Class: Kwaff::Parser

Inherits:
Object
  • Object
show all
Defined in:
lib/kwaff/parser.rb

Overview

parse Kwaff string and create Kwaff::Document.

ex.

kwaff_str = File.open('file.kwaff') { |f| f.read }
parser = Kwaff::Parser.new(kwaff_str)
kwaff_doc = parser.parse_document()
translator = Kwaff::XmlTranslator.new()
xml_str = translator.translate(kwaff_doc)

if you want to get REXML::Document instead of Kwaff::Document, use Kwaff::Rexml::Parser instead of Kwaff::Parser.

Direct Known Subclasses

Rexml::Parser

Instance Attribute Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(str, toppings = {}) ⇒ Parser

Returns a new instance of Parser.



42
43
44
45
46
47
 
# File 'lib/kwaff/parser.rb', line 42

def initialize(str, toppings={})
   reset(str)
   @toppings = toppings
   @_s_cache = {}			# for spclen()
   @indent_width = toppings[:indent_width] || 2
end

Instance Attribute Details

#index ⇒ Object (readonly)

Returns the value of attribute index.



48
49
50
 
# File 'lib/kwaff/parser.rb', line 48

def index
  @index
end

#lines ⇒ Object (readonly)

Returns the value of attribute lines.



48
49
50
 
# File 'lib/kwaff/parser.rb', line 48

def lines
  @lines
end

#newline ⇒ Object (readonly)

Returns the value of attribute newline.



48
49
50
 
# File 'lib/kwaff/parser.rb', line 48

def newline
  @newline
end

Instance Method Details

#create_comment(str) ⇒ Object

factory method



94
95
96
 
# File 'lib/kwaff/parser.rb', line 94

def create_comment(str)
   return Comment.new(str)
end

#create_document(headers, root) ⇒ Object

factory method



84
85
86
 
# File 'lib/kwaff/parser.rb', line 84

def create_document(headers, root)
   return Document.new(headers, root)
end

#create_element(tag, children, attrs) ⇒ Object

factory method



89
90
91
 
# File 'lib/kwaff/parser.rb', line 89

def create_element(tag, children, attrs)
   return Element.new(tag, children, attrs)
end

#create_text(str, flag_escape = true) ⇒ Object

factory method



99
100
101
 
# File 'lib/kwaff/parser.rb', line 99

def create_text(str, flag_escape=true)
   return Text.new(str, flag_escape)
end

#parse_cdata(level = 0) ⇒ Object 



303
304
305
 
# File 'lib/kwaff/parser.rb', line 303

def parse_cdata(level=0)
   return _parse_textline(/^([ \t]*)\~ ?(.*)$/, false, level)	# *pattern*
end

#parse_comment(level = 0) ⇒ Object 



288
289
290
291
292
293
294
295
296
 
# File 'lib/kwaff/parser.rb', line 288

def parse_comment(level=0)
   line = _current_line()
   line =~ /^([ \t]*)[\#\!][ \t]*(.*)$/			# *pattern*
   Kwaff::assert() unless level == spclen($1)
   str = $2
   comment = create_comment(str)
   line = _read_line()
   return comment
end

#parse_document(str = nil) ⇒ Object 



109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
 
# File 'lib/kwaff/parser.rb', line 109

def parse_document(str=nil)
   reset(str) if str
   headers = OrderedHash.new
   line = _current_line()
   while line
      if line =~ /^\?[ \t]*([-:\w]+)[ \t]*[:=][ \t]*(.*)/	## *pattern*
         headers[$1.intern] = $2.strip
      elsif line =~ /^\s*$/
         # ignore empty line
      elsif line =~ /^\s*\/\//				## *pattern*
         # ignore comment line
      else
         break
      end
      line = _read_line()
   end
   if headers[:doctype] =~ /\A@[-:\w]+\z/
      nickname = headers[:doctype]
      doctype = DocType.get(nickname)
      unless doctype
         raise ParseError.new("doctype `#{headers[:doctype]}' not registered.", self)
      end
      headers[:doctype] = doctype
   end
   root_elem = parse_element()
   doc = create_document(headers, root_elem)
   set_node_option(doc, :empty_lines, @empty_lines)
   return doc
end

#parse_element(level = 0) ⇒ Object 



211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
 
# File 'lib/kwaff/parser.rb', line 211

def parse_element(level=0)
   ## tag and text
   line = _current_line()
   line =~ /^([ \t]*)\*[ \t]*([-:\w]+)([ \t]*=[ \t]*(.*))?/	## *pattern*
   Kwaff::assert() unless level == spclen($1)
   tag = $2 || ''
   value = $4
   tag.strip!
   if tag.empty?
      raise ParseError.new("tag is missing.", self)
   end
   flag_escape = true
   if value =~ /^<<<?('?\w+'?)$/
      terminator = $1
      if terminator =~ /\A'(.*)'\z/
         terminator = $1
         flag_escape = false
      end
      value = _parse_heredoc(terminator)
   end

