htmltools 1.10
Sign up to get free protection for your applications and to get access to all the features.
- data/INSTALL +58 -0
- data/README +162 -0
- data/demo/degolive.rb +89 -0
- data/demo/ebaySearch.rb +93 -0
- data/demo/xpath.rb +62 -0
- data/lib/html/element.rb +323 -0
- data/lib/html/rexml-nodepath.rb +49 -0
- data/lib/html/sgml-parser.rb +372 -0
- data/lib/html/stparser.rb +280 -0
- data/lib/html/tags.rb +288 -0
- data/lib/html/tree.rb +140 -0
- data/lib/html/xmltree.rb +173 -0
- data/lib/html/xpath.rb +72 -0
- data/test/suite.rb +5 -0
- data/test/tc_html-element.rb +73 -0
- data/test/tc_html-tree.rb +201 -0
- data/test/tc_source-parser.rb +160 -0
- data/test/tc_stacking-parser.rb +196 -0
- data/test/tc_xpath.rb +87 -0
- metadata +58 -0
data/lib/html/element.rb
ADDED
@@ -0,0 +1,323 @@
|
|
1
|
+
# This module is a mix-in that provides parent/child behavior to real
|
2
|
+
# Element classes. Because it defines <tt>each()</tt> and includes Enumerable,
|
3
|
+
# you can iterate through a tree using the usual Enumerable methods.
|
4
|
+
#
|
5
|
+
# Copyright:: Copyright (C) 2003, Johannes Brodwall <johannes@brodwall.com>,
|
6
|
+
# Copyright (C) 2002, Ned Konz <ned@bike-nomad.com>
|
7
|
+
# License:: Same as Ruby's
|
8
|
+
# CVS ID: $Id: element.rb,v 1.4 2004/09/24 23:28:55 jhannes Exp $
|
9
|
+
|
10
|
+
require 'html/tags'
|
11
|
+
|
12
|
+
module HTMLTree
|
13
|
+
module TreeElement
|
14
|
+
include Enumerable
|
15
|
+
|
16
|
+
protected
|
17
|
+
|
18
|
+
def initialize_tree_element(parent_or_nil = nil, contents_or_nil = nil)
|
19
|
+
@_content, @_parent = contents_or_nil, parent_or_nil
|
20
|
+
if parent_or_nil
|
21
|
+
parent_or_nil.add_child(self)
|
22
|
+
end
|
23
|
+
end
|
24
|
+
|
25
|
+
attr_accessor :_parent
|
26
|
+
|
27
|
+
public
|
28
|
+
|
29
|
+
# Add one or more children to this node.
|
30
|
+
def add_child(*children_to_add)
|
31
|
+
if can_have_children?
|
32
|
+
children_to_add.each do |child|
|
33
|
+
@_content << child
|
34
|
+
child._parent = self
|
35
|
+
end
|
36
|
+
else
|
37
|
+
raise(ArgumentError.exception('node cannot have children'))
|
38
|
+
end
|
39
|
+
end
|
40
|
+
|
41
|
+
alias_method :add_children, :add_child
|
42
|
+
|
43
|
+
# Remove one or more children from this node.
|
44
|
+
def remove_child(*children_to_remove)
|
45
|
+
if can_have_children?
|
46
|
+
children_to_remove.each do |child|
|
47
|
+
child._parent = nil if @_content.delete(child)
|
48
|
+
end
|
49
|
+
else
|
50
|
+
raise(ArgumentError.exception('node cannot have children'))
|
51
|
+
end
|
52
|
+
end
|
53
|
+
|
54
|
+
alias_method :remove_children, :remove_child
|
55
|
+
|
56
|
+
# Change my parent. Disconnects from prior parent, if any.
|
57
|
+
def parent=(parent_or_nil)
|
58
|
+
@_parent.remove_child(self) if @_parent
|
59
|
+
parent_or_nil.add_child(self) if parent_or_nil
|
60
|
+
end
|
61
|
+
|
62
|
+
# Return true if my content is a collection of Elements
|
63
|
+
# rather than actual data.
|
64
|
+
def can_have_children?
|
65
|
+
@_content.kind_of?(Array)
|
66
|
+
end
|
67
|
+
|
68
|
+
# Return a collection of my children. Returns an empty Array if I am a
|
69
|
+
# data element, just to keep other methods simple.
|
70
|
+
def children
|
71
|
+
can_have_children? ? @_content : []
|
72
|
+
end
|
73
|
+
|
74
|
+
# Return my content; either my children or my data.
|
75
|
+
def content
|
76
|
+
@_content
|
77
|
+
end
|
78
|
+
|
79
|
+
# Return my parent element.
|
80
|
+
def parent
|
81
|
+
@_parent
|
82
|
+
end
|
83
|
+
|
84
|
+
def path
|
85
|
+
"/"
|
86
|
+
end
|
87
|
+
|
88
|
+
# Return the ultimate parent.
|
89
|
+
def root
|
90
|
+
@_parent ? self : @_parent.root
|
91
|
+
end
|
92
|
+
|
93
|
+
# Return true if I have any children.
|
94
|
+
def has_children?
|
95
|
+
children.size > 0
|
96
|
+
end
|
97
|
+
|
98
|
+
# Breadth-first iterator, required by Enumerable.
|
99
|
+
def each(&block)
|
100
|
+
block.call(self)
|
101
|
+
children.each { |ch| ch.each(&block) }
|
102
|
+
end
|
103
|
+
|
104
|
+
# Print out to $stdout (or given IO or String)
|
105
|
+
# a formatted dump of my structure.
|
106
|
+
def dump(indent=0, io=$stdout)
|
107
|
+
io << " " * indent
|
108
|
+
io << self.to_s
|
109
|
+
io << "\n"
|
110
|
+
children.each { |ea| ea.dump(indent+1, io) }
|
111
|
+
end
|
112
|
+
|
113
|
+
end
|
114
|
+
|
115
|
+
# This is a Element that represents the whole document (and makes a
|
116
|
+
# scope for the DTD declaration)
|
117
|
+
class Document
|
118
|
+
include TreeElement
|
119
|
+
|
120
|
+
def initialize
|
121
|
+
initialize_tree_element(nil, [])
|
122
|
+
end
|
123
|
+
|
124
|
+
def to_s
|
125
|
+
''
|
126
|
+
end
|
127
|
+
|
128
|
+
def each(&block)
|
129
|
+
children.each { |ch| ch.each(&block) }
|
130
|
+
end
|
131
|
+
|
132
|
+
def write(io)
|
133
|
+
children.each { |t| t.write(io) }
|
134
|
+
end
|
135
|
+
|
136
|
+
def tag
|
137
|
+
''
|
138
|
+
end
|
139
|
+
|
140
|
+
# Return my child <html> node, if any.
|
141
|
+
def html_node
|
142
|
+
children.detect { |ea| ea.tag == 'html' }
|
143
|
+
end
|
144
|
+
end
|
145
|
+
|
146
|
+
# This is a TreeElement that represents tagged items in an HTML
|
147
|
+
# document.
|
148
|
+
class Element
|
149
|
+
include TreeElement
|
150
|
+
|
151
|
+
protected
|
152
|
+
|
153
|
+
# parent_or_nil:: TreeElement or nil
|
154
|
+
# tag_name:: String
|
155
|
+
def initialize(parent_or_nil = nil, tag_name = nil)
|
156
|
+
initialize_tree_element(parent_or_nil, [])
|
157
|
+
@_tag = tag_name
|
158
|
+
@_attributes = {}
|
159
|
+
@_attribute_order = []
|
160
|
+
end
|
161
|
+
|
162
|
+
public
|
163
|
+
|
164
|
+
def can_have_children?; true; end
|
165
|
+
|
166
|
+
# Return true if I'm data instead of a tag
|
167
|
+
def data?; false; end
|
168
|
+
|
169
|
+
def to_s
|
170
|
+
a = [ "<", tag ]
|
171
|
+
@_attribute_order.each { |k|
|
172
|
+
v = @_attributes[k]
|
173
|
+
a << " #{k.to_s}=\"#{v.to_s}\""
|
174
|
+
}
|
175
|
+
a << ">"
|
176
|
+
a.join('')
|
177
|
+
end
|
178
|
+
|
179
|
+
# Append an attribute. <tt>values</tt> are first flattened into an Array,
|
180
|
+
# then converted into strings.
|
181
|
+
#
|
182
|
+
# If there is a single attribute value, it will appear as a String,
|
183
|
+
# otherwise it will be an Array of Strings.
|
184
|
+
#
|
185
|
+
# Example:
|
186
|
+
# element.add_attribute("width", "123")
|
187
|
+
# element.add_attribute("value", [ "a", "b" ])
|
188
|
+
def add_attribute(name, *values)
|
189
|
+
values = values.flatten.collect { |ea| ea.to_s.strip }
|
190
|
+
name = name.downcase
|
191
|
+
if @_attributes.include?(name)
|
192
|
+
@_attributes[name] = @_attributes[name].to_a + values
|
193
|
+
else
|
194
|
+
@_attributes[name] = values.size > 1 ? values : values[0]
|
195
|
+
end
|
196
|
+
@_attribute_order << name
|
197
|
+
self
|
198
|
+
end
|
199
|
+
|
200
|
+
# Return my tag (should be a String)
|
201
|
+
def tag; @_tag; end
|
202
|
+
|
203
|
+
# Return an HTML::Tag for further information, or nil if this is an
|
204
|
+
# unknown tag.
|
205
|
+
def tag_info
|
206
|
+
begin
|
207
|
+
HTML::Tag.named(@_tag)
|
208
|
+
rescue NoSuchHTMLTagError
|
209
|
+
nil
|
210
|
+
end
|
211
|
+
end
|
212
|
+
|
213
|
+
# Return the path to this element from the root
|
214
|
+
def path
|
215
|
+
path = []
|
216
|
+
node = self
|
217
|
+
while node do
|
218
|
+
path.unshift node.tag
|
219
|
+
node = node.parent
|
220
|
+
end
|
221
|
+
path.join(".")
|
222
|
+
end
|
223
|
+
|
224
|
+
def show_structure(indent = 0)
|
225
|
+
puts(' ' * indent) + path
|
226
|
+
elements.each { |node| node.show_structure(indent + 2) }
|
227
|
+
nil
|
228
|
+
end
|
229
|
+
|
230
|
+
# Return the children of this node that are elements (not data)
|
231
|
+
def elements
|
232
|
+
children.select { |node| node.is_a? Element }
|
233
|
+
end
|
234
|
+
|
235
|
+
# Return my attributes Hash.
|
236
|
+
def attributes; @_attributes; end
|
237
|
+
|
238
|
+
# Return the order of my attributes
|
239
|
+
def attribute_order; @_attribute_order; end
|
240
|
+
|
241
|
+
# Return the value of a single attribute (a String or Array).
|
242
|
+
def attribute(name); @_attributes[name]; end
|
243
|
+
|
244
|
+
# Return the value of a single attribute (a String or Array).
|
245
|
+
def [](name); attribute(name); end
|
246
|
+
|
247
|
+
# Replace an attribute.
|
248
|
+
def []=(name, *values)
|
249
|
+
@_attributes[name] = values.size > 1 ? values : values[0]
|
250
|
+
@_attribute_order.delete(name)
|
251
|
+
self
|
252
|
+
end
|
253
|
+
|
254
|
+
# Print me (and my descendents) on the given IO stream.
|
255
|
+
def write(io)
|
256
|
+
io << self
|
257
|
+
children.each { |t| t.write(io) }
|
258
|
+
unless tag_info.is_empty_element
|
259
|
+
io.puts( "</#{tag()}>" )
|
260
|
+
end
|
261
|
+
end
|
262
|
+
|
263
|
+
end
|
264
|
+
|
265
|
+
# This is a TreeElement that represents leaf data nodes (CDATA, scripts,
|
266
|
+
# comments, processing directives). It forwards unknown messages to the
|
267
|
+
# content element, so it otherwise behaves like a String.
|
268
|
+
class Data
|
269
|
+
include TreeElement
|
270
|
+
|
271
|
+
protected
|
272
|
+
|
273
|
+
# parent_or_nil:: parent, TreeElement or nil
|
274
|
+
# str:: contents, String
|
275
|
+
def initialize(parent_or_nil = nil, str = '')
|
276
|
+
initialize_tree_element(parent_or_nil, str)
|
277
|
+
end
|
278
|
+
|
279
|
+
public
|
280
|
+
|
281
|
+
# Return true because I am a data Element.
|
282
|
+
def data?; true; end
|
283
|
+
|
284
|
+
# Return false because I have no children.
|
285
|
+
def can_have_children?; false; end
|
286
|
+
|
287
|
+
# Return an empty collection because I have no children.
|
288
|
+
def children; []; end
|
289
|
+
|
290
|
+
# Return my (empty) tag String.
|
291
|
+
def tag; ''; end
|
292
|
+
|
293
|
+
# Return my (empty) attributes Hash.
|
294
|
+
def attributes; {}; end
|
295
|
+
|
296
|
+
def to_s
|
297
|
+
@_content
|
298
|
+
end
|
299
|
+
|
300
|
+
# Print me on the given IO stream.
|
301
|
+
def write(io)
|
302
|
+
io << self
|
303
|
+
end
|
304
|
+
|
305
|
+
# Forward all other methods to my content, so I can otherwise behave
|
306
|
+
# like a String.
|
307
|
+
def method_missing(sym, *args)
|
308
|
+
@_content.method(sym).call(*args)
|
309
|
+
end
|
310
|
+
end
|
311
|
+
|
312
|
+
class Comment < Data
|
313
|
+
def to_s
|
314
|
+
'<!--' + @_content + '-->'
|
315
|
+
end
|
316
|
+
end
|
317
|
+
|
318
|
+
class Special < Data
|
319
|
+
def to_s
|
320
|
+
'<' + @_content + '>'
|
321
|
+
end
|
322
|
+
end
|
323
|
+
end
|
@@ -0,0 +1,49 @@
|
|
1
|
+
# This module adds method 'full_path' to REXML hierarchy
|
2
|
+
# This method will print out the path to the given node, for example
|
3
|
+
# /html/body/p[0]/@text
|
4
|
+
#
|
5
|
+
# Copyright:: Copyright (C) 2004, Johannes Brodwall <johannes@brodwall.com>
|
6
|
+
# License:: Same as Ruby's
|
7
|
+
# CVS ID: $Id: rexml-nodepath.rb,v 1.5 2005/05/25 17:38:45 jhannes Exp $
|
8
|
+
|
9
|
+
|
10
|
+
module REXML
|
11
|
+
class Child
|
12
|
+
def parent_path
|
13
|
+
parent ? parent.full_path : ''
|
14
|
+
end
|
15
|
+
end
|
16
|
+
|
17
|
+
class Document
|
18
|
+
def full_path
|
19
|
+
''
|
20
|
+
end
|
21
|
+
end
|
22
|
+
|
23
|
+
class Element
|
24
|
+
def child_index
|
25
|
+
return "" unless parent
|
26
|
+
siblings = parent.to_a.select do |node|
|
27
|
+
node.kind_of? Element and node.expanded_name == self.expanded_name
|
28
|
+
end
|
29
|
+
return "" if siblings.size < 2
|
30
|
+
"[" + (siblings.index(self)+1).to_s + "]"
|
31
|
+
end
|
32
|
+
|
33
|
+
def full_path
|
34
|
+
parent_path + '/' + expanded_name + child_index
|
35
|
+
end
|
36
|
+
end
|
37
|
+
|
38
|
+
class Text
|
39
|
+
def full_path
|
40
|
+
parent_path + '/text()' + " " + expanded_name
|
41
|
+
end
|
42
|
+
end
|
43
|
+
|
44
|
+
class Attribute
|
45
|
+
def full_path
|
46
|
+
element.full_path + '/@' + name
|
47
|
+
end
|
48
|
+
end
|
49
|
+
end
|
@@ -0,0 +1,372 @@
|
|
1
|
+
# A parser for SGML, using the derived class as static DTD.
|
2
|
+
#
|
3
|
+
# Taken from http://raa.ruby-lang.org/list.rhtml?name=html-parser-2
|
4
|
+
# This file seems to be included in the current install of Ruby,
|
5
|
+
# but with a bug related to attributes quoted with '"', so I have
|
6
|
+
# included in in the HTML package of this distribution
|
7
|
+
#
|
8
|
+
# Copyright:: Copyright (C) 2003, Johannes Brodwall <johannes@brodwall.com>,
|
9
|
+
# Copyright (C) 2002, Ned Konz <ned@bike-nomad.com>
|
10
|
+
# License:: Same as Ruby's
|
11
|
+
# CVS ID: $Id: sgml-parser.rb,v 1.4 2004/09/24 23:28:55 jhannes Exp $
|
12
|
+
|
13
|
+
|
14
|
+
module HTML
|
15
|
+
|
16
|
+
class SGMLParser
|
17
|
+
|
18
|
+
attr_reader :src_range
|
19
|
+
|
20
|
+
# Regular expressions used for parsing:
|
21
|
+
Interesting = /[&<]/
|
22
|
+
Incomplete = Regexp.compile('&([a-zA-Z][a-zA-Z0-9]*|#[0-9]*)?|' +
|
23
|
+
'<([a-zA-Z][^<>]*|/([a-zA-Z][^<>]*)?|' +
|
24
|
+
'![^<>]*)?')
|
25
|
+
|
26
|
+
Entityref = /&([a-zA-Z][-.a-zA-Z0-9]*)[^-.a-zA-Z0-9]/
|
27
|
+
Charref = /&#([0-9]+)[^0-9]/
|
28
|
+
|
29
|
+
Starttagopen = /<[>a-zA-Z]/
|
30
|
+
Endtagopen = /<\/[<>a-zA-Z]/
|
31
|
+
Endbracket = /[<>]/
|
32
|
+
Special = /<![^<>]*>/
|
33
|
+
Commentopen = /<!--/
|
34
|
+
Commentclose = /--[ \t\n]*>/
|
35
|
+
Tagfind = /[a-zA-Z][a-zA-Z0-9.-]*/
|
36
|
+
Attrfind = Regexp.compile('[\s,]*([a-zA-Z_][a-zA-Z_0-9.-]*)' +
|
37
|
+
'(\s*=\s*' +
|
38
|
+
"('[^']*'" +
|
39
|
+
'|"[^"]*"' +
|
40
|
+
'|[-~a-zA-Z0-9,./:+*%?!()_#=]*))?')
|
41
|
+
Endtagfind = /\s*\/\s*>/
|
42
|
+
Entitydefs =
|
43
|
+
{'lt'=>'<', 'gt'=>'>', 'amp'=>'&', 'quot'=>'"', 'apos'=>'\''}
|
44
|
+
|
45
|
+
def initialize(verbose=false)
|
46
|
+
@verbose = verbose
|
47
|
+
reset
|
48
|
+
end
|
49
|
+
|
50
|
+
def reset
|
51
|
+
@rawdata = ''
|
52
|
+
@stack = []
|
53
|
+
@lasttag = '???'
|
54
|
+
@nomoretags = false
|
55
|
+
@literal = false
|
56
|
+
@offset = 0
|
57
|
+
@ranges = []
|
58
|
+
end
|
59
|
+
|
60
|
+
def get_source(range)
|
61
|
+
start = range.first
|
62
|
+
end_index = range.end
|
63
|
+
exclusive = range.exclude_end?
|
64
|
+
offset_range = Range.new(start-@offset, end_index-@offset, exclusive)
|
65
|
+
return @rawdata[offset_range]
|
66
|
+
end
|
67
|
+
|
68
|
+
def set_range(start, end_index)
|
69
|
+
@src_range = Range.new(start+@offset, end_index+@offset, exclusive = true)
|
70
|
+
#puts "setting range #{@src_range}, text = \"#{get_source(src_range)}\""
|
71
|
+
end
|
72
|
+
|
73
|
+
def has_context(gi)
|
74
|
+
@stack.include? gi
|
75
|
+
end
|
76
|
+
|
77
|
+
def setnomoretags
|
78
|
+
@nomoretags = true
|
79
|
+
@literal = true
|
80
|
+
end
|
81
|
+
|
82
|
+
def setliteral(*args)
|
83
|
+
@literal = true
|
84
|
+
end
|
85
|
+
|
86
|
+
def feed(data)
|
87
|
+
@rawdata << data
|
88
|
+
goahead(false)
|
89
|
+
end
|
90
|
+
|
91
|
+
def close
|
92
|
+
goahead(true)
|
93
|
+
end
|
94
|
+
|
95
|
+
def handle_data_range(rawdata, start, end_index)
|
96
|
+
if end_index > start
|
97
|
+
set_range(start, end_index)
|
98
|
+
handle_data(rawdata[start...end_index])
|
99
|
+
end
|
100
|
+
return end_index
|
101
|
+
end
|
102
|
+
|
103
|
+
def goahead(_end)
|
104
|
+
rawdata = @rawdata
|
105
|
+
i = 0
|
106
|
+
n = rawdata.length
|
107
|
+
while i < n
|
108
|
+
if @nomoretags
|
109
|
+
i = handle_data_range(rawdata, i, n)
|
110
|
+
break
|
111
|
+
end
|
112
|
+
j = rawdata.index(Interesting, i)
|
113
|
+
j = n unless j
|
114
|
+
i = handle_data_range(rawdata, i, j)
|
115
|
+
break if (i == n)
|
116
|
+
if rawdata[i] == ?< #
|
117
|
+
if rawdata.index(Starttagopen, i) == i
|
118
|
+
if @literal
|
119
|
+
i = handle_data_range(rawdata, i, i+1)
|
120
|
+
next
|
121
|
+
end
|
122
|
+
k = parse_starttag(i)
|
123
|
+
break unless k
|
124
|
+
i = k
|
125
|
+
next
|
126
|
+
end
|
127
|
+
if rawdata.index(Endtagopen, i) == i
|
128
|
+
k = parse_endtag(i)
|
129
|
+
break unless k
|
130
|
+
i = k
|
131
|
+
@literal = false
|
132
|
+
next
|
133
|
+
end
|
134
|
+
if rawdata.index(Commentopen, i) == i
|
135
|
+
if @literal
|
136
|
+
i = handle_data_range(rawdata, i, i+1)
|
137
|
+
next
|
138
|
+
end
|
139
|
+
k = parse_comment(i)
|
140
|
+
break unless k
|
141
|
+
i += k
|
142
|
+
next
|
143
|
+
end
|
144
|
+
if rawdata.index(Special, i) == i
|
145
|
+
if @literal
|
146
|
+
i = handle_data_range(rawdata, i, i+1)
|
147
|
+
next
|
148
|
+
end
|
149
|
+
k = parse_special(i)
|
150
|
+
break unless k
|
151
|
+
i += k
|
152
|
+
next
|
153
|
+
end
|
154
|
+
elsif rawdata[i] == ?& #
|
155
|
+
if rawdata.index(Charref, i) == i
|
156
|
+
end_index = i + $&.length
|
157
|
+
end_index -= 1 unless rawdata[end_index-1] == ?;
|
158
|
+
set_range(i, end_index)
|
159
|
+
handle_charref($1)
|
160
|
+
i = end_index
|
161
|
+
next
|
162
|
+
end
|
163
|
+
if rawdata.index(Entityref, i) == i
|
164
|
+
end_index = i + $&.length
|
165
|
+
end_index -= 1 unless rawdata[end_index-1] == ?;
|
166
|
+
set_range(i, end_index)
|
167
|
+
handle_entityref($1)
|
168
|
+
i = end_index
|
169
|
+
next
|
170
|
+
end
|
171
|
+
else
|
172
|
+
raise RuntimeError, 'neither < nor & ??'
