rubysl-rexml 1.0.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +7 -0
- data/.gitignore +17 -0
- data/.travis.yml +8 -0
- data/Gemfile +4 -0
- data/LICENSE +25 -0
- data/README.md +29 -0
- data/Rakefile +1 -0
- data/lib/rexml/attlistdecl.rb +62 -0
- data/lib/rexml/attribute.rb +185 -0
- data/lib/rexml/cdata.rb +67 -0
- data/lib/rexml/child.rb +96 -0
- data/lib/rexml/comment.rb +80 -0
- data/lib/rexml/doctype.rb +271 -0
- data/lib/rexml/document.rb +230 -0
- data/lib/rexml/dtd/attlistdecl.rb +10 -0
- data/lib/rexml/dtd/dtd.rb +51 -0
- data/lib/rexml/dtd/elementdecl.rb +17 -0
- data/lib/rexml/dtd/entitydecl.rb +56 -0
- data/lib/rexml/dtd/notationdecl.rb +39 -0
- data/lib/rexml/element.rb +1227 -0
- data/lib/rexml/encoding.rb +71 -0
- data/lib/rexml/encodings/CP-1252.rb +103 -0
- data/lib/rexml/encodings/EUC-JP.rb +35 -0
- data/lib/rexml/encodings/ICONV.rb +22 -0
- data/lib/rexml/encodings/ISO-8859-1.rb +7 -0
- data/lib/rexml/encodings/ISO-8859-15.rb +72 -0
- data/lib/rexml/encodings/SHIFT-JIS.rb +37 -0
- data/lib/rexml/encodings/SHIFT_JIS.rb +1 -0
- data/lib/rexml/encodings/UNILE.rb +34 -0
- data/lib/rexml/encodings/US-ASCII.rb +30 -0
- data/lib/rexml/encodings/UTF-16.rb +35 -0
- data/lib/rexml/encodings/UTF-8.rb +18 -0
- data/lib/rexml/entity.rb +166 -0
- data/lib/rexml/formatters/default.rb +109 -0
- data/lib/rexml/formatters/pretty.rb +138 -0
- data/lib/rexml/formatters/transitive.rb +56 -0
- data/lib/rexml/functions.rb +382 -0
- data/lib/rexml/instruction.rb +70 -0
- data/lib/rexml/light/node.rb +196 -0
- data/lib/rexml/namespace.rb +47 -0
- data/lib/rexml/node.rb +75 -0
- data/lib/rexml/output.rb +24 -0
- data/lib/rexml/parent.rb +166 -0
- data/lib/rexml/parseexception.rb +51 -0
- data/lib/rexml/parsers/baseparser.rb +503 -0
- data/lib/rexml/parsers/lightparser.rb +60 -0
- data/lib/rexml/parsers/pullparser.rb +196 -0
- data/lib/rexml/parsers/sax2parser.rb +238 -0
- data/lib/rexml/parsers/streamparser.rb +46 -0
- data/lib/rexml/parsers/treeparser.rb +97 -0
- data/lib/rexml/parsers/ultralightparser.rb +56 -0
- data/lib/rexml/parsers/xpathparser.rb +698 -0
- data/lib/rexml/quickpath.rb +266 -0
- data/lib/rexml/rexml.rb +32 -0
- data/lib/rexml/sax2listener.rb +97 -0
- data/lib/rexml/source.rb +251 -0
- data/lib/rexml/streamlistener.rb +92 -0
- data/lib/rexml/syncenumerator.rb +33 -0
- data/lib/rexml/text.rb +344 -0
- data/lib/rexml/undefinednamespaceexception.rb +8 -0
- data/lib/rexml/validation/relaxng.rb +559 -0
- data/lib/rexml/validation/validation.rb +155 -0
- data/lib/rexml/validation/validationexception.rb +9 -0
- data/lib/rexml/xmldecl.rb +119 -0
- data/lib/rexml/xmltokens.rb +18 -0
- data/lib/rexml/xpath.rb +66 -0
- data/lib/rexml/xpath_parser.rb +792 -0
- data/lib/rubysl/rexml.rb +1 -0
- data/lib/rubysl/rexml/version.rb +5 -0
- data/rubysl-rexml.gemspec +23 -0
- data/spec/attribute/clone_spec.rb +10 -0
- data/spec/attribute/element_spec.rb +22 -0
- data/spec/attribute/equal_value_spec.rb +17 -0
- data/spec/attribute/hash_spec.rb +12 -0
- data/spec/attribute/initialize_spec.rb +28 -0
- data/spec/attribute/inspect_spec.rb +19 -0
- data/spec/attribute/namespace_spec.rb +23 -0
- data/spec/attribute/node_type_spec.rb +9 -0
- data/spec/attribute/prefix_spec.rb +17 -0
- data/spec/attribute/remove_spec.rb +19 -0
- data/spec/attribute/to_s_spec.rb +13 -0
- data/spec/attribute/to_string_spec.rb +14 -0
- data/spec/attribute/value_spec.rb +14 -0
