textutils 0.8.3 → 0.8.4

data/Manifest.txt

@@ -26,6 +26,7 @@ lib/textutils/sanitizier.rb
 lib/textutils/utils.rb
 lib/textutils/version.rb
 test/helper.rb
+test/test_address_helper.rb
 test/test_hypertext_helper.rb
 test/test_title_helper.rb
 test/test_unicode_helper.rb

data/lib/textutils/helper/address_helper.rb

@@ -4,21 +4,34 @@
 module TextUtils
   module AddressHelper
 
-  def normalize_addr( old_address )
-    # for now only checks german 5-digit zip code
+  def normalize_addr( old_address, country_key=nil )
+
+    # for now only checks german (de) 5-digit zip code and
+    #                    austrian (at) 4-digit zip code
     #
     #  e.g.  Alte Plauener Straße 24 // 95028 Hof  becomes
     #        95028 Hof // Alte Plauener Straße 24 
 
+    if country_key.nil?
+      puts "TextUtils.normalize_addr drepreciated call - country_key now required; please add !!"
+      return old_address
+    end
+    
     new_address = old_address   # default - do nothing - just path through
-
+    
     lines = old_address.split( '//' )
-
+    
     if lines.size == 2   # two lines / check for switching lines
+      
       line1 = lines[0].strip
       line2 = lines[1].strip
-      if line2 =~ /^[0-9]{5}\s/
-        new_address = "#{line2} // #{line1}"   # swap - let line w/ 5-digit zip code go first
+
+      regex_nnnn  = /^[0-9]{4}\s+/   # four digits postal code
+      regex_nnnnn = /^[0-9]{5}\s+/   # five digits postal code
+
+      if (country_key == 'at' && line2 =~ regex_nnnn ) ||
+         (country_key == 'de' && line2 =~ regex_nnnnn )
+        new_address = "#{line2} // #{line1}"
       end
     end
 
@@ -26,28 +39,127 @@ module TextUtils
   end
 
 
-  #  todo/fix: add _in_adr or _in_addr to name - why? why not?
-  #  -- make country_key optional - why? why not?
-  #      n move to second pos; use opts={} why? why not?
+  def find_city_in_addr_without_postal_code( address )
 
-  def find_city_in_addr( address, country_key )
+    ## general rule; not country-specific; no postal code/zip code or state
+    #  - must be like two lines (one line empty) e.g.
+    #  // London   or
+    # London //
+    #  will assume entry is city
+    #  note: city may NOT include numbers, or pipe (|) or comma (,) chars
+
+    # fix: use blank?
+    return nil if address.nil? || address.empty?    # do NOT process nil or empty address lines; sorry
 
-    return nil if address.blank?    # do NOT process nil or empty address lines; sorry
+    old_lines = address.split( '//' )
+
+    ###
+    # note:   London //   will get split into arry with size 1 e.g. ['London ']
+    #   support it, that is, add missing empty line
+
+    # 1) strip lines
+    # 2) remove blank lines
+    lines = []
+    
+    old_lines.each do |line|
+      linec = line.strip
+      next if linec.empty?
+      lines << linec
+    end
+
+    if lines.size == 1
+      linec = lines[0]
+        #  note: city may NOT include
+        #   numbers  (e.g. assumes zip/postal code etc.) or
+        #   pipe (|) or
+        #   comma (,)
+      if linec =~ /[0-9|,]/
+        return nil
+      end
+        #   more than two uppercase letters e.g. TX NY etc.
+        #  check if city exists wit tow uppercase letters??
+      if linec =~ /[A-Z]{2,}/
+        return nil
+      end
+      return linec   # bingo!!! assume candidate line is a city name
+    end
+
+    nil  # no generic city match found
+  end
+
+
+  def find_city_in_addr_with_postal_code( address, country_key )
+
+    # fix: use blank?
+    return nil if address.nil? || address.empty?    # do NOT process nil or empty address lines; sorry
 
     lines = address.split( '//' )
 
