camdict 1.0.0 → 1.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 86357de26d1ad3547925075d9592facaadcb39f4
-  data.tar.gz: e16be0c399d94659828ad176a9c7557d2289dfda
+  metadata.gz: 556f016868d71a2d1dde23319a9d59dfa60ba05c
+  data.tar.gz: cb0090a31d8f56da0d366819f0de47309d9d5dfa
 SHA512:
-  metadata.gz: b35d9c52c329f91c84cd644520d440d17268f52b29c1eb4328d2ea5b90bedf2cbbcd7308eeb37a919ed0be4fbed16a1ff31d6a2909c72786ec42feaaee92e224
-  data.tar.gz: ac6ce320b885258a6108365e056cfd4b8bb8a6601d950862fcfe5806ce77525807b6894f564063e6b949970843c936f8193cc12d157ed89cf0725becfc24c969
+  metadata.gz: 3c110ea5cfe31a35bd69952f983669377a577aa1f7960ac7f69dd7d073d25ff6ff2f031b4a9ba9aa7190b169ded33dba3b06df70788bccb17be77fb93cb1196f
+  data.tar.gz: dda29ecc637f0550bddfd78b23071210649e8c9aa2d58dbc34cf14b764ec0f7e92bc5c917182bd73a4da0e62516c7cdf03dac647c1b30ce5e584ae30fba54604

data/README.md

@@ -19,6 +19,8 @@ One test may fail if the gem nokogiri hasn't pulled in the fix [here][2]. But
 it is safe to apply the patch to your nokogiri copy.
 
 ## Usage
+
+```ruby
     require 'camdict'
 
     # Look up a new word
@@ -48,7 +50,7 @@ it is safe to apply the patch to your nokogiri copy.
       # I had to give up drinking for health reasons.
       # He gave up work because of ill health.
     }
-
+```
 
 There are some useful testing examples in test directory of this gem.
 

data/lib/camdict/client.rb

@@ -124,7 +124,7 @@ module Camdict
     def matched_word?(word, node)
       li = node.css(".base")
       resword = li.size == 1 ? li.text : li[0].text
-      if resword.include? '/'
+      if resword.include? '/' or resword.include? ';'
         resword.flatten.include?(word)
       else
         word == resword

data/lib/camdict/common.rb

@@ -5,51 +5,76 @@ module Camdict
     String.class_eval do
       # 'blow a kiss to/at sb'.flatten =>
       # %q(blow a kiss to sb, blow a kiss at sb)
-      # if it doesn't include a slash, returns unchanged itself 
+      # if it doesn't include a slash, returns stripped string
       def flatten
-        return self unless self.include? '/'
+        str = self.strip
+        # remove the space surrounding '/'
+        str = str.gsub /\s*\/\s*/, '/'
+        return str unless str.include? '/'
+        len = str.length
         ret = []
-        len = self.length
+        # when two strings are passed in separated with ';', then separate them
+        if pos = str.index(';')
+          ret += str[0..pos-1].flatten
+          ret += str[pos+1..len-1].flatten
+          return ret
+        end
+        # when a string has round brackets meaning optional part
+        if str.include? '('
+          head, bracket, tail = str.partition(/\(.*\)/)
+          unless bracket.empty?
+            ret << (head.strip + tail).flatten
+            result = bracket.delete("()").flatten
+            result = [result] if result.is_a? String
+            result.each { |s|
+              ret << (head + s + tail).flatten
+            }
+          end
+          return ret.flatten
+        end
         j=0     # count of the alternative words, 'to/at' has two.
-        b=[]    # b[]/e[] index of the beginning/end of a alternative word
+        b=[]    # b[]/e[] index of the beginning/end of alternative words
         e=[]
         # set this flag when next word is expected an alternate word after slash
         include_next = false 
-        for i in (0..len-1)
-          c = self[i] 
-          case 
-          when c =~ /[[:alpha:]\-\(\)]/
+        for i in 0..len-1
+          c = str[i] 
+          case c
+          # valid char in a word
+          when /[[:alnum:]\-']/
             if b[j].nil?
               b[j] = i
               e[j] = i
             else
               e[j] = i
             end
-          when c == " " 
+          # char means a word has ended
+          when " ", "!", "?", ",", "."
             if include_next
               break
             else
               b[j] = nil
               e[j] = nil
             end
-          when c == "/"
+          # 'or' separator
+          when "/"
             j += 1
             include_next = true
           else
-            raise "Invalid char '#{c}' found in a string."
+            raise NotImplementedError, "char '#{c}' found in '#{self}'."
           end
         end
         if j > 0
           for i in (0..j)
             # alternative word is not the last word and not at the beginning
             if (e[j]+1 < len) && (b[0] > 0)
-              ret << self[0..b[0]-1] + self[b[i]..e[i]] + self[e[j]+1..len-1]
+              ret << str[0..b[0]-1] + str[b[i]..e[i]] + str[e[j]+1..len-1]
             elsif (e[j]+1 == len) && (b[0] > 0)
-              ret << self[0..b[0]-1] + self[b[i]..e[i]]
+              ret << str[0..b[0]-1] + str[b[i]..e[i]]
             elsif (e[j]+1 < len) && (b[0] == 0)
-              ret << self[b[i]..e[i]] + self[e[j]+1..len-1]
+              ret << str[b[i]..e[i]] + str[e[j]+1..len-1]
             else
-              ret << self[b[i]..e[i]]
+              ret << str[b[i]..e[i]]
             end
           end 
         end

