Class: DwcAgent::Cleaner

Inherits:
Object
  • Object
show all
Defined in:
lib/dwc_agent/cleaner.rb

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initializeCleaner

Returns a new instance of Cleaner.



11
12
13
14
15
16
# File 'lib/dwc_agent/cleaner.rb', line 11

def initialize
  @blacklist = BLACKLIST
  @given_blacklist = GIVEN_BLACKLIST
  @family_blacklist = FAMILY_BLACKLIST
  @particles = PARTICLES
end

Class Method Details

.instanceObject



6
7
8
# File 'lib/dwc_agent/cleaner.rb', line 6

def instance
  Thread.current[:dwc_agent_cleaner] ||= new
end

Instance Method Details

#clean(parsed_namae) ⇒ Object

Cleans the passed-in namae object from the parse method and re-organizes it to better match expected Darwin Core output.

Parameters:

  • parsed_namae (Namae::Name)

    a Namae object



27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
# File 'lib/dwc_agent/cleaner.rb', line 27

def clean(parsed_namae)

  if parsed_namae.family &&
     parsed_namae.family == NameCase(parsed_namae.family) &&
     parsed_namae.display_order.split.join == parsed_namae.initials
    return default
  end

  if parsed_namae.given &&
     @given_blacklist.any?{ |s| s.casecmp(parsed_namae.given) == 0 }
    return
  end

  if parsed_namae.family &&
     parsed_namae.family.length == 3 &&
     parsed_namae.family.count('.') == 1
    return default
  end

  if parsed_namae.given && parsed_namae.given.length > 35
    return default
  end

  if parsed_namae.given &&
     parsed_namae.given.count('.') >= 3 &&
     /\.\s*[a-zA-Z]{4,}\s+[a-zA-Z]{1,}\./.match(parsed_namae.given)
    return default
  end

  if parsed_namae.display_order =~ @blacklist
    return default
  end

  if parsed_namae.family &&
     parsed_namae.family.count(".") == 1 &&
     parsed_namae.family[-1] == "." &&
     parsed_namae.family.length > 3
    parsed_namae.family = parsed_namae.family.delete_suffix(".")
  end

  if parsed_namae.given &&
     parsed_namae.family &&
     parsed_namae.family.count(".") > 0 &&
     parsed_namae.family.length - parsed_namae.family.count(".") <= 3
      given = parsed_namae.given
      family = parsed_namae.family
      parsed_namae.family = given
      parsed_namae.given = family
  end

  if parsed_namae.given &&
     parsed_namae.family &&
     parsed_namae.family.length <=3 &&
     parsed_namae.family == parsed_namae.family.upcase &&
     #parsed_namae.family != NameCase(parsed_namae.family) &&
     parsed_namae.given[-1] != "."
      given = parsed_namae.given
      family = parsed_namae.family
      parsed_namae.family = given
      parsed_namae.given = family
  end

  if !parsed_namae.given &&
     parsed_namae.particle &&
     parsed_namae.family &&
     /^[A-Za-z]{3,}\s+(?:[A-Z]\.\s?){1,}$/.match(parsed_namae.family)
     matched = /^(?<family>[A-Za-z]{3,})\s+(?<given>([A-Z]\.\s?){1,})$/.match(parsed_namae.family)
     parsed_namae.family = matched[:family]
     parsed_namae.given = matched[:given]
   end

  if parsed_namae.given &&
    (parsed_namae.given == parsed_namae.given.upcase ||
    parsed_namae.given == parsed_namae.given.downcase) &&
    !parsed_namae.given.include?(".") &&
    parsed_namae.given.tr(".","").length >= 4
      parsed_namae.given = NameCase(parsed_namae.given)
  end

  if parsed_namae.given && /\.[A-Z]$/.match(parsed_namae.given)
    parsed_namae.given += "."
  end

  if parsed_namae.given && /[A-Za-z]\./.match(parsed_namae.given)
    parsed_namae.given = NameCase(parsed_namae.given)
  end

  if parsed_namae.family &&
     @family_blacklist.any?{ |s| s.casecmp(parsed_namae.family) == 0 }
    return default
  end

  if parsed_namae.family.nil? &&
     !parsed_namae.given.nil? &&
     !parsed_namae.given.include?(".")
    parsed_namae.family = parsed_namae.given
    parsed_namae.given = nil
  end

  parsed_namae.normalize_initials

  family = parsed_namae.family
  given = parsed_namae.given.strip rescue nil
  particle = parsed_namae.particle.strip rescue nil
  appellation = parsed_namae.appellation.strip rescue nil
  suffix = parsed_namae.suffix.strip rescue nil
  title = parsed_namae.title.strip rescue nil

  if !given.nil? && given.match(/[A-Z]\.[A-Za-z]{2,}/)
    given = given.gsub(".", ". ").strip
  end

  if !family.nil? &&
     given.nil? &&
     !particle.nil? &&
     !@particles.include?(particle.downcase)
    given = particle.sub(/[a-z]\./, &:upcase).sub(/^(.)/) { $1.capitalize }
    particle = nil
  end

  if !particle.nil? && particle.include?(".")
    particle = nil
  end

  if !family.nil? && (family == family.upcase || family == family.downcase)
    family = NameCase(family)
  end

  if !family.nil? && family.match(/[A-Z]{1,3}$/)
    family = NameCase(family.upcase)
  end

  if given.nil? && !family.nil? && family.match(/^[A-Z]{2}/)
    return default
  end

  if !family.nil? && @family_blacklist.any?{ |s| s.casecmp(family) == 0 }
    return default
  end

  if !given.nil? && @given_blacklist.any?{ |s| s.casecmp(given) == 0 }
    return default
  end

  name = {
    title: title,
    appellation: appellation,
    given: given,
    particle: particle,
    family: family,
    suffix: suffix,
    nick: parsed_namae.nick,
    dropping_particle: parsed_namae.dropping_particle
  }
  Namae::Name.new(name)
end

#defaultObject



18
19
20
# File 'lib/dwc_agent/cleaner.rb', line 18

def default
  Namae::Name.new
end