dwc_agent 1.4.12 → 1.5.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f0eca9af87fc999b9d1e2313ea88df6bee1aec54a080e46fc8245abd3c41d615
-  data.tar.gz: 38f98d4fc7b30a2f80521d7d064666f5da9f64889af7f08b15a852619d1ab3be
+  metadata.gz: 87cfd6b6ab9ee83156f9503e48669691d486f143f64a623f93a40f064d2dc7f7
+  data.tar.gz: 7f40287aa6bf7e7b90408d72594fd9ab5498401c5ee402da6f2a78163f011a9a
 SHA512:
-  metadata.gz: 43dafd9ab3d3cf8513a579614d56429551803f15b39a814f25076d09d023f335c74fb8d22aaed49bbc2318587963972d74f035951221d5b12057fc7a0691ae57
-  data.tar.gz: 35d99c98fd8d7334e5cdac7bf781f0ad85adb6fd0684ffa6c9c058264ade10d2a6b0b544ceec4c52ca64192833c3edcb896ed29e8f45c4df970680b6db719eaa
+  metadata.gz: 4cb91ffc7530bd1a93fbbc70c716ad96d49cab139bf95507e3a688566ce997cb71510b13013bb07cb52dffe2ce7419d294392028eedd28f1caba8b032eb98c9a
+  data.tar.gz: ae226c63106a2b41899b6d4c85c92aa372d8d1cbce8a90023cf9b8d1a4c153a28bd33a9eb8064c492fa79d5d4229ec8979d5ca4f751d97fc1ec09132b0e46ca3

data/lib/dwc_agent/cleaner.rb

@@ -16,7 +16,7 @@ module DwcAgent
     # @param parsed_namae [Object] the namae object
     # @return [Hash] the given, family hash
     def clean(parsed_namae)
-      blank_name = { given: nil, family: nil, particle: nil }
+      blank_name = { title: nil, appellation: nil, given: nil, particle: nil, family: nil, suffix: nil }
 
       if parsed_namae.family && FAMILY_BLACKLIST.any?{ |s| s.casecmp(parsed_namae.family) == 0 }
         return blank_name
@@ -84,6 +84,9 @@ module DwcAgent
       family = parsed_namae.family.gsub(/\.\z/, '').strip rescue nil
       given = parsed_namae.given.strip rescue nil
       particle = parsed_namae.particle.strip rescue nil
+      appellation = parsed_namae.appellation.strip rescue nil
+      suffix = parsed_names.suffix.strip rescue nil
+      title = parsed_names.title.strip rescue nil
 
       if !given.nil? && given.match(/[A-Z]\.[A-Za-z]{2,}/)
         given = given.gsub(".", ". ").strip
@@ -123,7 +126,7 @@ module DwcAgent
         return blank_name
       end
 
-      { given: given, family: family, particle: particle }
+      { title: nil, appellation: nil, given: given, particle: particle, family: family, suffix: nil }
     end
 
   end

data/lib/dwc_agent/constants.rb

@@ -159,16 +159,12 @@ module DwcAgent
   }
 
   PHRASE_SUBS = {
-    'dr\.' => 'Dr. ',
-    'mr\.' => 'Mr. ',
-    'mrs\.' => 'Mrs. ',
-    'ms\.' => 'Ms. ',
     'prof\.' => 'Prof. ',
     '\, ph\.d\.' => ' Ph.D.',
     '\, bro\.' => ' Bro.',
-    ' jr\.,' => ' Jr.;',
-    ' jr,' => ' Jr.;',
-    '-Jr' => ' Jr.'
+    ' jr\.\,' => ' Jr.;',
+    ' jr\,' => ' Jr.;',
+    '\-jr' => ' Jr.'
   }
 
   COMPLEX_SEPARATORS = %r{
@@ -274,4 +270,8 @@ module DwcAgent
 
   TITLE = /\s*\b(sir|count(ess)?|colonel|(gen|adm|col|maj|capt|cmdr|lt|sgt|cpl|pvt|prof|dr|md|ph\.?d|rev|mme|abbé|ptre|bro|esq)\.?|docteur|father|cantor|vicar|père|pastor|rabbi|reverend|pere|soeur|sister|professor)(\s+|$)/i
 
+  APPELLATION = /\s*\b((mrs?|ms|fr|hr)\.?|miss|herr|frau)(\s+|$)/i
+
+  SUFFIX = /\s*\b(JR|Jr|jr|SR|Sr|sr|[IVX]{2,})(\.|\b)/
+
 end

data/lib/dwc_agent/parser.rb

@@ -11,14 +11,16 @@ module DwcAgent
       options = {
         prefer_comma_as_separator: true,
         separator: SPLIT_BY,
-        title: TITLE
+        title: TITLE,
+        appellation: APPELLATION,
+        suffix: SUFFIX
       }
       @namae = Namae::Parser.new(options)
 
       @strip_out_regex = Regexp.new STRIP_OUT.to_s
       @residual_terminators_regex = Regexp.new SPLIT_BY.to_s + %r{\s*\z}.to_s
       @char_subs_regex = Regexp.new [CHAR_SUBS.keys.join].to_s
-      @phrase_subs_regex = Regexp.new((PHRASE_SUBS.keys.join('|')).to_s, Regexp::IGNORECASE)
+      @phrase_subs_regex = Regexp.new PHRASE_SUBS.keys.join('|').to_s, Regexp::IGNORECASE
       @complex_separators_regex = Regexp.new COMPLEX_SEPARATORS.to_s
       @add_separators_regex = Regexp.new %r{(\S{1}\.)([[:alpha:]]{2,})}.to_s
     end

data/lib/dwc_agent/version.rb

@@ -2,8 +2,8 @@ module DwcAgent
   class Version
 
     MAJOR = 1
-    MINOR = 4
-    PATCH = 12
+    MINOR = 5
+    PATCH = 0
     BUILD = nil
 
     def self.version

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: dwc_agent
 version: !ruby/object:Gem::Version
-  version: 1.4.12
+  version: 1.5.0
 platform: ruby
 authors:
 - David P. Shorthouse
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-08-05 00:00:00.000000000 Z
+date: 2020-08-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: namae