blackwinter-perseus_match 0.0.4 → 0.0.6

data/README

@@ -2,7 +2,7 @@
 
 == VERSION
 
-This documentation refers to perseus_match version 0.0.4
+This documentation refers to perseus_match version 0.0.6
 
 
 == DESCRIPTION

data/bin/perseus_match

@@ -8,6 +8,7 @@ require 'set'
 require 'rubygems'
 require 'nuggets/enumerable/minmax'
 require 'nuggets/numeric/duration'
+require 'nuggets/string/evaluate'
 
 $: << File.join(File.dirname(__FILE__), '..', 'lib')
 
@@ -26,6 +27,7 @@ options = {
   :minimal      => false,
   :separate     => false,
   :lingo        => false,
+  :format       => nil,
   :check        => false,
   :failed_only  => false,
   :align        => false,
@@ -84,6 +86,10 @@ OptionParser.new { |opts|
     options[:lingo] = true
   }
 
+  opts.on('-F', '--format FORMAT', 'Custom output format. Available placeholders:', '  %p = phrase', '  %P = phrase, CSV-ready', '  %t = target', '  %T = target, CSV-ready', '  %d = distance', '  %s = similarity') { |f|
+    options[:format] = f
+  }
+
   opts.separator ' '
   opts.separator '  * Checking pairs'
   opts.separator ' '
@@ -265,10 +271,37 @@ action = if options[:check]
     _action
   end
 else
-  format =
-    options[:lingo] ? lambda { |pm| "#{pm.phrase}*#{pm.target}" } :
-    options[:sort]  ? lambda { |pm| "  #{[pm.target, pm.distance, pm.similarity].inspect}" } :
-                      lambda { |pm| [pm.phrase, pm.target, pm.distance, pm.similarity].inspect }
+  format = if _format = options[:format]
+    substitutions = {
+      'p' => ['#{pm.phrase}',                     's'],
+      'P' => ['"#{pm.phrase.gsub(/"/, %q{""})}"', 's'],
+      't' => ['#{pm.target}',                     's'],
+      'T' => ['"#{pm.target.gsub(/"/, %q{""})}"', 's'],
+      'd' => ['#{pm.distance}',                   'd'],
+      's' => ['#{pm.similarity}',                 'f']
+    }
+
+    lambda { |pm|
+      _format.gsub(/(%-?[.\d]*)([pPtTds])/) {
+        value, field = substitutions[$2]
+        "#{$1}#{field}" % value.evaluate(binding)
+      }
+    }
+  else
+    if options[:lingo]
+      if options[:minimal]
+        lambda { |pm| ["#{pm.phrase}*#{pm.target}", "#{pm.target}*#{pm.phrase}"] }
+      else
+        lambda { |pm| "#{pm.phrase}*#{pm.target}" }
+      end
+    else
+      if options[:sort]
+        lambda { |pm| "  #{[pm.target, pm.distance, pm.similarity].inspect}" }
+      else
+        lambda { |pm| [pm.phrase, pm.target, pm.distance, pm.similarity].inspect }
+      end
+    end
+  end
 
   if options[:sort]
     lambda {
@@ -290,13 +323,14 @@ else
       PerseusMatch::List.pair(phrases, pm_options, list_options) { |pm|
         count_all += 1
 
-        if separator && pm.phrase != previous_phrase ||= pm.phrase
-          puts separator
-          previous_phrase = pm.phrase
-        end
-
         if pm.similarity >= threshold
           count += 1
+
+          if separator && pm.phrase != previous_phrase ||= pm.phrase
+            puts separator
+            previous_phrase = pm.phrase
+          end
+
           puts format[pm]
         end
       }

data/lib/perseus_match/token_set.rb

@@ -26,7 +26,7 @@
 ###############################################################################
 #++
 
-$KCODE = 'u'
+$KCODE = 'u' unless RUBY_VERSION >= '1.9'
 
 require 'pathname'
 require 'rbconfig'
@@ -81,12 +81,16 @@ class PerseusMatch
     def self.tokenize(form, unknowns = false)
       return @tokens[form] if @tokens
 
