text_alignment 0.12.1 → 0.12.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '097a3da5324e29b872a7927e2725ce82120f328ce1961cb24aaa5e058cf1b265'
-  data.tar.gz: 727bcfbffe4c30535a3b0bd01f4d14cdd1f4ccf64c5f337549da116e484e5e35
+  metadata.gz: 2345340266a0e66e9d26daaa51db1c9239bb837f52a5112a5c525a6d87b120d5
+  data.tar.gz: d3d1d118786e89a4bd7f9a6a9315643967c3ae099f8072913862ded9c895bfa5
 SHA512:
-  metadata.gz: 0bb88bc2f353bca72b1cc1488063c3ac3d9f3164c3d877bbef8b17876311a99f08143fced8d8ccaa7283cce1b4e6a1ba265435ef71d6c30979da0a5045462290
-  data.tar.gz: ef1a61192ade8cabe0b8960619f3200e1312b3cb6766d4b4dfc4a4685ab8af8397c9647cbc6e50cc6047b5ca4a2daaf7b47e669655616082fa99c06f6d3f1e26
+  metadata.gz: 9fc6c3235373f0e0174a922f006dd2cdf687361dfd567056137d20707b674ae75a40c13862d5a02946a225c19dfdae239ebf3f274a47bb4df1c8b2256bd968e2
+  data.tar.gz: 50dede22ed93d9e93a21dbbec5de39f3e342147c512d620a341331c4ddee6d0cd2cd08535e484732fd33e24ec1d8789d4daa7ef79191b13bcb9908634fcafe13

data/lib/text_alignment/char_mapping.rb

@@ -93,6 +93,7 @@ class TextAlignment::CharMapping
 
 		@text  = _text
 
+		# sort by the length of the spell-outs is important
 		char_mapping ||= TextAlignment::CHAR_MAPPING.sort{|a, b| b[1].length <=> a[1].length}
 		@mapped_text, offset_mapping = enmap_text(_text, char_mapping)
 		@index_enmap = offset_mapping.to_h
@@ -100,11 +101,11 @@ class TextAlignment::CharMapping
 	end
 
 	def enmap_position(position)
-		@index_enmap[position] || raise(ArgumentError, "Unusual position of annotation: #{position}")
+		@index_enmap[position]
 	end
 
 	def demap_position(position)
-		@index_demap[position] || raise(ArgumentError, "Unusual position of annotation: #{position}")
+		@index_demap[position]
 	end
 
 	def enmap_denotations(denotations)
@@ -112,23 +113,6 @@ class TextAlignment::CharMapping
 
 		denotations.map do |d|
 			d.dup.merge(span:{begin:enmap_position(d[:span][:begin]), end:enmap_position(d[:span][:end])})
-		rescue ArgumentError => e
-			snippet_begin = d[:span][:begin] - 5
-			if snippet_begin < 0
-				snippet_begin = 0
-			end
-			snippet_end = d[:span][:end] + 5
-			if snippet_end > @text.length
-				snippet_end = @text.length
-			end
-			snippet = @text[snippet_begin ... d[:span][:begin]] + '[' + @text[d[:span][:begin] ... d[:span][:end]] + ']' + @text[d[:span][:end] ... snippet_end]
-			if snippet_begin > 0
-				snippet = '...' + snippet
-			end
-			if snippet_end < @text.length
-				snippet = snippet + '...'
-			end
-			raise ArgumentError, e.message + " (#{snippet})"
 		end
 	end
 
@@ -137,7 +121,7 @@ class TextAlignment::CharMapping
 	def enmap_text(_text, char_mapping, no_ws = false)
 		text = _text.dup
 
-		# To execute the single letter mapping replacement
+		# To perform the single letter mapping replacement
 		char_mapping.each do |one, long|
 			text.gsub!(one, long) if long.length == 1
 		end
@@ -149,12 +133,16 @@ class TextAlignment::CharMapping
 
 			init_next = 0
 			while loc = text.index(long, init_next)
-				rpositions << [loc, long.length, 1]
+				# Huristics to check if the surrounding letters are sufficiently distinguished.
+				if long.length > 3 || ((text[loc - 1, 2] !~ /[a-z][a-z]/) && (text[loc + long.length - 1, 2] !~ /[a-z][a-z]/))
+				# if true
+					rpositions << [loc, long.length, 1]
+
+					# a workaround to avoid messing-up due to embedding
+					text[loc, long.length] = one * long.length
+				end
 				init_next = loc + long.length
 			end
-
-			# a workaround to avoid messing-up due to embedding
-			text.gsub!(long, one * long.length)
 		end
 
 		# To get the replacement positions, (position, old_length, new_length), for consecutive whitespaces
@@ -182,6 +170,7 @@ class TextAlignment::CharMapping
 
 		# To execute the long letter mapping
 		char_mapping.each do |one, long|
+			next unless text =~ /#{one}/
 			text.gsub!(one * long.length, one) if long.length > 1
 		end
 

data/lib/text_alignment/mixed_alignment.rb

@@ -17,7 +17,7 @@ class TextAlignment::MixedAlignment
 	attr_reader :similarity
 	attr_reader :str1_match_initial, :str1_match_final, :str2_match_initial, :str2_match_final
 
-	def initialize(_str1, _str2, _mappings = nil)
+	def initialize(_str1, _str2, mappings = nil)
 		raise ArgumentError, "nil string" if _str1.nil? || _str2.nil?
 
 		mappings ||= TextAlignment::CHAR_MAPPING

data/lib/text_alignment/text_alignment.rb

@@ -40,12 +40,12 @@ class TextAlignment::TextAlignment
 		end
 
 		@mapped_text = @text_mapping.mapped_text
-		denotations_mapped = @text_mapping.enmap_denotations(denotations)
 
 		## To generate the block_alignment of the input text against the reference text
 		@blocks = if r = whole_block_alignment(@mapped_text, @mapped_reference_text, @cultivation_map)
 			r
 		else
+			denotations_mapped = @text_mapping.enmap_denotations(denotations)
 			find_block_alignment(@mapped_text, @mapped_reference_text, denotations_mapped, @cultivation_map)
 		end
 
@@ -343,7 +343,7 @@ class TextAlignment::TextAlignment
 
 		## term-based alignment
 		tblocks = if denotations
-			denotations_in_scope = denotations.select{|d| d[:span][:begin] >= b1 && d[:span][:end] <= e1}.
+			denotations_in_scope = denotations.select{|d| d[:span][:begin] && d[:span][:end] && d[:span][:begin] >= b1 && d[:span][:end] <= e1}.
 							sort{|d1, d2| d1[:span][:begin] <=> d2[:span][:begin] || d2[:span][:end] <=> d1[:span][:end] }.
 							map{|d| d.merge(lex:str1[d[:span][:begin] ... d[:span][:end]])}
 

data/lib/text_alignment/version.rb

@@ -1,3 +1,3 @@
 class TextAlignment
-	VERSION = '0.12.1'
+	VERSION = '0.12.3'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: text_alignment
 version: !ruby/object:Gem::Version
-  version: 0.12.1
+  version: 0.12.3
 platform: ruby
 authors:
 - Jin-Dong Kim
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-12-26 00:00:00.000000000 Z
+date: 2024-01-05 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-dictionary