2007)
+      # If no match, returns any years which caused mismatch, for error reporting
+      def isobib_results_filter(result, year)
+        missed_years = []
+        result.each_slice(3) do |s| # ISO website only allows 3 connections
+          fetch_pages(s, 3).each_with_index do |r, i|
+            return { ret: r } if !year
+            r.dates.select { |d| d.type == "published" }.each do |d|
+              return { ret: r } if year.to_i == d.on.year
+              missed_years << d.on.year
+            end
+          end
+        end
+        { years: missed_years }
+      end
+
+      def isobib_get1(code, year, opts)
+        return iev if code.casecmp? "IEV"
+        result = isobib_search_filter(code) or return nil
+        ret = isobib_results_filter(result, year)
+        return ret[:ret] if ret[:ret]
+        fetch_ref_err(code, year, ret[:years])
+      end
     end
   end
 end

data/lib/isobib/scrapper.rb

@@ -220,7 +220,7 @@ module Isobib
           technical_committee: {
             name:   wg_link.text + doc.css('div.entry-title')[0].text,
             type:   'technicalCommittee',
-            number: workgroup[1].match(/\d+/).to_s.to_i
+            number: workgroup[1]&.match(/\d+/)&.to_s&.to_i
           } }
       end
 

data/lib/isobib/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Isobib
-  VERSION = '0.1.6'
+  VERSION = '0.1.7'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: isobib
 version: !ruby/object:Gem::Version
-  version: 0.1.6
+  version: 0.1.7
 platform: ruby
 authors:
 - Ribose Inc.
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-06-17 00:00:00.000000000 Z
+date: 2018-06-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -38,6 +38,20 @@ dependencies:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
+- !ruby/object:Gem::Dependency
+  name: byebug
+  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'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement