Generated on Fri Jul 11 15:55:29 -0700 2008 with rcov 0.7.0
Code reported as executed by Ruby looks like this... and this: this line is also marked as covered. Lines considered as run by rcov, but not reported by Ruby, look like this, and this: these lines were inferred by rcov (using simple heuristics). Finally, here's a line marked as not executed.
| Name | Total lines | Lines of code | Total coverage | Code coverage | ||||||||
| lib/openid/yadis/htmltokenizer.rb | 305 | 175 |
|
|
1 # = HTMLTokenizer 2 # 3 # Author:: Ben Giddings (mailto:bg-rubyforge@infofiend.com) 4 # Copyright:: Copyright (c) 2004 Ben Giddings 5 # License:: Distributes under the same terms as Ruby 6 # 7 # 8 # This is a partial port of the functionality behind Perl's TokeParser 9 # Provided a page it progressively returns tokens from that page 10 # 11 # $Id: htmltokenizer.rb,v 1.7 2005/06/07 21:05:53 merc Exp $ 12 13 # 14 # A class to tokenize HTML. 15 # 16 # Example: 17 # 18 # page = "<HTML> 19 # <HEAD> 20 # <TITLE>This is the title</TITLE> 21 # </HEAD> 22 # <!-- Here comes the <a href=\"missing.link\">blah</a> 23 # comment body 24 # --> 25 # <BODY> 26 # <H1>This is the header</H1> 27 # <P> 28 # This is the paragraph, it contains 29 # <a href=\"link.html\">links</a>, 30 # <img src=\"blah.gif\" optional alt='images 31 # are 32 # really cool'>. Ok, here is some more text and 33 # <A href=\"http://another.link.com/\" target=\"_blank\">another link</A>. 34 # </P> 35 # </body> 36 # </HTML> 37 # " 38 # toke = HTMLTokenizer.new(page) 39 # 40 # assert("<h1>" == toke.getTag("h1", "h2", "h3").to_s.downcase) 41 # assert(HTMLTag.new("<a href=\"link.html\">") == toke.getTag("IMG", "A")) 42 # assert("links" == toke.getTrimmedText) 43 # assert(toke.getTag("IMG", "A").attr_hash['optional']) 44 # assert("_blank" == toke.getTag("IMG", "A").attr_hash['target']) 45 # 46 class HTMLTokenizer 47 @@version = 1.0 48 49 # Get version of HTMLTokenizer lib 50 def self.version 51 @@version 52 end 53 54 attr_reader :page 55 56 # Create a new tokenizer, based on the content, used as a string. 57 def initialize(content) 58 @page = content.to_s 59 @cur_pos = 0 60 end 61 62 # Reset the parser, setting the current position back at the stop 63 def reset 64 @cur_pos = 0 65 end 66 67 # Look at the next token, but don't actually grab it 68 def peekNextToken 69 if @cur_pos == @page.length then return nil end 70 71 if ?< == @page[@cur_pos] 72 # Next token is a tag of some kind 73 if '!--' == @page[(@cur_pos + 1), 3] 74 # Token is a comment 75 tag_end = @page.index('-->', (@cur_pos + 1)) 76 if tag_end.nil? 77 raise HTMLTokenizerError, "No end found to started comment:\n#{@page[@cur_pos,80]}" 78 end 79 # p @page[@cur_pos .. (tag_end+2)] 80 HTMLComment.new(@page[@cur_pos .. (tag_end + 2)]) 81 else 82 # Token is a html tag 83 tag_end = @page.index('>', (@cur_pos + 1)) 84 if tag_end.nil? 85 raise HTMLTokenizerError, "No end found to started tag:\n#{@page[@cur_pos,80]}" 86 end 87 # p @page[@cur_pos .. tag_end] 88 HTMLTag.new(@page[@cur_pos .. tag_end]) 89 end 90 else 91 # Next token is text 92 text_end = @page.index('<', @cur_pos) 93 text_end = text_end.nil? ? -1 : (text_end - 1) 94 # p @page[@cur_pos .. text_end] 95 HTMLText.new(@page[@cur_pos .. text_end]) 96 end 97 end 98 99 # Get the next token, returns an instance of 100 # * HTMLText 101 # * HTMLToken 102 # * HTMLTag 103 def getNextToken 104 token = peekNextToken 105 if token 106 # @page = @page[token.raw.length .. -1] 107 # @page.slice!(0, token.raw.length) 108 @cur_pos += token.raw.length 109 end 110 #p token 111 #print token.raw 112 return token 113 end 114 115 # Get a tag from the specified set of desired tags. 116 # For example: 117 # <tt>foo = toke.getTag("h1", "h2", "h3")</tt> 118 # Will return the next header tag encountered. 119 def getTag(*sought_tags) 120 sought_tags.collect! {|elm| elm.downcase} 121 122 while (tag = getNextToken) 123 if tag.kind_of?(HTMLTag) and 124 (0 == sought_tags.length or sought_tags.include?