Class: Geocoder::US::Address

Inherits:
Object
  • Object
show all
Defined in:
lib/geocoder/us/address.rb

Overview

The Address class takes a US street address or place name and constructs a list of possible structured parses of the address string.

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(text) ⇒ Address

Takes an address or place name string as its sole argument.

Raises:

  • (ArgumentError) 


28
29
30
31
32
33
34
35
36
37
 
# File 'lib/geocoder/us/address.rb', line 28

def initialize (text)
  raise ArgumentError, "no text provided" unless text and !text.empty?
  if text.class == Hash
    @text = ""
    assign_text_to_address text
  else
    @text = clean text
    parse
  end
end

Instance Attribute Details

#city ⇒ Object

Returns the value of attribute city.



23
24
25
 
# File 'lib/geocoder/us/address.rb', line 23

def city
  @city
end

#number ⇒ Object

Returns the value of attribute number.



21
22
23
 
# File 'lib/geocoder/us/address.rb', line 21

def number
  @number
end

#plus4 ⇒ Object

Returns the value of attribute plus4.



25
26
27
 
# File 'lib/geocoder/us/address.rb', line 25

def plus4
  @plus4
end

#prenum ⇒ Object

Returns the value of attribute prenum.



21
22
23
 
# File 'lib/geocoder/us/address.rb', line 21

def prenum
  @prenum
end

#state ⇒ Object

Returns the value of attribute state.



24
25
26
 
# File 'lib/geocoder/us/address.rb', line 24

def state
  @state
end

#street ⇒ Object

Returns the value of attribute street.



22
23
24
 
# File 'lib/geocoder/us/address.rb', line 22

def street
  @street
end

#sufnum ⇒ Object

Returns the value of attribute sufnum.



21
22
23
 
# File 'lib/geocoder/us/address.rb', line 21

def sufnum
  @sufnum
end

#text ⇒ Object

Returns the value of attribute text.



20
21
22
 
# File 'lib/geocoder/us/address.rb', line 20

def text
  @text
end

#zip ⇒ Object

Returns the value of attribute zip.



25
26
27
 
# File 'lib/geocoder/us/address.rb', line 25

def zip
  @zip
end

Instance Method Details

#assign_text_to_address(text) ⇒ Object 



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
 
# File 'lib/geocoder/us/address.rb', line 47

def assign_text_to_address(text)
  if !text[:address].nil?
    @text = clean text[:address]
    parse
  else
    @street = []
    @prenum = text[:prenum] 
    @sufnum = text[:sufnum] 
    if !text[:street].nil?
      @street = text[:street].scan(Match[:street])
    end
    @number = ""
    if !@street.nil?
      if text[:number].nil?
         @street.map! { |single_street|
           single_street.downcase!
           @number = single_street.scan(Match[:number])[0].to_s
           single_street.sub! @number, ""
           single_street.sub! /^\s*,?\s*/o, ""
          }
     else
        @number = text[:number].to_s 
      end
     @street = expand_streets(@street)
      street_parts
    end
    @city = []
    if !text[:city].nil?
      @city.push(text[:city])
      @text = text[:city].to_s
    else
      @city.push("")
    end
    if !text[:region].nil?
     # @state = []
     @state = text[:region]
      if @state.length > 2
       # full_state = @state.strip # special case: New York
        @state = State[@state]
      end
    elsif !text[:country].nil?
      @state = text[:country]
    elsif !text[:state].nil?
      @state = text[:state]
    end

    @zip = text[:postal_code] 
    @plus4 = text[:plus4] 
    if !@zip
       @zip = @plus4 = ""
    end
  end
end

#city_parts ⇒ Object 



260
261
262
263
264
265
266
267
268
269
270
271
272
273
 
# File 'lib/geocoder/us/address.rb', line 260

def city_parts
  strings = []
  @city.map {|string|
    tokens = string.split(" ")
    strings |= (0...tokens.length).to_a.reverse.map {|i|
               (i...tokens.length).map {|j| tokens[i..j].join(" ")}}.flatten
  }
  # Don't return strings that consist solely of abbreviations.
  # NOTE: Is this a micro-optimization that has edge cases that will break?
  # Answer: Yes, it breaks on "Prairie"
  good_strings = strings.reject {|s| Std_Abbr.key? s}
  strings = good_strings if !good_strings.empty?
  strings.uniq
end

#clean(value) ⇒ Object

Removes any characters that aren’t strictly part of an address string.