data/lib/camdict/definition.rb

@@ -44,7 +44,7 @@ module Camdict
     # tense of this verb.
     attr_reader :verb
 
-    # Input are +word+ and +entry_html+ is 
+    # Input +word+ and +entry_html+ are
     # { entry ID => its html definition source }
     def initialize(word, entry_html)
       @word = word
@@ -138,13 +138,13 @@ module Camdict
       location = "idiom" if @is_idiom && @title_word.include?(@word)
       unless @spelling_variant.nil?
         # spelling variant is treated as "title word"
-        location = "spellvar" if @spelling_variant.include? @word
+        location = "spellvar" if @spelling_variant == @word
       end
       unless @head_variant.nil?
         location = "head_variant" if @head_variant.include? @word
       end
-      location ="body_variant" if @body_variant && @body_variant.include?(@word)
-      location = "inflection" if @inflection && @inflection.include?(@word)
+      location ="body_variant" if @body_variant && @body_variant == @word
+      location = "inflection" if @inflection && @inflection == @word
       unless @derived_words.nil?
         if @derived_words.include? @word
           unless location.nil?
@@ -268,6 +268,7 @@ module Camdict
           # US IPA is always followed by a symbol US
           # favorite: UK/US ipa (spellvar US s:favorite) => normal title word
           usnode = @html.css ".di-info img.ussymbol + .pron .ipa"
+          usnode = usnode.first
           usbase = parse_ipa(usnode) unless usnode.nil?
         when 'inflection'
           usnode = @html.css ".info-group img.ussymbol + .pron .ipa"
@@ -302,7 +303,7 @@ module Camdict
             derived_uk = parse_ipa(node.first) unless node.first.nil?
           }
           derived_css("img.ussymbol + .pron .ipa") { |node|
-            usbase = parse_ipa(node) unless node.nil?
+            usbase = parse_ipa(node.first) unless node.first.nil?
           }
           if derived_uk && derived_uk[:baseipa].include?('-')
             ukbase = join_ipa(ukbase, derived_uk)
@@ -366,7 +367,15 @@ module Camdict
             ret = full[0..position-slen+2] + center + full[position+1..flen-1]
             return {baseipa: ret, sindex: findex}
           end
-          raise "unmatched head-tail hyphen IPA"
+          # this is a simple solution to workaround the issue since no common
+          # chars are found between the full and short ipa. Such as the word
+          # 'difference', so just assign full to short
+          begin
+            raise "head-tail hyphen IPA #{short} for the word #{@word}" +
+              "unmatched with #{full}."
+          rescue RuntimeError
+            return full_sp
+          end
         else
           # head hyphen
           right = short[1, slen-1]
@@ -392,23 +401,16 @@ module Camdict
       # tail hyphen
       elsif short[-1] == '-'
         left = short[0, slen-1]
-        # match left
-        # unicode of secondary stress & stress mark are considered
-        if  ["\u{2cc}", "\u{2c8}"].include? left[0]
-          if left[0,2] == full[0,2]
-            ret = left + full[slen-1..flen-1]  
-            findex = mix_spi( ussp, 0, basesp, slen-1..flen-1)
-            return {baseipa: ret, sindex: findex}
-          end
-        elsif left[0] == full[0]
-          ret = left + full[slen-1..flen-1]  
-          findex = mix_spi( ussp, 0, basesp, slen-1..flen-1)
-          return {baseipa: ret, sindex: findex}
-        else
-          raise "tail hyphen has uncovered case - code needs update."
-        end
+        ret = left + full[slen-1..flen-1]
+        findex = mix_spi( ussp, 0, basesp, slen-1..flen-1)
+        return {baseipa: ret, sindex: findex}
+      # begin with a primary or secondary stress mark like reunion
+      elsif ["\u{2cc}", "\u{2c8}"].include? short[0]
+        return full_sp # for simple, use uk ipa instead
       else
-        raise "IPA doesn't begin or end with a hyphen, nothing is done."
+        raise ArgumentError,
+          "IPA doesn't begin with a hyphen or stress, nor end with a hyphen. " +
+          "Nothing is done."
       end
     end
 