(tag.tag_name)) 125 break 126 end 127 end 128 tag 129 end 130 131 # Get all the text between the current position and the next tag 132 # (if specified) or a specific later tag 133 def getText(until_tag = nil) 134 if until_tag.nil? 135 if ?< == @page[@cur_pos] 136 # Next token is a tag, not text 137 "" 138 else 139 # Next token is text 140 getNextToken.text 141 end 142 else 143 ret_str = "" 144 145 while (tag = peekNextToken) 146 if tag.kind_of?(HTMLTag) and tag.tag_name == until_tag 147 break 148 end 149 150 if ("" != tag.text) 151 ret_str << (tag.text + " ") 152 end 153 getNextToken 154 end 155 156 ret_str 157 end 158 end 159 160 # Like getText, but squeeze all whitespace, getting rid of 161 # leading and trailing whitespace, and squeezing multiple 162 # spaces into a single space. 163 def getTrimmedText(until_tag = nil) 164 getText(until_tag).strip.gsub(/\s+/m, " ") 165 end 166 167 end 168 169 class HTMLTokenizerError < Exception 170 end 171 172 # The parent class for all three types of HTML tokens 173 class HTMLToken 174 attr_accessor :raw 175 176 # Initialize the token based on the raw text 177 def initialize(text) 178 @raw = text 179 end 180 181 # By default, return exactly the string used to create the text 182 def to_s 183 raw 184 end 185 186 # By default tokens have no text representation 187 def text 188 "" 189 end 190 191 def trimmed_text 192 text.strip.gsub(/\s+/m, " ") 193 end 194 195 # Compare to another based on the raw source 196 def ==(other) 197 raw == other.to_s 198 end 199 end 200 201 # Class representing text that isn't inside a tag 202 class HTMLText < HTMLToken 203 def text 204 raw 205 end 206 end 207 208 # Class representing an HTML comment 209 class HTMLComment < HTMLToken 210 attr_accessor :contents 211 def initialize(text) 212 super(text) 213 temp_arr = text.scan(/^<!--\s*(.*?)\s*-->$/m) 214 if temp_arr[0].nil? 215 raise HTMLTokenizerError, "Text passed to HTMLComment.initialize is not a comment" 216 end 217 218 @contents = temp_arr[0][0] 219 end 220 end 221 222 # Class representing an HTML tag 223 class HTMLTag < HTMLToken 224 attr_reader :end_tag, :tag_name 225 def initialize(text) 226 super(text) 227 if ?< != text[0] or ?> != text[-1] 228 raise HTMLTokenizerError, "Text passed to HTMLComment.initialize is not a comment" 229 end 230 231 @attr_hash = Hash.new 232 @raw = text 233 234 tag_name = text.scan(/[\w:-]+/)[0] 235 if tag_name.nil? 236 raise HTMLTokenizerError, "Error, tag is nil: #{tag_name}" 237 end 238 239 if ?/ == text[1] 240 # It's an end tag 241 @end_tag = true 242 @tag_name = '/' + tag_name.downcase 243 else 244 @end_tag = false 245 @tag_name = tag_name.downcase 246 end 247 248 @hashed = false 249 end 250 251 # Retrieve a hash of all the tag's attributes. 252 # Lazily done, so that if you don't look at a tag's attributes 253 # things go quicker 254 def attr_hash 255 # Lazy initialize == don't build the hash until it's needed 256 if !@hashed 257 if !@end_tag 258 # Get the attributes 259 attr_arr = @raw.scan(/<[\w:-]+\s+(.*?)\/?>/m)[0] 260 if attr_arr.kind_of?(Array) 261 # Attributes found, parse them 262 attrs = attr_arr[0] 263 attr_arr = attrs.scan(/\s*([\w:-]+)(?:\s*=\s*("[^"]*"|'[^']*'|([^"'>][^\s>]*)))?/m) 264 # clean up the array by: 265 # * setting all nil elements to true 266 # * removing enclosing quotes 267 attr_arr.each { 268 |item| 269 val = if item[1].nil? 270 item[0] 271 elsif '"'[0] == item[1][0] or '\''[0] == item[1][0] 272 item[1][1 .. -2] 273 else 274 item[1] 275 end 276 @attr_hash[item[0].downcase] = val 277 } 278 end 279 end 280 @hashed = true 281 end 282 283 #p self 284 285 @attr_hash 286 end 287 288 # Get the 'alt' text for a tag, if it exists, or an empty string otherwise 289 def text 290 if !end_tag 291 case tag_name 292 when 'img' 293 if !attr_hash['alt'].nil? 294 return attr_hash['alt'] 295 end 296 when 'applet' 297 if !attr_hash['alt'].nil? 298 return attr_hash['alt'] 299 end 300 end 301 end 302 return '' 303 end 304 end 305
Generated using the rcov code coverage analysis tool for Ruby version 0.7.0.