- data/spec/attribute/write_spec.rb +22 -0
- data/spec/attribute/xpath_spec.rb +19 -0
- data/spec/attributes/add_spec.rb +6 -0
- data/spec/attributes/append_spec.rb +6 -0
- data/spec/attributes/delete_all_spec.rb +30 -0
- data/spec/attributes/delete_spec.rb +26 -0
- data/spec/attributes/each_attribute_spec.rb +24 -0
- data/spec/attributes/each_spec.rb +24 -0
- data/spec/attributes/element_reference_spec.rb +18 -0
- data/spec/attributes/element_set_spec.rb +25 -0
- data/spec/attributes/get_attribute_ns_spec.rb +13 -0
- data/spec/attributes/get_attribute_spec.rb +28 -0
- data/spec/attributes/initialize_spec.rb +18 -0
- data/spec/attributes/length_spec.rb +6 -0
- data/spec/attributes/namespaces_spec.rb +5 -0
- data/spec/attributes/prefixes_spec.rb +23 -0
- data/spec/attributes/shared/add.rb +17 -0
- data/spec/attributes/shared/length.rb +12 -0
- data/spec/attributes/size_spec.rb +6 -0
- data/spec/attributes/to_a_spec.rb +20 -0
- data/spec/cdata/clone_spec.rb +9 -0
- data/spec/cdata/initialize_spec.rb +24 -0
- data/spec/cdata/shared/to_s.rb +11 -0
- data/spec/cdata/to_s_spec.rb +6 -0
- data/spec/cdata/value_spec.rb +6 -0
- data/spec/document/add_element_spec.rb +30 -0
- data/spec/document/add_spec.rb +60 -0
- data/spec/document/clone_spec.rb +19 -0
- data/spec/document/doctype_spec.rb +14 -0
- data/spec/document/encoding_spec.rb +21 -0
- data/spec/document/expanded_name_spec.rb +15 -0
- data/spec/document/new_spec.rb +37 -0
- data/spec/document/node_type_spec.rb +7 -0
- data/spec/document/root_spec.rb +11 -0
- data/spec/document/stand_alone_spec.rb +18 -0
- data/spec/document/version_spec.rb +13 -0
- data/spec/document/write_spec.rb +38 -0
- data/spec/document/xml_decl_spec.rb +14 -0
- data/spec/element/add_attribute_spec.rb +40 -0
- data/spec/element/add_attributes_spec.rb +21 -0
- data/spec/element/add_element_spec.rb +38 -0
- data/spec/element/add_namespace_spec.rb +23 -0
- data/spec/element/add_text_spec.rb +23 -0
- data/spec/element/attribute_spec.rb +16 -0
- data/spec/element/attributes_spec.rb +18 -0
- data/spec/element/cdatas_spec.rb +23 -0
- data/spec/element/clone_spec.rb +28 -0
- data/spec/element/comments_spec.rb +20 -0
- data/spec/element/delete_attribute_spec.rb +38 -0
- data/spec/element/delete_element_spec.rb +50 -0
- data/spec/element/delete_namespace_spec.rb +24 -0
- data/spec/element/document_spec.rb +17 -0
- data/spec/element/each_element_with_attribute_spec.rb +34 -0
- data/spec/element/each_element_with_text_spec.rb +30 -0
- data/spec/element/get_text_spec.rb +17 -0
- data/spec/element/has_attributes_spec.rb +16 -0
- data/spec/element/has_elements_spec.rb +17 -0
- data/spec/element/has_text_spec.rb +15 -0
- data/spec/element/inspect_spec.rb +26 -0
- data/spec/element/instructions_spec.rb +20 -0
- data/spec/element/namespace_spec.rb +26 -0
- data/spec/element/namespaces_spec.rb +31 -0
- data/spec/element/new_spec.rb +34 -0
- data/spec/element/next_element_spec.rb +18 -0
- data/spec/element/node_type_spec.rb +7 -0
- data/spec/element/prefixes_spec.rb +22 -0
- data/spec/element/previous_element_spec.rb +19 -0
- data/spec/element/raw_spec.rb +23 -0
- data/spec/element/root_spec.rb +27 -0
- data/spec/element/text_spec.rb +45 -0
- data/spec/element/texts_spec.rb +15 -0
- data/spec/element/whitespace_spec.rb +22 -0
- data/spec/node/each_recursive_spec.rb +20 -0
- data/spec/node/find_first_recursive_spec.rb +24 -0
- data/spec/node/index_in_parent_spec.rb +14 -0
