proiel-cli 1.2.0 → 1.3.0

data/lib/proiel/cli/converters/text.rb

@@ -1,98 +1,96 @@
-module PROIEL
-  module Converter
-    class Text
-      class << self
-        def process(tb, options)
-          tb.sources.each do |source|
-            puts "% id = #{source.id}"
-            puts "% export_time = #{source.export_time}"
-            puts "% title = #{source.title}"
-            puts "% author = #{source.author}"
-            puts "% citation_part = #{source.citation_part}"
-            puts "% language = #{source.language}"
-
-            source.divs.each do |div|
-              puts
-              puts "# #{div.title}"
-              puts
-
-              if options['diffable']
-                print_diffable_div(div)
-              else
-                print_formatted_div(div)
-              end
+module PROIEL::Converter
+  class Text
+    class << self
+      def process(tb, options)
+        tb.sources.each do |source|
+          puts "% id = #{source.id}"
+          puts "% export_time = #{source.export_time}"
+          puts "% title = #{source.title}"
+          puts "% author = #{source.author}"
+          puts "% citation_part = #{source.citation_part}"
+          puts "% language = #{source.language}"
+
+          source.divs.each do |div|
+            puts
+            puts "# #{div.title}"
+            puts
+
+            if options['diffable']
+              print_diffable_div(div)
+            else
+              print_formatted_div(div)
             end
           end
         end
+      end
 
-        def print_formatted_div(div)
-          p = ''
-          p += div.presentation_before unless div.presentation_before.nil?
-
-          current_citation = nil
+      def print_formatted_div(div)
+        p = ''
+        p += div.presentation_before unless div.presentation_before.nil?
 
-          div.sentences.each do |sentence|
-            p += sentence.presentation_before unless sentence.presentation_before.nil?
+        current_citation = nil
 
-            sentence.tokens.each do |token|
-              if token.has_content?
-                new_citation = token.citation_part
+        div.sentences.each do |sentence|
+          p += sentence.presentation_before unless sentence.presentation_before.nil?
 
-                if current_citation != new_citation
-                  p += "§#{new_citation.gsub(/\s+/, '_')} "
-                  current_citation = new_citation
-                end
+          sentence.tokens.each do |token|
+            if token.has_content?
+              new_citation = token.citation_part
 
-                p += [token.presentation_before,
-                      token.form,
-                      token.presentation_after].compact.join
+              if current_citation != new_citation
+                p += "§#{new_citation.gsub(/\s+/, '_')} "
+                current_citation = new_citation
               end
-            end
 
-            p += sentence.presentation_after unless sentence.presentation_after.nil?
+              p += [token.presentation_before,
+                    token.form,
+                    token.presentation_after].compact.join
+            end
           end
 
-          p += div.presentation_after unless div.presentation_after.nil?
+          p += sentence.presentation_after unless sentence.presentation_after.nil?
+        end
 
-          p = p.strip.gsub(/ +/, ' ').split("\n").collect do |line|
-              line.length > 80 ? line.gsub(/(.{1,80})(\s+|$)/, "\\1\n").strip : line
-            end * "\n"
+        p += div.presentation_after unless div.presentation_after.nil?
 
-          puts p
-        end
+        p = p.strip.gsub(/ +/, ' ').split("\n").collect do |line|
+            line.length > 80 ? line.gsub(/(.{1,80})(\s+|$)/, "\\1\n").strip : line
+          end * "\n"
 
-        def print_diffable_div(div)
-          current_citation = nil
+        puts p
+      end
 
-          p = ''
-          pb = div.presentation_before || ''
+      def print_diffable_div(div)
+        current_citation = nil
 
-          div.sentences.each do |sentence|
-            pb += sentence.presentation_before || ''
+        p = ''
+        pb = div.presentation_before || ''
 
-            sentence.tokens.each do |token|
-              if token.has_content?
-                if current_citation != token.citation_part
-                  puts p unless p.empty?
-                  p = "§#{token.citation_part.gsub(/\s+/, '_')} "
-                  current_citation = token.citation_part
-                end
+        div.sentences.each do |sentence|
+          pb += sentence.presentation_before || ''
 
-                p += [pb,
-                      token.presentation_before,
-                      token.form,
-                      token.presentation_after].compact.join
-                pb = ''
+          sentence.tokens.each do |token|
+            if token.has_content?
+              if current_citation != token.citation_part
+                puts p unless p.empty?
+                p = "§#{token.citation_part.gsub(/\s+/, '_')} "
+                current_citation = token.citation_part
               end
-            end
 
-            p += sentence.presentation_after unless sentence.presentation_after.nil?
+              p += [pb,
+                    token.presentation_before,
+                    token.form,
+                    token.presentation_after].compact.join
+              pb = ''
+            end
           end
 