   ## text
   text = value ? create_text(value, flag_escape) : nil

   ## attrs
   attrs = OrderedHash.new
   line = _read_line()
   while line
      if line =~ /^\s*\/\/$/	# line comment				## *pattern*
         # nothing
      #elsif line =~ /^\s*$/	# empty line
      #   @empty_lines += 1
      elsif line =~ /^\s*\-/	# attribute				## *pattern*
         unless line =~ /\A([ \t]*)\-[ \t]+([-:\w]+)[ \t]*=[ \t]*(.*)/	## *pattern*
            raise ParseError.new("invalid attribute format.", self)
         end
         attr_level = spclen($1)
         attr_name = $2
         attr_value = $3
         unless attr_level == level + @indent_width			## *indent*
            raise ParseError.new("attr indent width is not #{@indent_width}.", self)
         end
         if attr_value =~ /^<<([_A-Z]+)$/
            terminator = $1
            attr_value = _parse_heredoc(terminator)
         end
         attrs[attr_name] = attr_value
      else
         break
      end
      line = _read_line()
   end

   ## empty lines
   @empty_lines = 0
   while line && line =~ /\A\s*\z/
      @empty_lines += 1
      line = _read_line()
   end

   ## children
   if line && line =~ /^(\s*)[*\#!.\~]/ && spclen($1) > level	## *pattern*
      #children = parse_node_list(spclen($1))
      unless spclen($1) == level + @indent_width			## *indent*
         raise ParseError.new("indent width is not #{@indent_width}.", self)
      end
      children = parse_node_list(level+@indent_width)
   else
      children = []
   end
   children.unshift(text) if text

   ## element
   return create_element(tag, children, attrs)
end

#parse_node_list(level = 0) ⇒ Object 



140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
 
# File 'lib/kwaff/parser.rb', line 140

def parse_node_list(level=0)
   list = []
   str  = nil
   line = _current_line()
   while line
      if line =~ /^(\s*)([*\#!.\~])/	# node		## *pattern*
         curr_level = spclen($1)
         ch = $2
         if str
            list << create_text(str)
            str = nil
         end
         if curr_level == level		# sibling node
            empty_lines = @empty_lines || 0
            @empty_lines = 0
            case ch
            when '*'					## *pattern*
               node = parse_element(level)
            when '#', '!'					## *pattern*
               node = parse_comment(level)
            when '.'					## *pattern*
               node = parse_text(level)
            when '~'					## *pattern*
               node = parse_cdata(level)
            else
               raise ParseError.new("internal error", self)
            end
            set_node_option(node, :empty_lines, empty_lines)
            list << node
            line = _current_line()
            next
         elsif curr_level < level		# parent node
            break
         else				# child node
            raise ParseError.new("illegal indentation.", self)	# or internal error
         end
      elsif line =~ /^\s*$/		# empty lines
         @empty_lines ||= 0
         @empty_lines += 1
         line = _read_line()
      elsif line =~ /^\s*\/\//		# line-comment	# *pattern*
         line = _read_line()
      else
         #raise ParseError.new("invalid line.", self)
         str ||= ''
         str << line
         line = _read_line()
      end
   end
   list << create_text(str) if str
   return list
end

#parse_text(level = 0) ⇒ Object 



299
300
301
 
# File 'lib/kwaff/parser.rb', line 299

def parse_text(level=0)
   return _parse_textline(/^([ \t]*)\. ?(.*)$/, true, level)	# *pattern*
end

#reset(str) ⇒ Object 



51
52
53
54
55
56
57
58
59
 
# File 'lib/kwaff/parser.rb', line 51

def reset(str)
   @str = str
   idx = str.index(?\n)
   @newline  = (idx && str[idx-1] == ?\r) ? "\r\n" : "\n"
   @lines    = str.split(@newline, -1)
   @lines.pop()		# remove the last line
   @index    = 0
   @limit    = @lines.length
end

#set_node_option(node, name, value) ⇒ Object 



104
105
106
 
# File 'lib/kwaff/parser.rb', line 104

def set_node_option(node, name, value)
   node.empty_lines = value if name == :empty_lines
end

#spclen(str) ⇒ Object

space length, with tab expanding



76
77
78
79
80
 
# File 'lib/kwaff/parser.rb', line 76

def spclen(str)
   return 0 unless str
   len = (@_s_cache[str] ||= (str.gsub(/([^\t]{8})|([^\t]*)\t/n){[$+].pack("A8")}).length)
   return len
end