|
173
|
+
end
|
174
|
+
# We get here only if incomplete matches but
|
175
|
+
# nothing else
|
176
|
+
match = rawdata.index(Incomplete, i)
|
177
|
+
unless match == i
|
178
|
+
i = handle_data_range(rawdata, i, i+1)
|
179
|
+
next
|
180
|
+
end
|
181
|
+
j = match + $&.length
|
182
|
+
break if j == n # Really incomplete
|
183
|
+
i = handle_data_range(rawdata, i, j)
|
184
|
+
end
|
185
|
+
# end while
|
186
|
+
if _end and i < n
|
187
|
+
i = handle_data_range(rawdata, i, n)
|
188
|
+
end
|
189
|
+
@rawdata = rawdata[i..-1]
|
190
|
+
@offset += i
|
191
|
+
end
|
192
|
+
|
193
|
+
def parse_comment(i)
|
194
|
+
rawdata = @rawdata
|
195
|
+
if rawdata[i, 4] != '<!--'
|
196
|
+
raise RuntimeError, 'unexpected call to handle_comment'
|
197
|
+
end
|
198
|
+
match = rawdata.index(Commentclose, i)
|
199
|
+
return nil unless match
|
200
|
+
matched_length = $&.length
|
201
|
+
j = match
|
202
|
+
src_length = match + matched_length - i
|
203
|
+
set_range(i, i + src_length)
|
204
|
+
handle_comment(rawdata[i+4..(j-1)])
|
205
|
+
return src_length
|
206
|
+
end
|
207
|
+
|
208
|
+
def parse_starttag(i)
|
209
|
+
rawdata = @rawdata
|
210
|
+
j = rawdata.index(Endbracket, i + 1)
|
211
|
+
return nil unless j
|
212
|
+
attrs = []
|
213
|
+
if rawdata[i+1] == ?> #
|
214
|
+
# SGML shorthand: <> == <last open tag seen>
|
215
|
+
k = j
|
216
|
+
tag = @lasttag
|
217
|
+
else
|
218
|
+
match = rawdata.index(Tagfind, i + 1)
|
219
|
+
unless match
|
220
|
+
raise RuntimeError, 'unexpected call to parse_starttag'
|
221
|
+
end
|
222
|
+
k = i + 1 + ($&.length)
|
223
|
+
tag = $&.downcase
|
224
|
+
@lasttag = tag
|
225
|
+
end
|
226
|
+
while k < j
|
227
|
+
break if rawdata.index(Endtagfind, k)
|
228
|
+
break unless rawdata.index(Attrfind, k)
|
229
|
+
matched_length = $&.length
|
230
|
+
attrname, rest, attrvalue = $1, $2, $3
|
231
|
+
if not rest
|
232
|
+
attrvalue = '' # was: = attrname
|
233
|
+
elsif (attrvalue[0] == ?' && attrvalue[-1] == ?') or
|
234
|
+
(attrvalue[0] == ?" && attrvalue[-1] == ?")
|
235
|
+
attrvalue = attrvalue[1..-2]
|
236
|
+
end
|
237
|
+
attrs << [attrname.downcase, attrvalue]
|
238
|
+
k += matched_length
|
239
|
+
end
|
240
|
+
if rawdata[j] == ?> #
|
241
|
+
j += 1
|
242
|
+
end
|
243
|
+
set_range(i, j)
|
244
|
+
finish_starttag(tag, attrs)
|
245
|
+
return j
|
246
|
+
end
|
247
|
+
|
248
|
+
def parse_endtag(i)
|
249
|
+
rawdata = @rawdata
|
250
|
+
j = rawdata.index(Endbracket, i + 1)
|
251
|
+
return nil unless j
|
252
|
+
tag = (rawdata[i+2..j-1].strip).downcase
|
253
|
+
if rawdata[j] == ?> #
|
254
|
+
j += 1
|
255
|
+
end
|
256
|
+
set_range(i, j)
|
257
|
+
finish_endtag(tag)
|
258
|
+
return j
|
259
|
+
end
|
260
|
+
|
261
|
+
def finish_starttag(tag, attrs)
|
262
|
+
method = 'start_' + tag
|
