rsyntaxtree 0.8.8 → 1.0.1

data/lib/rsyntaxtree/string_parser.rb

@@ -7,241 +7,234 @@
 #
 # Parses a phrase into leafs and nodes and store the result in an element list
 # (see element_list.rb)
-#
-# This file is part of RSyntaxTree, which is a ruby port of Andre Eisenbach's
-# excellent program phpSyntaxTree.
-#
 # Copyright (c) 2007-2021 Yoichiro Hasebe <yohasebe@gmail.com>
-# Copyright (c) 2003-2004 Andre Eisenbach <andre@ironcreek.net>
 
 require 'elementlist'
 require 'element'
-
-# def escape_high_ascii(string)
-#   html = ""
-#   string.length.times do |i|
-#     ch = string[i]
-#     if(ch < 127)
-#       html += ch.chr
-#     else
-#       html += sprintf("&#%d;", ch)
-#     end
-#   end  
-#   html
-# end
-
-class StringParser
-
-  attr_accessor :data, :elist, :pos, :id, :level, :tncnt
-  def initialize(str)
-    # Clean up the data a little to make processing easier
-    string = str.gsub(/\t/, "") rescue ""
-    string.gsub!(/\s+/, " ")
-    string.gsub!(/\] \[/, "][")
-    string.gsub!(/ \[/, "[")
-
-    @data = string # Store it for later...
-    @elist = ElementList.new # Initialize internal element list 
-    @pos = 0 # Position in the sentence
-    @id = 1 # ID for the next element
-    @level = 0 # Level in the diagram
-    @tncnt = Hash.new # Node type counts
-  end
-
-  # caution: quick and dirty solution      
-  def valid?
-    if(@data.length < 1)
-      return false
-    end
-
-    if /\[\s*\]/m =~ @data
-      return false  
-    end
-
-    if /\[\_/ =~ @data
-      return false
-    end
-
-    text = @data.strip
-    text_r = text.split(//)
-    open_br, close_br = [], []
-    escape = false
-    text_r.each do |chr|
-      if chr == "\\"
-        escape = true
-      elsif chr == '[' && !escape
-        open_br.push(chr)
-      elsif chr == ']' && !escape
-        close_br.push(chr)
-        if open_br.length < close_br.length
-          break
+require 'utils'
+
+module RSyntaxTree
+  class StringParser
+
+    attr_accessor :data, :elist, :pos, :id, :level
+    def initialize(str, fontset, fontsize)
+      # Clean up the data a little to make processing easier
+      # repeated newlines => a newline
+      string = str.gsub(/[\n\r]+/m, "\n")
+      # a backslash followed by a newline => a backslash followed by an 'n'
+      string.gsub!(/\\\n\s*/m, "\\n")
+      # repeated whitespace characters => " "
+      string.gsub!(/\s+/, " ")
+      string.gsub!(/\]\s+\[/, "][")
+      string.gsub!(/\s+\[/, "[")
+      string.gsub!(/\[\s+/, "[")
+      string.gsub!(/\s+\]/, "]")
+      string.gsub!(/\]\s+/, "]")
+      string.gsub!(/<(\d*)>/) do
+        num_padding = $1.to_i
+        if num_padding > 0
+          result = WHITESPACE_BLOCK * num_padding
+        else
+          result = WHITESPACE_BLOCK
         end
-      elsif escape
-        escape = false
+        result
       end
-    end
-
-    return false unless open_br.length == close_br.length
-    # make_tree(0)
-    # return false if @tncnt.empty?
-    # @tncnt.each do |key, value|
-    #   return false if key == ""
-    # end
-    return true
-  end 
-
-  def parse
-    make_tree(0);
-  end
 
-  def get_elementlist
-    @elist;
-  end
+      @data = string # Store it for later...
+      if @data.contains_cjk?
+        fontset[:normal] = fontset[:cjk]
+      end
+      @elist = ElementList.new # Initialize internal element list
+      @pos = 0 # Position in the sentence
+      @id = 1 # ID for the next element
+      @level = 0 # Level in the diagram
+      @fontset = fontset
+      @fontsize = fontsize
+    end
 
-  def auto_subscript
-    elements = @elist.get_elements
-    tmpcnt   = Hash.new
-    elements.each do |element|
-      if(element.type == ETYPE_NODE)
-        count = 1
-        content = element.content
+    def self.valid?(data)
+      if(data.length < 1)
+        raise RSTError, "Error: input text is empty"
+      end
 
-        if @tncnt[content]
-          count = @tncnt[content]
-        end
+      if /\[\s*\]/m =~ data
+        raise RSTError, "Error: inside the brackets is empty"
+      end
 
-        if(count > 1)
-          if tmpcnt[content]
-            tmpcnt[content] += 1
+      text = data.strip
+      text_r = text.split(//)
+      open_br, close_br = [], []
+      escape = false
+      text_r.each do |chr|
+        if chr == "\\"
+          if escape
+            escape = false
           else
-            tmpcnt[content] = 1
+            escape = true
           end
+          next
+        end
 
-          element.content += ("_" + tmpcnt[content].to_s)
+        if escape && /[\[\]]/ =~ chr
+          escape = false
+          next
+        elsif chr == '['
+          open_br.push(chr)
+        elsif chr == ']'
+          close_br.push(chr)
+          if open_br.length < close_br.length
+            break
+          end
         end
+        escape = false
+      end
 
+      if open_br.empty? && close_br.empty?
+        raise RSTError, "Error: input text does not contain paired brackets"
+      elsif open_br.length == close_br.length
+        return true
+      else
+        raise RSTError, "Error: open and close brackets do not match"
       end
-    end  
-    @tncnt
-  end
- 
-  def count_node(name)
-    name = name.strip
-    if @tncnt[name]
-      @tncnt[name] += 1
-    else
-      @tncnt[name] = 1
     end
-  end
 
