virastar 0.0.2 → 0.0.3

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    virastar (0.0.1)
+    virastar (0.0.2)
 
 GEM
   remote: http://rubygems.org/

data/TODO

@@ -0,0 +1,12 @@
+- do not put space after : in the context of numbers like hour 19:45 => ۱۹:۴۵
+- spacing after , : ; causing a lot of problem in this case (,) => (, )
+- do not destroy urls dots and colons
+
+- translate to js
+
+
+DONE:
+- he yeh => hamzeh should consider arabic yeh and zwnj chars
+- replacing quotes shouldn't be greedy 
+- for suffixes also consider tari
+- (IMP)  do not replace \n probably caused by fixing spaces after dots and commas

data/lib/virastar/version.rb

@@ -1,3 +1,3 @@
 module Virastar
-  VERSION = "0.0.2"
+  VERSION = "0.0.3"
 end

data/lib/virastar.rb

@@ -39,10 +39,10 @@ module Virastar
       text.gsub!(/\s*\.{3,}/,'…') if @fix_three_dots
 
       # replace English quotes with their Persian equivalent
-      text.gsub!(/(["'`]+)(.+)(\1)/, '«\2»') if @fix_english_quotes
+      text.gsub!(/(["'`]+)(.+?)(\1)/, '«\2»') if @fix_english_quotes
 
       # should convert ه ی to ه
-      text.gsub!(/(\S)(ه[\s‌]+ی)(\s)/, '\1هٔ\3') if @fix_hamzeh
+      text.gsub!(/(\S)(ه[\s‌]+[یي])(\s)/, '\1هٔ\3') if @fix_hamzeh
 
       # remove unnecessary zwnj char that are succeeded/preceded by a space
       text.gsub!(/\s+‌|‌\s+/,' ') if @cleanup_zwnj
@@ -80,7 +80,7 @@ module Virastar
       # put zwnj between word and suffix (*tar *tarin *ha *haye)
       # there's a possible bug here: های and تر could be separate nouns and not suffix
       if @fix_suffix_spacing
-        text.gsub!(/\s+(تر(ین)?|ها(ی)?)\s+/,'‌\1 ')
+        text.gsub!(/\s+(تر(ی(ن)?)?|ها(ی)?)\s+/,'‌\1 ') # in case you can not read it: \s+(tar(i(n)?)?|ha(ye)?)\s+
       end
 
       # -- Aggressive Editing ------------------------------------------
@@ -100,13 +100,13 @@ module Virastar
 
       # : ; , . ! ? and their persian equivalents should have one space after and no space before
       if @fix_spacing_for_braces_and_quotes
-        text.gsub!(/\s*([:;,؛،.؟!]{1})\s*/, '\1 ')
+        text.gsub!(/[ ‌	]*([:;,؛،.؟!]{1})[ ‌	]*/, '\1 ')
       end
 
       # should replace more than one space with just a single one
       if @cleanup_spacing
         text.gsub!(/[ ]+/,' ')
-        text.gsub!(/\s*[\n]+\s*/," \n")
+        #text.gsub!(/\s*[\n]+\s*/," \n")
       end
 
       # remove spaces, tabs, and new lines from the beginning and enf of file

data/spec/virastar_spec.rb

@@ -47,8 +47,10 @@ describe Virastar do
   it "should correct :;,.?! spacing (one space after and no space before)" do
     test   = "گفت : سلام"
     result = "گفت: سلام"
-    #puts Diffy::Diff.new(test, result).to_s(:color) # TODO: char diff
+    test2 = "salam.\n\nkhoobi"
+    result2 = "salam. \n\nkhoobi"
     test.persian_cleanup.should == result
+    test2.persian_cleanup.should == result2
   end
 
   it "should replace English quotes with their Persian equivalent" do
@@ -59,11 +61,15 @@ describe Virastar do
     test5 = "``تست``"
     result = result2 = result4 = result5 = "«تست»"
     result3 = "«گفت: سلام»"
+    # not greedy
+    test6 = '"this" or "that"'
+    result6 = '«this» or «that»'
     test.persian_cleanup.should  == result
     test2.persian_cleanup.should == result2
     test3.persian_cleanup.should == result3
     test4.persian_cleanup.should == result4
     test5.persian_cleanup.should == result5
+    test6.persian_cleanup.should == result6
   end
 
   it "should replace three dots with ellipsis" do
@@ -84,9 +90,11 @@ describe Virastar do
   it "should convert ه ی to هٔ"  do
     test = "خانه ی ما"
     test2 = "خانه ی ما"
-    result = result2 = "خانهٔ ما"
+    test3 = "خانه ي ما"
+    result = result2 = result3 = "خانهٔ ما"
     test.persian_cleanup.should   == result
     test2.persian_cleanup.should  == result2
+    test3.persian_cleanup.should  == result3
   end
 
   it "should replace double dash to ndash and triple dash to mdash" do
@@ -162,8 +170,8 @@ describe Virastar do
     result2 = "ما نمی‌توانیم"
     test3   = "این بهترین کتاب ها است"
     result3 = "این بهترین کتاب‌ها است"
-    test4   = "بزرگ تر و قدرتمند ترین زبان های دنیا"
-    result4 = "بزرگ‌تر و قدرتمند‌ترین زبان‌های دنیا"
+    test4   = "بزرگ تری و قدرتمند ترین زبان های دنیا"
+    result4 = "بزرگ‌تری و قدرتمند‌ترین زبان‌های دنیا"
     test.persian_cleanup.should == result
   end
 

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: virastar
 version: !ruby/object:Gem::Version 
-  hash: 27
+  hash: 25
   prerelease: 
   segments: 
   - 0
   - 0
-  - 2
-  version: 0.0.2
+  - 3
+  version: 0.0.3
 platform: ruby
 authors: 
 - Allen A. Bargi
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2011-01-19 00:00:00 +01:00
+date: 2011-01-20 00:00:00 +01:00
 default_executable: 
 dependencies: 
 - !ruby/object:Gem::Dependency 
@@ -67,6 +67,7 @@ files:
 - LICENSE
 - README.md
 - Rakefile
+- TODO
 - lib/virastar.rb
 - lib/virastar/version.rb
 - spec/spec_helper.rb