- data/spec/node/next_sibling_node_spec.rb +20 -0
- data/spec/node/parent_spec.rb +20 -0
- data/spec/node/previous_sibling_node_spec.rb +20 -0
- data/spec/shared/each_element.rb +35 -0
- data/spec/shared/elements_to_a.rb +35 -0
- data/spec/text/append_spec.rb +9 -0
- data/spec/text/clone_spec.rb +9 -0
- data/spec/text/comparison_spec.rb +24 -0
- data/spec/text/empty_spec.rb +11 -0
- data/spec/text/indent_text_spec.rb +23 -0
- data/spec/text/inspect_spec.rb +7 -0
- data/spec/text/new_spec.rb +48 -0
- data/spec/text/node_type_spec.rb +7 -0
- data/spec/text/normalize_spec.rb +7 -0
- data/spec/text/read_with_substitution_spec.rb +12 -0
- data/spec/text/to_s_spec.rb +17 -0
- data/spec/text/unnormalize_spec.rb +7 -0
- data/spec/text/value_spec.rb +36 -0
- data/spec/text/wrap_spec.rb +20 -0
- data/spec/text/write_with_substitution_spec.rb +32 -0
- metadata +385 -0
@@ -0,0 +1,155 @@
|
|
1
|
+
require 'rexml/validation/validationexception'
|
2
|
+
|
3
|
+
module REXML
|
4
|
+
module Validation
|
5
|
+
module Validator
|
6
|
+
NILEVENT = [ nil ]
|
7
|
+
def reset
|
8
|
+
@current = @root
|
9
|
+
@root.reset
|
10
|
+
@root.previous = true
|
11
|
+
@attr_stack = []
|
12
|
+
self
|
13
|
+
end
|
14
|
+
def dump
|
15
|
+
puts @root.inspect
|
16
|
+
end
|
17
|
+
def validate( event )
|
18
|
+
#puts "Current: #@current"
|
19
|
+
#puts "Event: #{event.inspect}"
|
20
|
+
@attr_stack = [] unless defined? @attr_stack
|
21
|
+
match = @current.next(event)
|
22
|
+
raise ValidationException.new( "Validation error. Expected: "+
|
23
|
+
@current.expected.join( " or " )+" from #{@current.inspect} "+
|
24
|
+
" but got #{Event.new( event[0], event[1] ).inspect}" ) unless match
|
25
|
+
@current = match
|
26
|
+
|
27
|
+
# Check for attributes
|
28
|
+
case event[0]
|
29
|
+
when :start_element
|
30
|
+
#puts "Checking attributes"
|
31
|
+
@attr_stack << event[2]
|
32
|
+
begin
|
33
|
+
sattr = [:start_attribute, nil]
|
34
|
+
eattr = [:end_attribute]
|
35
|
+
text = [:text, nil]
|
36
|
+
k,v = event[2].find { |k,v|
|
37
|
+
sattr[1] = k
|
38
|
+
#puts "Looking for #{sattr.inspect}"
|
39
|
+
m = @current.next( sattr )
|
40
|
+
#puts "Got #{m.inspect}"
|
41
|
+
if m
|
42
|
+
# If the state has text children...
|
43
|
+
#puts "Looking for #{eattr.inspect}"
|
44
|
+
#puts "Expect #{m.expected}"
|
45
|
+
if m.matches?( eattr )
|
46
|
+
#puts "Got end"
|
47
|
+
@current = m
|
48
|
+
else
|
49
|
+
#puts "Didn't get end"
|
50
|
+
text[1] = v
|
51
|
+
#puts "Looking for #{text.inspect}"
|
52
|
+
m = m.next( text )
|
53
|
+
#puts "Got #{m.inspect}"
|
54
|
+
text[1] = nil
|
55
|
+
return false unless m
|
56
|
+
@current = m if m
|
57
|
+
end
|
58
|
+
m = @current.next( eattr )
|
59
|
+
if m
|
60
|
+
@current = m
|
61
|
+
true
|
62
|
+
else
|
63
|
+
false
|
64
|
+
end
|
65
|
+
else
|
66
|
+
false
|
67
|
+
end
|
68
|
+
}
|
69
|
+
event[2].delete(k) if k
|
70
|
+
end while k
|
71
|
+
when :end_element
|
72
|
+
attrs = @attr_stack.pop
|
73
|
+
raise ValidationException.new( "Validation error. Illegal "+
|
74
|
+
" attributes: #{attrs.inspect}") if attrs.length > 0
|
75
|
+
end
|
76
|
+
end
|
77
|
+
end
|
78
|
+
|
79
|
+
class Event
|
80
|
+
def initialize(event_type, event_arg=nil )
|
81
|
+
@event_type = event_type
|
82
|
+
@event_arg = event_arg
|
83
|
+
end
|
84
|
+
|
85
|
+
attr_reader :event_type
|
86
|
+
attr_accessor :event_arg
|
87
|
+
|
88
|
+
def done?
|
89
|
+
@done
|
90
|
+
end
|
91
|
+
|
92
|
+
def single?