-    if country_key == 'at' || country_key == 'de'
-      # first line strip numbers (assuming zip code) and whitespace
-      line1 = lines[0]
-      line1 = line1.gsub( /\b[0-9]+\b/, '' )   # use word boundries (why? why not?)
-      line1 = line1.strip
-      
-      return nil if line1.blank?   # nothing left sorry; better return nil
-      
-      line1   # assume its the city
+    if country_key == 'at' || country_key == 'be'
+      # support for now
+      #  - 2018 Antwerpen or 2870 Breendonk-Puurs (be)
+      lines.each do |line|
+        linec = line.strip
+        regex_nnnn = /^[0-9]{4}\s+/ 
+        if linec =~ regex_nnnn   # must start w/ four digit postal code ? assume its the city line
+          return linec.sub( regex_nnnn, '' )  # cut off leading postal code; assume rest is city
+        end
+      end
+    elsif country_key == 'de'
+      lines.each do |line|
+        linec = line.strip
+        regex_nnnnn = /^[0-9]{5}\s+/
+        if linec =~ regex_nnnnn   # must start w/ five digit postal code ? assume its the city line
+          return linec.sub( regex_nnnnn, '' )  # cut off leading postal code; assume rest is city
+        end
+      end
+    elsif country_key == 'cz' || country_key == 'sk'
+      # support for now
+      #  - 284 15  Kutná Hora or  288 25  Nymburk (cz)
+      #  - 036 42  Martin     or  974 05  Banská Bystrica (sk)
+      lines.each do |line|
+        linec = line.strip
+        regex_nnn_nn = /^[0-9]{3}\s[0-9]{2}\s+/
+        if linec =~ regex_nnn_nn   # must start w/ five digit postal code ? assume its the city line
+          return linec.sub( regex_nnn_nn, '' )  # cut off leading postal code; assume rest is city
+        end
+      end
+    elsif country_key == 'us'
+      # support for now
+      #  - Brooklyn | NY 11249  or Brooklyn, NY 11249
+      #  - Brooklyn | NY   or Brooklyn, NY
+
+      lines.each do |line|
+        linec = line.strip
+        regexes_us = [/\s*[|,]\s+[A-Z]{2}\s+[0-9]{5}\s*$/,
+                      /\s*[|,]\s+[A-Z]{2}\s*$/]
+        
+        regexes_us.each do |regex|
+          if linec =~ regex
+            return linec.sub( regex, '' )  # cut off leading postal code; assume rest is city
+          end
+        end
+      end
     else
-      nil   # unsupported country/address schema for now; sorry
+      # unsupported country/address schema for now; sorry
     end
+    return nil   # sorry nothing found
+  end
+
+
+  def find_city_in_addr( address, country_key )
+
+    # fix: use blank?
+    return nil if address.nil? || address.empty?    # do NOT process nil or empty address lines; sorry
+
+    ## try geneneric rule first (e.g. w/o postal code/zip code or state), see above
+    city = find_city_in_addr_without_postal_code( address )
+    return city unless city.nil?
+    
+    city = find_city_in_addr_with_postal_code( address, country_key )
+    return city unless city.nil?
+
+    nil # sorry; no city found (using known patterns)
   end
 
 

data/lib/textutils/helper/title_helper.rb

@@ -49,10 +49,11 @@ module TextUtils
   end
 
   def strip_special_chars( title )
-      # remove special chars (e.g. %°&)
+      # remove special chars (e.g. %°&$)
       # e.g. +Malta
       #      Minerva 8:60
-      title.gsub( /[%&°+:]/, '' )
+      #      $Alianz$ Arena
+      title.gsub( /[%&°+:$]/, '' )
   end
 
   def title_to_key( title )
@@ -112,6 +113,7 @@ module TextUtils
         ['ő', 'o' ],  # e.g. Győri
         ['ó', 'o' ],  # e.g. Colón, Łódź, Kraków
         ['õ', 'o' ],  # e.g. Nõmme
+        ['ô', 'o' ],  # e.g. Amazônia (pt)
         ['ø', 'o' ],  # e.g. Fuglafjørdur, København
         ['ř', 'r' ],  # e.g. Třeboň
         ['ș', 's' ],  # e.g. Chișinău, București

data/lib/textutils/version.rb

@@ -1,7 +1,7 @@
 
 module TextUtils
 
-  VERSION = '0.8.3'
+  VERSION = '0.8.4'
 