-          p += div.presentation_after unless div.presentation_after.nil?
-
-          puts p unless p.empty?
+          p += sentence.presentation_after unless sentence.presentation_after.nil?
         end
+
+        p += div.presentation_after unless div.presentation_after.nil?
+
+        puts p unless p.empty?
       end
     end
   end

data/lib/proiel/cli/converters/tiger.rb

@@ -1,154 +1,150 @@
-require 'builder'
-
-module PROIEL
-  module Converter
-    # Converter for the TigerXML format
-    # (http://www.ims.uni-stuttgart.de/projekte/TIGER/TIGERSearch/doc/html/TigerXML.html)
-    # in the variant used by VISL under the name 'TIGER dependency format'
-    # (http://beta.visl.sdu.dk/treebanks.html#TIGER_dependency_format).
-    class Tiger
-      SCHEMA_FILE = File.join('tigerxml', 'TigerXML.xsd')
-
-      MORPHOLOGICAL_FEATURES = %i(person_number tense_mood_voice case_number gender degree strength inflection)
-      OTHER_FEATURES = %i(lemma pos information_status antecedent_id word)
-
-      class << self
-        def process(tb, options)
-          selected_features = MORPHOLOGICAL_FEATURES + OTHER_FEATURES
-          @features = selected_features.map { |f| [f, 'FREC'] }.to_h
-
-          builder = Builder::XmlMarkup.new(target: STDOUT, indent: 2)
-          builder.instruct! :xml, version: '1.0', encoding: 'UTF-8'
-
-          tb.sources.each do |source|
-            @hack = tb.annotation_schema
-            write_source(builder, source) do
-              source.divs.each do |div|
-                div.sentences.each do |sentence|
-                  write_sentence(builder, sentence)
-                end
+module PROIEL::Converter
+  # Converter for the TigerXML format
+  # (http://www.ims.uni-stuttgart.de/projekte/TIGER/TIGERSearch/doc/html/TigerXML.html)
+  # in the variant used by VISL under the name 'TIGER dependency format'
+  # (http://beta.visl.sdu.dk/treebanks.html#TIGER_dependency_format).
+  class Tiger
+    SCHEMA_FILE = File.join('tigerxml', 'TigerXML.xsd')
+
+    MORPHOLOGICAL_FEATURES = %i(person_number tense_mood_voice case_number gender degree strength inflection)
+    OTHER_FEATURES = %i(lemma pos information_status antecedent_id word)
+
+    class << self
+      def process(tb, _)
+        selected_features = MORPHOLOGICAL_FEATURES + OTHER_FEATURES
+        @features = selected_features.map { |f| [f, 'FREC'] }.to_h
+
+        builder = Builder::XmlMarkup.new(target: STDOUT, indent: 2)
+        builder.instruct! :xml, version: '1.0', encoding: 'UTF-8'
+
+        tb.sources.each do |source|
+          @hack = tb.annotation_schema
+          write_source(builder, source) do
+            source.divs.each do |div|
+              div.sentences.each do |sentence|
+                write_sentence(builder, sentence)
               end
             end
           end
         end
+      end
 
-        def write_source(builder, s)
-          builder.corpus(id: s.id) do
-            builder.head do
-              builder.meta do
-                builder.name(s.title)
-              end
-
-              declare_annotation(builder, @features, @hack)
+      def write_source(builder, s)
+        builder.corpus(id: s.id) do
+          builder.head do
+            builder.meta do
+              builder.name(s.title)
             end
 
-            builder.body do
-              yield
-            end
+            declare_annotation(builder, @features, @hack)
+          end
+
+          builder.body do
+            yield
           end
         end
+      end
 
-        def declare_annotation(builder, features, annotation_schema)
-          builder.annotation do
-            features.each do |name, domain|
-              # FIXME: we may want to list possible values for some of these
-              builder.feature(name: name, domain: domain)
-            end
+      def declare_annotation(builder, features, annotation_schema)
+        builder.annotation do
+          features.each do |name, domain|
+            # FIXME: we may want to list possible values for some of these
+            builder.feature(name: name, domain: domain)
+          end
 
-            builder.edgelabel do
-              builder.value(name: '--')
+          builder.edgelabel do
+            builder.value(name: '--')
 
-              annotation_schema.primary_relations.each do |tag, features|
-                builder.value({ name: tag }, features.summary)
-              end
+            annotation_schema.primary_relations.each do |tag, features|
+              builder.value({ name: tag }, features.summary)
             end
+          end
 
-            builder.secedgelabel do
-              annotation_schema.secondary_relations.each do |tag, features|
-                builder.value({name: tag }, features.summary)
-              end
+          builder.secedgelabel do
+            annotation_schema.secondary_relations.each do |tag, features|
+              builder.value({name: tag }, features.summary)
             end
           end
         end
+      end
 