|
93
|
+
return (@event_type != :start_element and @event_type != :start_attribute)
|
94
|
+
end
|
95
|
+
|
96
|
+
def matches?( event )
|
97
|
+
#puts "#@event_type =? #{event[0]} && #@event_arg =? #{event[1]} "
|
98
|
+
return false unless event[0] == @event_type
|
99
|
+
case event[0]
|
100
|
+
when nil
|
101
|
+
return true
|
102
|
+
when :start_element
|
103
|
+
return true if event[1] == @event_arg
|
104
|
+
when :end_element
|
105
|
+
return true
|
106
|
+
when :start_attribute
|
107
|
+
return true if event[1] == @event_arg
|
108
|
+
when :end_attribute
|
109
|
+
return true
|
110
|
+
when :end_document
|
111
|
+
return true
|
112
|
+
when :text
|
113
|
+
return (@event_arg.nil? or @event_arg == event[1])
|
114
|
+
=begin
|
115
|
+
when :processing_instruction
|
116
|
+
false
|
117
|
+
when :xmldecl
|
118
|
+
false
|
119
|
+
when :start_doctype
|
120
|
+
false
|
121
|
+
when :end_doctype
|
122
|
+
false
|
123
|
+
when :externalentity
|
124
|
+
false
|
125
|
+
when :elementdecl
|
126
|
+
false
|
127
|
+
when :entity
|
128
|
+
false
|
129
|
+
when :attlistdecl
|
130
|
+
false
|
131
|
+
when :notationdecl
|
132
|
+
false
|
133
|
+
when :end_doctype
|
134
|
+
false
|
135
|
+
=end
|
136
|
+
else
|
137
|
+
false
|
138
|
+
end
|
139
|
+
end
|
140
|
+
|
141
|
+
def ==( other )
|
142
|
+
return false unless other.kind_of? Event
|
143
|
+
@event_type == other.event_type and @event_arg == other.event_arg
|
144
|
+
end
|
145
|
+
|
146
|
+
def to_s
|
147
|
+
inspect
|
148
|
+
end
|
149
|
+
|
150
|
+
def inspect
|
151
|
+
"#{@event_type.inspect}( #@event_arg )"
|
152
|
+
end
|
153
|
+
end
|
154
|
+
end
|
155
|
+
end
|
@@ -0,0 +1,119 @@
|
|
1
|
+
require 'rexml/encoding'
|
2
|
+
require 'rexml/source'
|
3
|
+
|
4
|
+
module REXML
|
5
|
+
# NEEDS DOCUMENTATION
|
6
|
+
class XMLDecl < Child
|
7
|
+
include Encoding
|
8
|
+
|
9
|
+
DEFAULT_VERSION = "1.0";
|
10
|
+
DEFAULT_ENCODING = "UTF-8";
|
11
|
+
DEFAULT_STANDALONE = "no";
|
12
|
+
START = '<\?xml';
|
13
|
+
STOP = '\?>';
|
14
|
+
|
15
|
+
attr_accessor :version, :standalone
|
16
|
+
attr_reader :writeencoding, :writethis
|
17
|
+
|
18
|
+
def initialize(version=DEFAULT_VERSION, encoding=nil, standalone=nil)
|
19
|
+
@writethis = true
|
20
|
+
@writeencoding = !encoding.nil?
|
21
|
+
if version.kind_of? XMLDecl
|
22
|
+
super()
|
23
|
+
@version = version.version
|
24
|
+
self.encoding = version.encoding
|
25
|
+
@writeencoding = version.writeencoding
|
26
|
+
@standalone = version.standalone
|
27
|
+
else
|
28
|
+
super()
|
29
|
+
@version = version
|
30
|
+
self.encoding = encoding
|
31
|
+
@standalone = standalone
|
32
|
+
end
|
33
|
+
@version = DEFAULT_VERSION if @version.nil?
|
34
|
+
end
|
35
|
+
|
36
|
+
def clone
|
37
|
+
XMLDecl.new(self)
|
38
|
+
end
|
39
|
+
|
40
|
+
# indent::
|
41
|
+
# Ignored. There must be no whitespace before an XML declaration
|
42
|
+
# transitive::
|
43
|
+
# Ignored
|
44
|
+
# ie_hack::
|
45
|
+
# Ignored
|
46
|
+
def write(writer, indent=-1, transitive=false, ie_hack=false)
|
47
|
+
return nil unless @writethis or writer.kind_of? Output
|
48
|
+
writer << START.sub(/\\/u, '')
|
49
|
+
if writer.kind_of? Output
|
50
|
+
writer << " #{content writer.encoding}"
|
51
|
+
else
|
52
|
+
writer << " #{content encoding}"
|
53
|
+
end
|
54
|
+
writer << STOP.sub(/\\/u, '')
|
55
|
+
end
|
56
|
+
|
57
|
+
def ==( other )
|
58
|
+
other.kind_of?(XMLDecl) and
|
59
|
+
other.version == @version and
|
60
|
+
other.encoding == self.encoding and
|
61
|
+
other.standalone == @standalone
|
62
|
+
end
|
63
|
+
|
64
|
+
def xmldecl version, encoding, standalone
|
65
|
+
@version = version
|
66
|
+
self.encoding = encoding
|
67
|
+
@standalone = standalone
|
68
|
+
end
|
69
|
+
|
70
|
+
def node_type
|
71
|
+
:xmldecl
|
72
|
+
end
|
73
|
+
|
74
|
+
alias :stand_alone? :standalone
|
75
|
+
alias :old_enc= :encoding=
|
76
|
+
|
77
|
+
def encoding=( enc )
|
78
|
+
if enc.nil?
|
79
|
+
self.old_enc = "UTF-8"
