pragmatic_tokenizer 0.1.9 → 0.1.10

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 0cad1673af08205c4d74c54e33a5c81af1868db1
-  data.tar.gz: 1c69082577b4a35d6cccb5120eb1f05520d6724f
+  metadata.gz: 4b2dd34906b6a89945200b65353a772089c2fdb2
+  data.tar.gz: dceed8cdbb9cefcd822ddf68cc99073b76aa9506
 SHA512:
-  metadata.gz: 15ba9389b8c229575c00834be9fb9eb43ebb5b674f2ad03ef19f2da1818e7152aaaece1af0e86cbc7a215bce7dd19075487c3bd865c56b205a6f06cd6c5659d5
-  data.tar.gz: fd97185a77a8cc88809a4a55387d28bb78b48833a1e8974ac59b583aee5ea18c9ae81e237180e0a4885443dd03efc2fd57802d1c2c199bc950dc567f160cb63e
+  metadata.gz: 21d892cc075635b1d1ba42c122421bb237bbc01b35ab2068648fb8c4eb9725a2174f22b89a41976d9df06546bdf92e1d4f4f8520a409edf1d6c2c62df1faf8cc
+  data.tar.gz: 2750c610896eeb49f8218e666be12f6deeb5675fd6bef7678f826d07031fa0fba94fd5113e0473f263899f20d5e740c4117748d48621e2a51fe1c3f835275a1a

data/lib/pragmatic_tokenizer/processor.rb

@@ -14,6 +14,7 @@ module PragmaticTokenizer
       shift_special_quotes(text)
       shift_colon(text)
       shift_bracket(text)
+      shift_semicolon(text)
       convert_dbl_quotes(text)
       convert_sgl_quotes(text)
       tokens = separate_full_stop(text.squeeze(' ').split.map { |t| convert_sym_to_punct(t.downcase) })
@@ -74,6 +75,10 @@ module PragmaticTokenizer
       text.gsub!(/:/o, ' :') || text
     end
 
+    def shift_semicolon(text)
+      text.gsub!(/([;])/o) { ' ' + $1 + ' ' } || text
+    end
+
     def shift_ellipse(text)
       text.gsub!(/(\.\.\.+)/o) { ' ' + $1 + ' ' } || text
     end

data/lib/pragmatic_tokenizer/tokenizer.rb

@@ -65,7 +65,13 @@ module PragmaticTokenizer
 
     def cleaner(tokens)
       return tokens unless clean
-      tokens.delete_if { |t| t =~ /\A_+\z/ || t =~ /\A-+\z/ || PragmaticTokenizer::Languages::Common::SPECIAL_CHARACTERS.include?(t) || t =~ /\A\.{2,}\z/ || t.include?("\\") || t.length > 50 }
+      tokens.delete_if { |t| t =~ /\A_+\z/ ||
+        t =~ /\A-+\z/ ||
+        PragmaticTokenizer::Languages::Common::SPECIAL_CHARACTERS.include?(t) ||
+        t =~ /\A\.{2,}\z/ || t.include?("\\") ||
+        t.length > 50 ||
+        (t.length > 1 && t =~ /[#&*+<=>@^|~]/i)
+      }
     end
 
     def remove_punctuation(tokens)

data/lib/pragmatic_tokenizer/version.rb

@@ -1,3 +1,3 @@
 module PragmaticTokenizer
-  VERSION = "0.1.9"
+  VERSION = "0.1.10"
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: pragmatic_tokenizer
 version: !ruby/object:Gem::Version
-  version: 0.1.9
+  version: 0.1.10
 platform: ruby
 authors:
 - Kevin S. Dias