263
|
+
if self.respond_to?(method)
|
264
|
+
@stack << tag
|
265
|
+
handle_starttag(tag, method, attrs)
|
266
|
+
return 1
|
267
|
+
else
|
268
|
+
method = 'do_' + tag
|
269
|
+
if self.respond_to?(method)
|
270
|
+
handle_starttag(tag, method, attrs)
|
271
|
+
return 0
|
272
|
+
else
|
273
|
+
unknown_starttag(tag, attrs)
|
274
|
+
return -1
|
275
|
+
end
|
276
|
+
end
|
277
|
+
end
|
278
|
+
|
279
|
+
def finish_endtag(tag)
|
280
|
+
if tag == ''
|
281
|
+
found = @stack.length - 1
|
282
|
+
if found < 0
|
283
|
+
unknown_endtag(tag)
|
284
|
+
return
|
285
|
+
end
|
286
|
+
else
|
287
|
+
unless @stack.include? tag
|
288
|
+
method = 'end_' + tag
|
289
|
+
unless self.respond_to?(method)
|
290
|
+
unknown_endtag(tag)
|
291
|
+
end
|
292
|
+
return
|
293
|
+
end
|
294
|
+
found = @stack.index(tag) #or @stack.length
|
295
|
+
end
|
296
|
+
while @stack.length > found
|
297
|
+
tag = @stack[-1]
|
298
|
+
method = 'end_' + tag
|
299
|
+
if respond_to?(method)
|
300
|
+
handle_endtag(tag, method)
|
301
|
+
else
|
302
|
+
unknown_endtag(tag)
|
303
|
+
end
|
304
|
+
@stack.pop
|
305
|
+
end
|
306
|
+
end
|
307
|
+
|
308
|
+
def parse_special(i)
|
309
|
+
rawdata = @rawdata
|
310
|
+
match = rawdata.index(Endbracket, i+1)
|
311
|
+
return nil unless match
|
312
|
+
matched_length = $&.length
|
313
|
+
src_length = match - i + matched_length
|
314
|
+
set_range(i, i + src_length)
|
315
|
+
handle_special(rawdata[i+1..(match-1)])
|
316
|
+
return src_length
|
317
|
+
end
|
318
|
+
|
319
|
+
def handle_starttag(tag, method, attrs)
|
320
|
+
self.send(method, attrs)
|
321
|
+
end
|
322
|
+
|
323
|
+
def handle_endtag(tag, method)
|
324
|
+
self.send(method)
|
325
|
+
end
|
326
|
+
|
327
|
+
def report_unbalanced(tag)
|
328
|
+
if @verbose
|
329
|
+
print '*** Unbalanced </' + tag + '>', "\n"
|
330
|
+
print '*** Stack:', self.stack, "\n"
|
331
|
+
end
|
332
|
+
end
|
333
|
+
|
334
|
+
def handle_charref(name)
|
335
|
+
n = Integer(name)
|
336
|
+
if !(0 <= n && n <= 255)
|
337
|
+
unknown_charref(name)
|
338
|
+
return
|
339
|
+
end
|
340
|
+
handle_data(n.chr)
|
341
|
+
end
|
342
|
+
|
343
|
+
def handle_entityref(name)
|
344
|
+
table = Entitydefs
|
345
|
+
if table.include?(name)
|
346
|
+
handle_data(table[name])
|
347
|
+
else
|
348
|
+
unknown_entityref(name)
|
349
|
+
return
|
350
|
+
end
|
351
|
+
end
|
352
|
+
|
353
|
+
def handle_data(data)
|
354
|
+
end
|
355
|
+
|
356
|
+
def handle_comment(data)
|
357
|
+
end
|
358
|
+
|
359
|
+
def handle_special(data)
|
360
|
+
end
|
361
|
+
|
362
|
+
def unknown_starttag(tag, attrs)
|
363
|
+
end
|
364
|
+
def unknown_endtag(tag)
|
365
|
+
end
|
366
|
+
def unknown_charref(ref)
|
367
|
+
end
|
368
|
+
def unknown_entityref(ref)
|
369
|
+
end
|
370
|
+
|
371
|
+
end
|
372
|
+
end
|