|
80
|
+
@writeencoding = false
|
81
|
+
else
|
82
|
+
self.old_enc = enc
|
83
|
+
@writeencoding = true
|
84
|
+
end
|
85
|
+
self.dowrite
|
86
|
+
end
|
87
|
+
|
88
|
+
# Only use this if you do not want the XML declaration to be written;
|
89
|
+
# this object is ignored by the XML writer. Otherwise, instantiate your
|
90
|
+
# own XMLDecl and add it to the document.
|
91
|
+
#
|
92
|
+
# Note that XML 1.1 documents *must* include an XML declaration
|
93
|
+
def XMLDecl.default
|
94
|
+
rv = XMLDecl.new( "1.0" )
|
95
|
+
rv.nowrite
|
96
|
+
rv
|
97
|
+
end
|
98
|
+
|
99
|
+
def nowrite
|
100
|
+
@writethis = false
|
101
|
+
end
|
102
|
+
|
103
|
+
def dowrite
|
104
|
+
@writethis = true
|
105
|
+
end
|
106
|
+
|
107
|
+
def inspect
|
108
|
+
START.sub(/\\/u, '') + " ... " + STOP.sub(/\\/u, '')
|
109
|
+
end
|
110
|
+
|
111
|
+
private
|
112
|
+
def content(enc)
|
113
|
+
rv = "version='#@version'"
|
114
|
+
rv << " encoding='#{enc}'" if @writeencoding || enc !~ /utf-8/i
|
115
|
+
rv << " standalone='#@standalone'" if @standalone
|
116
|
+
rv
|
117
|
+
end
|
118
|
+
end
|
119
|
+
end
|
@@ -0,0 +1,18 @@
|
|
1
|
+
module REXML
|
2
|
+
# Defines a number of tokens used for parsing XML. Not for general
|
3
|
+
# consumption.
|
4
|
+
module XMLTokens
|
5
|
+
NCNAME_STR= '[\w:][\-\w\d.]*'
|
6
|
+
NAME_STR= "(?:#{NCNAME_STR}:)?#{NCNAME_STR}"
|
7
|
+
|
8
|
+
NAMECHAR = '[\-\w\d\.:]'
|
9
|
+
NAME = "([\\w:]#{NAMECHAR}*)"
|
10
|
+
NMTOKEN = "(?:#{NAMECHAR})+"
|
11
|
+
NMTOKENS = "#{NMTOKEN}(\\s+#{NMTOKEN})*"
|
12
|
+
REFERENCE = "(?:&#{NAME};|&#\\d+;|&#x[0-9a-fA-F]+;)"
|
13
|
+
|
14
|
+
#REFERENCE = "(?:#{ENTITYREF}|#{CHARREF})"
|
15
|
+
#ENTITYREF = "&#{NAME};"
|
16
|
+
#CHARREF = "&#\\d+;|&#x[0-9a-fA-F]+;"
|
17
|
+
end
|
18
|
+
end
|
data/lib/rexml/xpath.rb
ADDED
@@ -0,0 +1,66 @@
|
|
1
|
+
require 'rexml/functions'
|
2
|
+
require 'rexml/xpath_parser'
|
3
|
+
|
4
|
+
module REXML
|
5
|
+
# Wrapper class. Use this class to access the XPath functions.
|
6
|
+
class XPath
|
7
|
+
include Functions
|
8
|
+
EMPTY_HASH = {}
|
9
|
+
|
10
|
+
# Finds and returns the first node that matches the supplied xpath.
|
11
|
+
# element::
|
12
|
+
# The context element
|
13
|
+
# path::
|
14
|
+
# The xpath to search for. If not supplied or nil, returns the first
|
15
|
+
# node matching '*'.
|
16
|
+
# namespaces::
|
17
|
+
# If supplied, a Hash which defines a namespace mapping.
|
18
|
+
#
|
19
|
+
# XPath.first( node )
|
20
|
+
# XPath.first( doc, "//b"} )
|
21
|
+
# XPath.first( node, "a/x:b", { "x"=>"http://doofus" } )
|
22
|
+
def XPath::first element, path=nil, namespaces=nil, variables={}
|
23
|
+
raise "The namespaces argument, if supplied, must be a hash object." unless namespaces.nil? or namespaces.kind_of?(Hash)
|
24
|
+
raise "The variables argument, if supplied, must be a hash object." unless variables.kind_of?(Hash)
|
25
|
+
parser = XPathParser.new
|
26
|
+
parser.namespaces = namespaces
|
27
|
+
parser.variables = variables
|
28
|
+
path = "*" unless path
|
29
|
+
element = [element] unless element.kind_of? Array
|
30
|
+
parser.parse(path, element).flatten[0]
|
31
|
+
end
|
32
|
+
|
33
|
+
# Iterates over nodes that match the given path, calling the supplied
|
34
|
+
# block with the match.
|
35
|
+
# element::
|
36
|
+
# The context element
|
37
|
+
# path::
|
38
|
+
# The xpath to search for. If not supplied or nil, defaults to '*'
|
39
|
+
# namespaces::
|
40
|
+
# If supplied, a Hash which defines a namespace mapping
|
41
|
+
#
|
42
|
+
# XPath.each( node ) { |el| ... }
|
43
|
+
# XPath.each( node, '/*[@attr='v']' ) { |el| ... }
|
44
|
+
# XPath.each( node, 'ancestor::x' ) { |el| ... }
|
45
|
+
def XPath::each element, path=nil, namespaces=nil, variables={}, &block
|
46
|
+
raise "The namespaces argument, if supplied, must be a hash object." unless namespaces.nil? or namespaces.kind_of?(Hash)
|
47
|
+
raise "The variables argument, if supplied, must be a hash object." unless variables.kind_of?(Hash)
|
48
|
+
parser = XPathParser.new
|
49
|
+
parser.namespaces = namespaces
|
50
|
+
parser.variables = variables
|
51
|
+
path = "*" unless path
|
52
|
+
element = [element] unless element.kind_of? Array
|
53
|
+
parser.parse(path, element).each( &block )
