Method: HTML::Node.parse

Defined in:
lib/html/node.rb

.parse(parent, line, pos, content, strict = true) ⇒ Object 



139
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
192
193
194
195
196
197
198
199
200
201
202
 
# File 'lib/html/node.rb', line 139

def parse(parent, line, pos, content, strict=true)
  if content !~ /^<\S/
    Text.new(parent, line, pos, content)
  else
    scanner = StringScanner.new(content)

    unless scanner.skip(/</)
      if strict
        raise "expected <"
      else
        return Text.new(parent, line, pos, content)
      end
    end

    if scanner.skip(/!\[CDATA\[/)
      scanner.scan_until(/\]\]>/)
      return CDATA.new(parent, line, pos, scanner.pre_match)
    end
    
    closing = ( scanner.scan(/\//) ? :close : nil )
    return Text.new(parent, line, pos, content) unless name = scanner.scan(/[\w:]+/)
    name.downcase!
  
    unless closing
      scanner.skip(/\s*/)
      attributes = {}
      while attr = scanner.scan(/[-\w:]+/)
        value = true
        if scanner.scan(/\s*=\s*/)
          if delim = scanner.scan(/['"]/)
            value = ""
            while text = scanner.scan(/[^#{delim}\\]+|./)
              case text
                when "\\" then
                  value << text
                  value << scanner.getch
                when delim
                  break
                else value << text
              end
            end
          else
            value = scanner.scan(/[^\s>\/]+/)
          end
        end
        attributes[attr.downcase] = value
        scanner.skip(/\s*/)
      end
    
      closing = ( scanner.scan(/\//) ? :self : nil )
    end
    
    unless scanner.scan(/\s*>/)
      if strict
        raise "expected > (got #{scanner.rest.inspect} for #{content}, #{attributes.inspect})" 
      else
        # throw away all text until we find what we're looking for
        scanner.skip_until(/>/) or scanner.terminate
      end
    end

    Tag.new(parent, line, pos, name, attributes, closing)
  end
end