 end   # module TextUtils
 

data/test/test_address_helper.rb

@@ -0,0 +1,108 @@
+# encoding: utf-8
+
+
+require 'helper'
+
+class TestAddressHelper < MiniTest::Unit::TestCase
+
+  def test_normalize_addr
+    
+    txt_io = [
+      ['Alte Plauener Straße 24 // 95028 Hof', nil, 'Alte Plauener Straße 24 // 95028 Hof'],
+      ['Alte Plauener Straße 24 // 95028 Hof', 'de', '95028 Hof // Alte Plauener Straße 24'],
+      ['Mautner Markhof-Straße 11 // 2320 Schwechat', nil, 'Mautner Markhof-Straße 11 // 2320 Schwechat'],
+      ['Mautner Markhof-Straße 11 // 2320 Schwechat', 'at', '2320 Schwechat // Mautner Markhof-Straße 11']
+    ]
+
+    txt_io.each_with_index do |txt,i|
+      puts "testing [#{i}] #{txt[0]}"
+      assert_equal txt[2], TextUtils.normalize_addr( txt[0], txt[1] )
+    end
+
+  end  # method test_normalize_addr
+
+
+  def test_addr_without_postal_code  # aka generic rule
+
+    txt_io = [
+      ['London //',  'London'],
+      ['// London',  'London'],
+      ['// London  ',  'London'],
+      ['  // London',  'London'],
+      ['// London, W4 2QB', nil],
+      ['// London | W4 2QB', nil],
+      ['// London  W4 2QB', nil],
+      ['Chiswick Lane South // London, W4 2QB', nil],
+      ['The Griffin Brewery // Chiswick Lane South // London', nil], # three lines will NOT work, sorry
+      ['// New York, NY', nil],
+      ['// New York NY', nil]    # check: does it exist in the real world (e.g. w/o comma or pipe?) support it?
+    ]
+
+    txt_io.each_with_index do |txt,i|
+      puts "testing [#{i}] #{txt[0]}"
+      assert_equal txt[1], TextUtils.find_city_in_addr_without_postal_code( txt[0] )
+    end
+  end # method test_addr_without_postal_code
+
+
+  def test_addr_with_postal_code
+
+    txt_io = [
+      ['2320 Schwechat // Mautner Markhof-Straße 11', 'at', 'Schwechat'],
+      ['Mautner Markhof-Straße 11 // 2320 Schwechat', 'at', 'Schwechat'],
+      ['3910 Zwettl // Syrnauer Straße 22-25', 'at', 'Zwettl'],
+      ['Syrnauer Straße 22-25 // 3910 Zwettl', 'at', 'Zwettl'],
+      ['2018 Antwerpen', 'be', 'Antwerpen'],
+      ['2870 Breendonk-Puurs', 'be', 'Breendonk-Puurs'],
+      ['Alte Plauener Straße 24 // 95028 Hof', 'de', 'Hof'],
+      ['95028 Hof // Alte Plauener Straße 24', 'de', 'Hof'],
+      ['284 15 Kutná Hora', 'cz', 'Kutná Hora'],
+      ['288 25 Nymburk', 'cz', 'Nymburk'],
+      ['036 42 Martin', 'sk', 'Martin'],
+      ['974 05 Banská Bystrica', 'sk', 'Banská Bystrica'],
+      ['Brooklyn | NY 11249', 'us', 'Brooklyn'],
+      ['Brooklyn, NY 11249', 'us', 'Brooklyn'],
+      ['Brooklyn | NY', 'us', 'Brooklyn'],
+      ['Brooklyn, NY', 'us', 'Brooklyn'],
+    ]
+
+    txt_io.each_with_index do |txt,i|
+      puts "testing [#{i}] #{txt[0]}"
+      assert_equal txt[2], TextUtils.find_city_in_addr_with_postal_code( txt[0], txt[1] )
+    end
+  end # method test_addr_with_postal_code
+
+
+  def test_addr
+
+    txt_io = [
+      ['London //', nil, 'London'],
+      ['// London', nil, 'London'],
+      ['// London  ', nil,  'London'],
+      ['  // London', nil, 'London'],
+      ['2320 Schwechat // Mautner Markhof-Straße 11', 'at', 'Schwechat'],
+      ['Mautner Markhof-Straße 11 // 2320 Schwechat', 'at', 'Schwechat'],
+      ['3910 Zwettl // Syrnauer Straße 22-25', 'at', 'Zwettl'],
+      ['Syrnauer Straße 22-25 // 3910 Zwettl', 'at', 'Zwettl'],
+      ['2018 Antwerpen', 'be', 'Antwerpen'],
+      ['2870 Breendonk-Puurs', 'be', 'Breendonk-Puurs'],
+      ['Alte Plauener Straße 24 // 95028 Hof', 'de', 'Hof'],
+      ['95028 Hof // Alte Plauener Straße 24', 'de', 'Hof'],
+      ['284 15 Kutná Hora', 'cz', 'Kutná Hora'],
+      ['288 25 Nymburk', 'cz', 'Nymburk'],
+      ['036 42 Martin', 'sk', 'Martin'],
+      ['974 05 Banská Bystrica', 'sk', 'Banská Bystrica'],
+      ['Brooklyn | NY 11249', 'us', 'Brooklyn'],
+      ['Brooklyn, NY 11249', 'us', 'Brooklyn'],
+      ['Brooklyn | NY', 'us', 'Brooklyn'],
+      ['Brooklyn, NY', 'us', 'Brooklyn'],
+    ]
+
+    txt_io.each_with_index do |txt,i|
+      puts "testing [#{i}] #{txt[0]}"
+      assert_equal txt[2], TextUtils.find_city_in_addr( txt[0], txt[1] )
+    end
+  end # method test_addr
+
+
+end # class TestAddressHelper