|
54
|
+
end
|
55
|
+
|
56
|
+
# Returns an array of nodes matching a given XPath.
|
57
|
+
def XPath::match element, path=nil, namespaces=nil, variables={}
|
58
|
+
parser = XPathParser.new
|
59
|
+
parser.namespaces = namespaces
|
60
|
+
parser.variables = variables
|
61
|
+
path = "*" unless path
|
62
|
+
element = [element] unless element.kind_of? Array
|
63
|
+
parser.parse(path,element)
|
64
|
+
end
|
65
|
+
end
|
66
|
+
end
|
@@ -0,0 +1,792 @@
|
|
1
|
+
require 'rexml/namespace'
|
2
|
+
require 'rexml/xmltokens'
|
3
|
+
require 'rexml/attribute'
|
4
|
+
require 'rexml/syncenumerator'
|
5
|
+
require 'rexml/parsers/xpathparser'
|
6
|
+
|
7
|
+
class Object
|
8
|
+
def dclone
|
9
|
+
clone
|
10
|
+
end
|
11
|
+
end
|
12
|
+
class Symbol
|
13
|
+
def dclone ; self ; end
|
14
|
+
end
|
15
|
+
class Fixnum
|
16
|
+
def dclone ; self ; end
|
17
|
+
end
|
18
|
+
class Float
|
19
|
+
def dclone ; self ; end
|
20
|
+
end
|
21
|
+
class Array
|
22
|
+
def dclone
|
23
|
+
klone = self.clone
|
24
|
+
klone.clear
|
25
|
+
self.each{|v| klone << v.dclone}
|
26
|
+
klone
|
27
|
+
end
|
28
|
+
end
|
29
|
+
|
30
|
+
module REXML
|
31
|
+
# You don't want to use this class. Really. Use XPath, which is a wrapper
|
32
|
+
# for this class. Believe me. You don't want to poke around in here.
|
33
|
+
# There is strange, dark magic at work in this code. Beware. Go back! Go
|
34
|
+
# back while you still can!
|
35
|
+
class XPathParser
|
36
|
+
include XMLTokens
|
37
|
+
LITERAL = /^'([^']*)'|^"([^"]*)"/u
|
38
|
+
|
39
|
+
def initialize( )
|
40
|
+
@parser = REXML::Parsers::XPathParser.new
|
41
|
+
@namespaces = nil
|
42
|
+
@variables = {}
|
43
|
+
end
|
44
|
+
|
45
|
+
def namespaces=( namespaces={} )
|
46
|
+
Functions::namespace_context = namespaces
|
47
|
+
@namespaces = namespaces
|
48
|
+
end
|
49
|
+
|
50
|
+
def variables=( vars={} )
|
51
|
+
Functions::variables = vars
|
52
|
+
@variables = vars
|
53
|
+
end
|
54
|
+
|
55
|
+
def parse path, nodeset
|
56
|
+
#puts "#"*40
|
57
|
+
path_stack = @parser.parse( path )
|
58
|
+
#puts "PARSE: #{path} => #{path_stack.inspect}"
|
59
|
+
#puts "PARSE: nodeset = #{nodeset.inspect}"
|
60
|
+
match( path_stack, nodeset )
|
61
|
+
end
|
62
|
+
|
63
|
+
def get_first path, nodeset
|
64
|
+
#puts "#"*40
|
65
|
+
path_stack = @parser.parse( path )
|
66
|
+
#puts "PARSE: #{path} => #{path_stack.inspect}"
|
67
|
+
#puts "PARSE: nodeset = #{nodeset.inspect}"
|
68
|
+
first( path_stack, nodeset )
|
69
|
+
end
|
70
|
+
|
71
|
+
def predicate path, nodeset
|
72
|
+
path_stack = @parser.parse( path )
|
73
|
+
expr( path_stack, nodeset )
|
74
|
+
end
|
75
|
+
|
76
|
+
def []=( variable_name, value )