-  def get_next_token
-    data = @data.split(//)
-    gottoken = false
-    token = ""
-    i = 0
+    def parse
+      make_tree(0);
+      @elist.set_hierarchy
+    end
 
-    if((@pos + 1) >= data.length)
-      return ""
+    def get_elementlist
+      @elist;
     end
 
-    escape = false
-    while(((@pos + i) < data.length) && !gottoken)
-      ch = data[@pos + i];
-      case ch
-      when "["
-        if escape
-          token += ch
-          escape = false
-        else
-          if(i > 0)
+    def get_next_token
+      data = @data.split(//)
+      gottoken = false
+      token = ""
+      i = 0
+
+      if((@pos + 1) >= data.length)
+        return ""
+      end
+
+      escape = false
+      while(((@pos + i) < data.length) && !gottoken)
+        ch = data[@pos + i];
+        case ch
+        when "["
+          if escape
+            token += ch
+            escape = false
+          else
+            if(i > 0)
+              gottoken = true
+            else
+              token += ch
+            end
+          end
+        when "]"
+          if escape
+            token += ch
+            escape = false
+          else
+            if(i == 0 )
+              token += ch
+            end
             gottoken = true
+          end
+        when "\\"
+          if escape
+            token += '\\\\'
+            escape = false
+          else
+            escape = true
+          end
+        when " "
+          if escape
+            token += '\\n'
+            escape = false
+          else
+            token += ch
+          end
+        when /[n{}<>^+*_=~\|\-]/
+          if escape
+            token += '\\' + ch
+            escape = false
           else
             token += ch
           end
-        end
-      when "]"
-        if escape
-          token += ch
-          escape = false
         else
-          if(i == 0 )
+          if escape
+            token += ch
+            escape = false
+          else
             token += ch
           end
-          gottoken = true
         end
-      when "\\"
-        escape = true
-      when /[\n\r]/
-        gottoken = false # same as do nothing  
-      else
-        token += ch
-        escape = false if escape
+        i += 1
       end
-      i += 1
-    end
 
-    if(i > 1)
-      @pos += (i - 1)
-    else
-      @pos += 1
+      if(i > 1)
+        @pos += (i - 1)
+      else
+        @pos += 1
+      end
+      return token
     end
-    return token
-  end
 
-  def make_tree(parent)
-    token = get_next_token.strip
-    parts = Array.new
-
-    while(token != "" && token != "]" )
-      token_r = token.split(//)
-      case token_r[0]
-      when "["
-        tl = token_r.length
-        token_r = token_r[1, tl - 1]
-        spaceat = token_r.index(" ")
-        newparent  = -1
-
-        if spaceat
-          parts[0] = token_r[0, spaceat].join
-          parts[0] = parts[0].gsub("<>", " ")
-
-          tl =token_r.length
-          parts[1] = token_r[spaceat, tl - spaceat].join
-          parts[1] = parts[1].gsub("<>", " ")
-
-          element = Element.new(@id, parent, parts[0], @level)
-          @id += 1
-          @elist.add(element)
-          newparent = element.id
-          count_node(parts[0])
-
-          element = Element.new(@id, @id - 1, parts[1], @level + 1 )
-          @id += 1          
-          @elist.add(element)
-        else
-          joined = token_r.join.gsub("<>", " ")
-          element = Element.new(@id, parent, joined, @level)
-          @id += 1          
-          newparent = element.id
-          @elist.add(element)
-          count_node(joined)
-        end 
+    def make_tree(parent)
+      token = get_next_token.strip
+      parts = Array.new
+
+      while(token != "" && token != "]" )
+        token_r = token.split(//)
+        case token_r[0]
+        when "["
+          tl = token_r.length
+          token_r = token_r[1, tl - 1]
+          spaceat = token_r.index(" ")
+          newparent  = -1
+
+          if spaceat
+            parts[0] = token_r[0, spaceat].join
+            # parts[0] = parts[0].gsub("\\>", "﹥").gsub("\\<", "﹤")
+
+            tl =token_r.length
+            parts[1] = token_r[spaceat, tl - spaceat].join
+            # parts[1] = parts[1].gsub("\\>", "﹥").gsub("\\<", "﹤")
+
+            element = Element.new(@id, parent, parts[0], @level, @fontset, @fontsize)
+            @id += 1
+            @elist.add(element)
+            newparent = element.id
+
+            element = Element.new(@id, @id - 1, parts[1], @level + 1, @fontset, @fontsize)
+            @id += 1
+            @elist.add(element)
+          else
+            # joined = token_r.join.gsub("\\>", "﹥").gsub("\\<", "﹤")
+            joined = token_r.join
+            element = Element.new(@id, parent, joined, @level, @fontset,  @fontsize)
+            @id += 1
+            newparent = element.id
+            @elist.add(element)
+          end
 
-        @level += 1
-        make_tree(newparent)
+          @level += 1
+          make_tree(newparent)
 
-      else
-        if token.strip != ""
-          element = Element.new(@id, parent, token, @level)
-          @id += 1          
-          @elist.add(element)
-          count_node(token)
+        else
+          if token.strip != ""
+            element = Element.new(@id, parent, token, @level, @fontset, @fontsize)
+            @id += 1
+            @elist.add(element)
+          end
         end
-      end
 
-      token = get_next_token
+        token = get_next_token
+      end
+      @level -= 1
     end
-    @level -= 1
   end
 end