data/test/test_common.rb

@@ -18,6 +18,38 @@ module Camdict
       str = "not give/budge/move an inch"
       expected = ['not give an inch', 'not budge an inch', 'not move an inch']
       assert_equal expected, str.flatten
+      str = "fall into the/sb's trap"
+      expected = ['fall into the trap', 'fall into sb\'s trap']
+      assert_equal expected, str.flatten
+      str = "what is sb/sth?"
+      expected = ['what is sb?', 'what is sth?']
+      assert_equal expected, str.flatten
+      str = "look lively/sharp!"
+      expected = ['look lively!', 'look sharp!']
+      assert_equal expected, str.flatten
+      str = "the like of sb/sth; sb's/sth's like"
+      expected = ['the like of sb', 'the like of sth', 
+        "sb's like", "sth's like"]
+      assert_equal expected, str.flatten
+      str = "go (like/down) a bomb"
+      expected = ['go a bomb', 'go like a bomb', 'go down a bomb']
+      assert_equal expected, str.flatten
+      str = "the other side/end (of sth)"
+      expected = ['the other side', 'the other end', 'the other side of sth',
+        'the other end of sth']
+      assert_equal expected, str.flatten
+      strs = ["20/20 vision", "public enemy number one/no. 1"]
+      assert_nothing_raised do
+        strs.each { |s| s.flatten }
+      end
+      str = "the more...the more/less"
+      expected = ['the more...the more', 'the more...the less']
+      assert_equal expected, str.flatten
+      # need more examples to support complex 'or' separators 
+      #   sound like/as if/as though
+      #   look on/upon sb/sth as sth
+      #   look at/see sth through rose-coloured/tinted glasses
+      #   give /quote sth/sb chapter and verse
     end
 
     def test_expand

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: camdict
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.0.2
 platform: ruby
 authors:
 - Pan Gaoyong
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-01-05 00:00:00.000000000 Z
+date: 2014-04-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: nokogiri
@@ -25,7 +25,7 @@ dependencies:
     - - '>='
       - !ruby/object:Gem::Version
         version: '0'
-description: get definitions, pronunciation and example sentences of a word or phrase
+description: Get definitions, pronunciation and example sentences of a word or phrase
   from the online Cambridge dictionaries.
 email: pan.gaoyong@gmail.com
 executables: []