|
77
|
+
@variables[ variable_name ] = value
|
78
|
+
end
|
79
|
+
|
80
|
+
|
81
|
+
# Performs a depth-first (document order) XPath search, and returns the
|
82
|
+
# first match. This is the fastest, lightest way to return a single result.
|
83
|
+
#
|
84
|
+
# FIXME: This method is incomplete!
|
85
|
+
def first( path_stack, node )
|
86
|
+
#puts "#{depth}) Entering match( #{path.inspect}, #{tree.inspect} )"
|
87
|
+
return nil if path.size == 0
|
88
|
+
|
89
|
+
case path[0]
|
90
|
+
when :document
|
91
|
+
# do nothing
|
92
|
+
return first( path[1..-1], node )
|
93
|
+
when :child
|
94
|
+
for c in node.children
|
95
|
+
#puts "#{depth}) CHILD checking #{name(c)}"
|
96
|
+
r = first( path[1..-1], c )
|
97
|
+
#puts "#{depth}) RETURNING #{r.inspect}" if r
|
98
|
+
return r if r
|
99
|
+
end
|
100
|
+
when :qname
|
101
|
+
name = path[2]
|
102
|
+
#puts "#{depth}) QNAME #{name(tree)} == #{name} (path => #{path.size})"
|
103
|
+
if node.name == name
|
104
|
+
#puts "#{depth}) RETURNING #{tree.inspect}" if path.size == 3
|
105
|
+
return node if path.size == 3
|
106
|
+
return first( path[3..-1], node )
|
107
|
+
else
|
108
|
+
return nil
|
109
|
+
end
|
110
|
+
when :descendant_or_self
|
111
|
+
r = first( path[1..-1], node )
|
112
|
+
return r if r
|
113
|
+
for c in node.children
|
114
|
+
r = first( path, c )
|
115
|
+
return r if r
|
116
|
+
end
|
117
|
+
when :node
|
118
|
+
return first( path[1..-1], node )
|
119
|
+
when :any
|
120
|
+
return first( path[1..-1], node )
|
121
|
+
end
|
122
|
+
return nil
|
123
|
+
end
|
124
|
+
|
125
|
+
|
126
|
+
def match( path_stack, nodeset )
|
127
|
+
#puts "MATCH: path_stack = #{path_stack.inspect}"
|
128
|
+
#puts "MATCH: nodeset = #{nodeset.inspect}"
|
129
|
+
r = expr( path_stack, nodeset )
|
130
|
+
#puts "MAIN EXPR => #{r.inspect}"
|
131
|
+
r
|
132
|
+
end
|
133
|
+
|
134
|
+
private
|
135
|
+
|
136
|
+
|
137
|
+
# Returns a String namespace for a node, given a prefix
|
138
|
+
# The rules are:
|
139
|
+
#
|
140
|
+
# 1. Use the supplied namespace mapping first.
|
141
|
+
# 2. If no mapping was supplied, use the context node to look up the namespace
|
142
|
+
def get_namespace( node, prefix )
|
143
|
+
if @namespaces
|
144
|
+
return @namespaces[prefix] || ''
|
145
|
+
else
|
146
|
+
return node.namespace( prefix ) if node.node_type == :element
|
147
|
+
return ''
|
148
|
+
end
|
149
|
+
end
|
150
|
+
|
151
|
+
|
152
|
+
# Expr takes a stack of path elements and a set of nodes (either a Parent
|
153
|
+
# or an Array and returns an Array of matching nodes
|
154
|
+
ALL = [ :attribute, :element, :text, :processing_instruction, :comment ]
|
155
|
+
ELEMENTS = [ :element ]
|
156
|
+
def expr( path_stack, nodeset, context=nil )
|
157
|
+
#puts "#"*15
|
158
|
+
#puts "In expr with #{path_stack.inspect}"
|
159
|
+
#puts "Returning" if path_stack.length == 0 || nodeset.length == 0
|
160
|
+
node_types = ELEMENTS
|
161
|
+
return nodeset if path_stack.length == 0 || nodeset.length == 0
|
162
|
+
while path_stack.length > 0
|
163
|
+
#puts "#"*5
|
164
|
+
#puts "Path stack = #{path_stack.inspect}"
|
165
|
+
#puts "Nodeset is #{nodeset.inspect}"
|
166
|
+
if nodeset.length == 0
|
167
|
+
path_stack.clear
|
168
|
+
return []
|
169
|
+
end
|
170
|
+
case (op = path_stack.shift)
|
171
|
+
when :document
|
172
|
+
nodeset = [ nodeset[0].root_node ]
|
173
|
+
#puts ":document, nodeset = #{nodeset.inspect}"
|
174
|
+
|
175
|
+
when :qname
|
176
|
+
#puts "IN QNAME"
|
177
|
+
prefix = path_stack.shift
|
178
|
+
name = path_stack.shift
|
179
|
+
nodeset.delete_if do |node|
|
180
|
+
# FIXME: This DOUBLES the time XPath searches take
|
181
|
+
ns = get_namespace( node, prefix )
|
182
|
+
#puts "NS = #{ns.inspect}"
|
183
|
+
#puts "node.node_type == :element => #{node.node_type == :element}"
|
184
|
+
if node.node_type == :element
|
185
|
+
#puts "node.name == #{name} => #{node.name == name}"
|
186
|
+
if node.name == name
|
187
|
+
#puts "node.namespace == #{ns.inspect} => #{node.namespace == ns}"
|
188
|
+
end
|
189
|
+
end
|
190
|
+
!(node.node_type == :element and
|
191
|
+
node.name == name and
|
192
|
+
node.namespace == ns )
|
193
|
+
end
|
194
|
+
node_types = ELEMENTS
|
195
|
+
|
196
|
+
when :any
|
197
|
+
#puts "ANY 1: nodeset = #{nodeset.inspect}"
|
198
|
+
#puts "ANY 1: node_types = #{node_types.inspect}"
|
199
|
+
nodeset.delete_if { |node| !node_types.include?(node.node_type) }