data/test/test_title_helper.rb

@@ -34,7 +34,9 @@ class TestTitleHelper < MiniTest::Unit::TestCase
       [ '+Lupulus', 'lupulus' ],
       [ '+Malta', 'malta' ],
       [ 'Minerva 8:60', 'minerva860' ],
-      [ 'Hop Crisis!', 'hopcrisis' ]
+      [ 'Hop Crisis!', 'hopcrisis' ],
+      [ '$Alianz$ Arena', 'alianzarena' ],
+      [ 'Arena Amazônia', 'arenaamazonia' ]
     ]
 
     txt_io.each do |txt|

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: textutils
 version: !ruby/object:Gem::Version
-  version: 0.8.3
+  version: 0.8.4
   prerelease: 
 platform: ruby
 authors:
@@ -9,11 +9,11 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-01-23 00:00:00.000000000 Z
+date: 2014-02-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: logutils
-  requirement: &21298812 !ruby/object:Gem::Requirement
+  requirement: &20379624 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -21,10 +21,10 @@ dependencies:
         version: '0.5'
   type: :runtime
   prerelease: false
-  version_requirements: *21298812
+  version_requirements: *20379624
 - !ruby/object:Gem::Dependency
   name: rdoc
-  requirement: &21298416 !ruby/object:Gem::Requirement
+  requirement: &20377728 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -32,10 +32,10 @@ dependencies:
         version: '4.0'
   type: :development
   prerelease: false
-  version_requirements: *21298416
+  version_requirements: *20377728
 - !ruby/object:Gem::Dependency
   name: hoe
-  requirement: &21297996 !ruby/object:Gem::Requirement
+  requirement: &20376972 !ruby/object:Gem::Requirement
     none: false
     requirements:
     - - ~>
@@ -43,7 +43,7 @@ dependencies:
         version: '3.7'
   type: :development
   prerelease: false
-  version_requirements: *21297996
+  version_requirements: *20376972
 description: textutils - Text Filters, Helpers, Readers and More
 email: ruby-talk@ruby-lang.org
 executables: []
@@ -81,6 +81,7 @@ files:
 - lib/textutils/utils.rb
 - lib/textutils/version.rb
 - test/helper.rb
+- test/test_address_helper.rb
 - test/test_hypertext_helper.rb
 - test/test_title_helper.rb
 - test/test_unicode_helper.rb
@@ -114,6 +115,7 @@ signing_key:
 specification_version: 3
 summary: textutils - Text Filters, Helpers, Readers and More
 test_files:
+- test/test_address_helper.rb
 - test/test_hypertext_helper.rb
 - test/test_title_helper.rb
 - test/test_unicode_helper.rb