40
41
42
43
44
 
# File 'lib/geocoder/us/address.rb', line 40

def clean (value)
  value.strip \
       .gsub(/[^a-z0-9 ,'&@\/-]+/io, "") \
       .gsub(/\s+/o, " ")
end

#expand_numbers(string) ⇒ Object

Expands a token into a list of possible strings based on the Geocoder::US::Name_Abbr constant, and expands numerals and number words into their possible equivalents.



104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
 
# File 'lib/geocoder/us/address.rb', line 104

def expand_numbers (string)
  if /\b\d+(?:st|nd|rd|th)?\b/o.match string
    match = $&
    num = $&.to_i
  elsif Ordinals.regexp.match string
    num = Ordinals[$&]
    match = $&
  elsif Cardinals.regexp.match string
    num = Cardinals[$&]
    match = $&
  end
  strings = []
  if num and num < 100
    [num.to_s, Ordinals[num], Cardinals[num]].each {|replace|
      strings << string.sub(match, replace)
    }
  else
    strings << string
  end
  strings
end

#expand_streets(street) ⇒ Object 



207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
 
# File 'lib/geocoder/us/address.rb', line 207

def expand_streets(street)
  if !street.empty? && !street[0].nil?
    street.map! {|s|s.strip}
    add = street.map {|item| item.gsub(Name_Abbr.regexp) {|m| Name_Abbr[m]}}
    street |= add
    add = street.map {|item| item.gsub(Std_Abbr.regexp) {|m| Std_Abbr[m]}}
    street |= add
    street.map! {|item| expand_numbers(item)}
    street.flatten!
    street.map! {|s| s.downcase}
    street.uniq!
  else
    street = []
  end
  street
end

#intersection? ⇒ Boolean

Returns:

  • (Boolean) 


286
287
288
 
# File 'lib/geocoder/us/address.rb', line 286

def intersection?
  Match[:at].match @text
end

#parse ⇒ Object 



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
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
 
# File 'lib/geocoder/us/address.rb', line 152

def parse
  text = @text.clone.downcase

  @zip = text.scan(Match[:zip])[-1]
  if @zip
    text = parse_zip($&, text) 
  else
    @zip = @plus4 = ""
  end
  
  @state = text.scan(Match[:state])[-1]
  if @state
    text = parse_state($&, text)
  else
    @full_state = ""
    @state = ""
  end
  
  @number = text.scan(Match[:number])[0]
  # FIXME: 230 Fish And Game Rd, Hudson NY 12534
  if @number # and not intersection?
    text = parse_number($&, text)
  else
    @prenum = @number = @sufnum = ""
  end

  # FIXME: special case: Name_Abbr gets a bit aggressive
  # about replacing St with Saint. exceptional case:
  # Sault Ste. Marie

  # FIXME: PO Box should geocode to ZIP
  @street = text.scan(Match[:street])
  @street = expand_streets(@street)
  # SPECIAL CASE: 1600 Pennsylvania 20050
  @street << @full_state if @street.empty? and @state.downcase != @full_state.downcase      
 
  @city = text.scan(Match[:city])
  if !@city.empty?
    @city = [@city[-1].strip]
    add = @city.map {|item| item.gsub(Name_Abbr.regexp) {|m| Name_Abbr[m]}} 
    @city |= add
    @city.map! {|s| s.downcase}
    @city.uniq!
  else
    @city = []
  end