-      @_tokens, @tokens = {}, Hash.new { |h, k| h[k] = new(
-        k, (@_tokens[k] || []) | k.scan(/\w+/).map { |i| @_tokens[i] }.flatten.compact
-      )}
+      @_tokens, @tokens = {}, Hash.new { |h, k|
+        h[k] = new(
+          k, (@_tokens[k] || []) | (
+            k.scan(/\w+/) + k.scan(/[\w-]+/)
+          ).map { |i| @_tokens[i] }.flatten.compact
+        )
+      }
 
       parse = lambda { |x|
-        x.each { |res|
+        x.each_line { |res|
           case res
             when /<(.*?)\s=\s\[(.*)\]>/
               a, b = $1, $2
@@ -130,10 +134,12 @@ class PerseusMatch
           file = temp.path
         end
 
+        ruby = Config::CONFIG.values_at('RUBY_INSTALL_NAME', 'EXEEXT').join
+
         begin
-          Dir.chdir(LINGO_BASE) { parse[%x{
-            #{Config::CONFIG['ruby_install_name']} lingo.rb -c "#{cfg.path}" < "#{file}"
-          }] }
+          Dir.chdir(LINGO_BASE) {
+            parse[%x{#{ruby} lingo.rb -c "#{cfg.path}" < "#{file}"}]
+          }
         ensure
           cfg.unlink
           temp.unlink if temp

data/lib/perseus_match/version.rb

@@ -4,7 +4,7 @@ class PerseusMatch
 
     MAJOR = 0
     MINOR = 0
-    TINY  = 4
+    TINY  = 6
 
     class << self
 

data/spec/perseus_match/cluster_spec.rb

@@ -3,14 +3,14 @@ describe PerseusMatch::Cluster do
   it 'should accept limit option in sort_by' do
     PerseusMatch::Cluster.new(%w[foo bar]).sort_by(:similarity, :limit => 1).all? { |phrase, matches|
       matches.size.should == 1
-      matches.size.should == matches.nitems
+      matches.should_not include(nil)
     }
   end
 
   it 'should accept threshold option in sort_by (1a)' do
     PerseusMatch::Cluster.new(%w[foo bar]).sort_by(:similarity, :threshold => 0.1).all? { |phrase, matches|
       matches.size.should == 1
-      matches.size.should == matches.nitems
+      matches.should_not include(nil)
       matches.each { |match| match.target.should == phrase }
     }
   end
@@ -18,28 +18,28 @@ describe PerseusMatch::Cluster do
   it 'should accept threshold option in sort_by (1b)' do
     PerseusMatch::Cluster.new(%w[foo bar]).sort_by(:similarity, :threshold => 0).all? { |phrase, matches|
       matches.size.should == 2
-      matches.size.should == matches.nitems
+      matches.should_not include(nil)
     }
   end
 
   it 'should accept threshold option in sort_by (2)' do
     PerseusMatch::Cluster.new(%w[foo bar]).sort_by(:target, :threshold => 'c').all? { |phrase, matches|
       matches.size.should == 1
-      matches.size.should == matches.nitems
+      matches.should_not include(nil)
     }
   end
 
   it 'should accept both limit and threshold options in sort_by (1)' do
     PerseusMatch::Cluster.new(%w[foo bar]).sort_by(:target, :threshold => 'z', :limit => 1).all? { |phrase, matches|
       matches.size.should == 1
-      matches.size.should == matches.nitems
+      matches.should_not include(nil)
     }
   end
 
   it 'should accept both limit and threshold options in sort_by (2)' do
     PerseusMatch::Cluster.new(%w[foo bar]).sort_by(:target, :threshold => 'a', :limit => 1).all? { |phrase, matches|
       matches.size.should be_zero
-      matches.size.should == matches.nitems
+      matches.should_not include(nil)
     }
   end
 

data/spec/perseus_match_spec.rb

@@ -1,3 +1,5 @@
+# encoding: utf-8
+
 require 'rubygems'
 require 'nuggets/tempfile/open'
 require 'nuggets/util/i18n'

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: blackwinter-perseus_match
 version: !ruby/object:Gem::Version 
-  version: 0.0.4
+  version: 0.0.6
 platform: ruby
 authors: 
 - Jens Wille
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2009-01-13 00:00:00 -08:00
+date: 2009-01-26 00:00:00 -08:00
 default_executable: perseus_match
 dependencies: 
 - !ruby/object:Gem::Dependency