-        def token_attrs(s, t, type)
-          attrs = {}
-
-          @features.each do |name, domain|
-            if domain == 'FREC' or domain == type
-              case name
-              when :word, :cat
-                attrs[name] = t.pro? ? "PRO-#{t.relation.upcase}" : t.form
-              when *@semantic_features
-                attrs[name] = t.sem_tags_to_hash[attr]
-              when :lemma
-                attrs[name] = t.lemma
-              when :pos
-                if t.empty_token_sort
-                  attrs[name] = t.empty_token_sort + "-"
-                else
-                  attrs[name] = t.pos
-                end
-              when *MORPHOLOGICAL_FEATURES
-                attrs[name] = name.to_s.split("_").map { |a| t.morphology_hash[a.to_sym] || '-' }.join
+      def token_attrs(t, type)
+        attrs = {}
+
+        @features.each do |name, domain|
+          if domain == 'FREC' or domain == type
+            case name
+            when :word, :cat
+              attrs[name] = t.pro? ? "PRO-#{t.relation.upcase}" : t.form
+            when *@semantic_features
+              attrs[name] = t.sem_tags_to_hash[attr]
+            when :lemma
+              attrs[name] = t.lemma
+            when :pos
+              if t.empty_token_sort
+                attrs[name] = t.empty_token_sort + '-'
+              else
+                attrs[name] = t.pos
+              end
+            when *MORPHOLOGICAL_FEATURES
+              attrs[name] = name.to_s.split('_').map { |a| t.morphology_hash[a.to_sym] || '-' }.join
+            else
+              if t.respond_to?(name)
+                attrs[name] = t.send(name)
               else
-                if t.respond_to?(name)
-                  attrs[name] = t.send(name)
-                else
-                  raise "Do not know how to get required attribute #{name}"
-                end
+                raise "Do not know how to get required attribute #{name}"
               end
-              attrs[name] ||= "--"
             end
+            attrs[name] ||= '--'
           end
-
-          attrs
         end
 
-        def write_terminals(builder, s)
-          builder.terminals do
-            s.tokens.each do |t|
-              builder.t(token_attrs(s, t, 'T').merge({ id: "w#{t.id}"}))
-            end
+        attrs
+      end
+
+      def write_terminals(builder, s)
+        builder.terminals do
+          s.tokens.each do |t|
+            builder.t(token_attrs(t, 'T').merge({ id: "w#{t.id}"}))
           end
         end
+      end
 
-        def write_nonterminals(builder, s)
-          builder.nonterminals do
-            # Add an empty root node
-            h = @features.select { |_, domain| ['FREC', 'NT'].include?(domain) }.map { |name, _| [name, '--'] }.to_h
-            h[:id] = "s#{s.id}_root"
+      def write_nonterminals(builder, s)
+        builder.nonterminals do
+          # Add an empty root node
+          h = @features.select { |_, domain| ['FREC', 'NT'].include?(domain) }.map { |name, _| [name, '--'] }.to_h
+          h[:id] = "s#{s.id}_root"
 
-            builder.nt(h) do
-              s.tokens.reject { |t| t.head or t.pro? }.each do |t|
-                builder.edge(idref: "p#{t.id}", label: t.relation)
-              end
+          builder.nt(h) do
+            s.tokens.reject { |t| t.head or t.pro? }.each do |t|
+              builder.edge(idref: "p#{t.id}", label: t.relation)
             end
+          end
 
-            # Add other NTs
-            s.tokens.each do |t|
-              builder.nt(token_attrs(s, t, 'NT').merge(id: "p#{t.id}")) do
-                # Add an edge to the correspoding terminal node
-                builder.edge(idref: "w#{t.id}", label: '--')
+          # Add other NTs
+          s.tokens.each do |t|
+            builder.nt(token_attrs(t, 'NT').merge(id: "p#{t.id}")) do
+              # Add an edge to the correspoding terminal node
+              builder.edge(idref: "w#{t.id}", label: '--')
 
-                # Add primary dependency edges
-                t.children.each { |d| builder.edge(idref: "p#{d.id}", label: d.relation) }
+              # Add primary dependency edges
+              t.children.each { |d| builder.edge(idref: "p#{d.id}", label: d.relation) }
 
-                # Add secondary dependency edges
-                t.slashes.each do |relation, target_id|
-                  builder.secedge(idref: "p#{target_id}", label: relation)
-                end
+              # Add secondary dependency edges
+              t.slashes.each do |relation, target_id|
+                builder.secedge(idref: "p#{target_id}", label: relation)
               end
             end
           end
         end
+      end
 
-        def write_sentence(builder, s)
-          builder.s(id: "s#{s.id}") do
-            builder.graph(root: "s#{s.id}_root") do
-              write_terminals(builder, s)
-              write_nonterminals(builder, s)
-            end
+      def write_sentence(builder, s)
+        builder.s(id: "s#{s.id}") do
+          builder.graph(root: "s#{s.id}_root") do
+            write_terminals(builder, s)
+            write_nonterminals(builder, s)
           end
         end
       end