  # SPECIAL CASE: no city, but a state with the same name. e.g. "New York"
  @city << @full_state if @state.downcase != @full_state.downcase

  # SPECIAL CASE: if given a single city string, and it's not the
  # same as the street string, remove it from the street parts
  self.city= @city if @city.length == 1 and @city != @street
end

#parse_number(regex_match, text) ⇒ Object 



143
144
145
146
147
148
149
150
 
# File 'lib/geocoder/us/address.rb', line 143

def parse_number(regex_match, text)
  # FIXME: What if this string appears twice?
  idx = text.index(regex_match)  
  text[idx...idx+regex_match.length] = ""
  text.sub! /^\s*,?\s*/o, ""
  @prenum, @number, @sufnum = @number.map {|s| s and s.strip}
  text
end

#parse_state(regex_match, text) ⇒ Object 



134
135
136
137
138
139
140
141
 
# File 'lib/geocoder/us/address.rb', line 134

def parse_state(regex_match, text)
  idx = text.rindex(regex_match)
  text[idx...idx+regex_match.length] = ""
  text.sub! /\s*,?\s*$/o, ""
  @full_state = @state[0].strip # special case: New York
  @state = State[@full_state]
  text
end

#parse_zip(regex_match, text) ⇒ Object 



126
127
128
129
130
131
132
 
# File 'lib/geocoder/us/address.rb', line 126

def parse_zip(regex_match, text)
  idx = text.rindex(regex_match)
  text[idx...idx+regex_match.length] = ""
  text.sub! /\s*,?\s*$/o, ""
  @zip, @plus4 = @zip.map {|s|s.strip} 
  text
end

#po_box? ⇒ Boolean

Returns:

  • (Boolean) 


282
283
284
 
# File 'lib/geocoder/us/address.rb', line 282

def po_box?
  Match[:po_box].match @text
end

#remove_noise_words(strings) ⇒ Object 



239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
 
# File 'lib/geocoder/us/address.rb', line 239

def remove_noise_words(strings)
  # Don't return strings that consist solely of abbreviations.
  # NOTE: Is this a micro-optimization that has edge cases that will break?
  # Answer: Yes, it breaks on simple things like "Prairie St" or "Front St"
  prefix = Regexp.new("^" + Prefix_Type.regexp.source + "\s*", Regexp::IGNORECASE)
  suffix = Regexp.new("\s*" + Suffix_Type.regexp.source + "$", Regexp::IGNORECASE)
  predxn = Regexp.new("^" + Directional.regexp.source + "\s*", Regexp::IGNORECASE)
  sufdxn = Regexp.new("\s*" + Directional.regexp.source + "$", Regexp::IGNORECASE)
  good_strings = strings.map {|s|
    s = s.clone
    s.gsub!(predxn, "")
    s.gsub!(sufdxn, "")
    s.gsub!(prefix, "")
    s.gsub!(suffix, "")
    s
  }
  good_strings.reject! {|s| s.empty?}
  strings = good_strings if !good_strings.empty? {|s| not Std_Abbr.key?(s) and not Name_Abbr.key?(s)}
  strings
end

#street_parts ⇒ Object 



224
225
226
227
228
229
230
231
232
233
234
235
236
237
 
# File 'lib/geocoder/us/address.rb', line 224

def street_parts
  strings = []
  # Get all the substrings delimited by whitespace
  @street.each {|string|
    tokens = string.split(" ")
    strings |= (0...tokens.length).map {|i|
               (i...tokens.length).map {|j| tokens[i..j].join(" ")}}.flatten
  }
  strings = remove_noise_words(strings)

  # Try a simpler case of adding the @number in case everything is an abbr.
  strings += [@number] if strings.all? {|s| Std_Abbr.key? s or Name_Abbr.key? s}
  strings.uniq
end