|
200
|
+
#puts "ANY 2: nodeset = #{nodeset.inspect}"
|
201
|
+
|
202
|
+
when :self
|
203
|
+
# This space left intentionally blank
|
204
|
+
|
205
|
+
when :processing_instruction
|
206
|
+
target = path_stack.shift
|
207
|
+
nodeset.delete_if do |node|
|
208
|
+
(node.node_type != :processing_instruction) or
|
209
|
+
( target!='' and ( node.target != target ) )
|
210
|
+
end
|
211
|
+
|
212
|
+
when :text
|
213
|
+
nodeset.delete_if { |node| node.node_type != :text }
|
214
|
+
|
215
|
+
when :comment
|
216
|
+
nodeset.delete_if { |node| node.node_type != :comment }
|
217
|
+
|
218
|
+
when :node
|
219
|
+
# This space left intentionally blank
|
220
|
+
node_types = ALL
|
221
|
+
|
222
|
+
when :child
|
223
|
+
new_nodeset = []
|
224
|
+
nt = nil
|
225
|
+
for node in nodeset
|
226
|
+
nt = node.node_type
|
227
|
+
new_nodeset += node.children if nt == :element or nt == :document
|
228
|
+
end
|
229
|
+
nodeset = new_nodeset
|
230
|
+
node_types = ELEMENTS
|
231
|
+
|
232
|
+
when :literal
|
233
|
+
return path_stack.shift
|
234
|
+
|
235
|
+
when :attribute
|
236
|
+
new_nodeset = []
|
237
|
+
case path_stack.shift
|
238
|
+
when :qname
|
239
|
+
prefix = path_stack.shift
|
240
|
+
name = path_stack.shift
|
241
|
+
for element in nodeset
|
242
|
+
if element.node_type == :element
|
243
|
+
#puts "Element name = #{element.name}"
|
244
|
+
#puts "get_namespace( #{element.inspect}, #{prefix} ) = #{get_namespace(element, prefix)}"
|
245
|
+
attrib = element.attribute( name, get_namespace(element, prefix) )
|
246
|
+
#puts "attrib = #{attrib.inspect}"
|
247
|
+
new_nodeset << attrib if attrib
|
248
|
+
end
|
249
|
+
end
|
250
|
+
when :any
|
251
|
+
#puts "ANY"
|
252
|
+
for element in nodeset
|
253
|
+
if element.node_type == :element
|
254
|
+
new_nodeset += element.attributes.to_a
|
255
|
+
end
|
256
|
+
end
|
257
|
+
end
|
258
|
+
nodeset = new_nodeset
|
259
|
+
|
260
|
+
when :parent
|
261
|
+
#puts "PARENT 1: nodeset = #{nodeset}"
|
262
|
+
nodeset = nodeset.collect{|n| n.parent}.compact
|
263
|
+
#nodeset = expr(path_stack.dclone, nodeset.collect{|n| n.parent}.compact)
|
264
|
+
#puts "PARENT 2: nodeset = #{nodeset.inspect}"
|
265
|
+
node_types = ELEMENTS
|
266
|
+
|
267
|
+
when :ancestor
|
268
|
+
new_nodeset = []
|
269
|
+
for node in nodeset
|
270
|
+
while node.parent
|
271
|
+
node = node.parent
|
272
|
+
new_nodeset << node unless new_nodeset.include? node
|
273
|
+
end
|
274
|
+
end
|
275
|
+
nodeset = new_nodeset
|
276
|
+
node_types = ELEMENTS
|
277
|
+
|
278
|
+
when :ancestor_or_self
|
279
|
+
new_nodeset = []
|
280
|
+
for node in nodeset
|
281
|
+
if node.node_type == :element
|
282
|
+
new_nodeset << node
|
283
|
+
while ( node.parent )
|
284
|
+
node = node.parent
|
285
|
+
new_nodeset << node unless new_nodeset.include? node
|
286
|
+
end
|
287
|
+
end
|
288
|
+
end
|
289
|
+
nodeset = new_nodeset
|
290
|
+
node_types = ELEMENTS
|
291
|
+
|
292
|
+
when :predicate
|
293
|
+
new_nodeset = []
|
294
|
+
subcontext = { :size => nodeset.size }
|
295
|
+
pred = path_stack.shift
|
296
|
+
nodeset.each_with_index { |node, index|
|
297
|
+
subcontext[ :node ] = node
|
298
|
+
#puts "PREDICATE SETTING CONTEXT INDEX TO #{index+1}"
|
299
|
+
subcontext[ :index ] = index+1
|
300
|
+
pc = pred.dclone
|
301
|
+
#puts "#{node.hash}) Recursing with #{pred.inspect} and [#{node.inspect}]"
|
302
|
+
result = expr( pc, [node], subcontext )
|
303
|
+
result = result[0] if result.kind_of? Array and result.length == 1
|
304
|
+
#puts "#{node.hash}) Result = #{result.inspect} (#{result.class.name})"
|
305
|
+
if result.kind_of? Numeric
|
306
|
+
#puts "Adding node #{node.inspect}" if result == (index+1)
|
307
|
+
new_nodeset << node if result == (index+1)
|
308
|
+
elsif result.instance_of? Array
|
309
|
+
if result.size > 0 and result.inject(false) {|k,s| s or k}
|
310
|
+
#puts "Adding node #{node.inspect}" if result.size > 0
|
311
|
+
new_nodeset << node if result.size > 0
|
312
|
+
end
|
313
|
+
else
|
314
|
+
#puts "Adding node #{node.inspect}" if result
|
315
|
+
new_nodeset << node if result
|
316
|
+
end
|
317
|
+
}
|
318
|
+
#puts "New nodeset = #{new_nodeset.inspect}"
|
319
|
+
#puts "Path_stack = #{path_stack.inspect}"
|
320
|
+
nodeset = new_nodeset
|
321
|
+
=begin
|
322
|
+
predicate = path_stack.shift
|
323
|
+
ns = nodeset.clone
|
324
|
+
result = expr( predicate, ns )
|
325
|
+
#puts "Result = #{result.inspect} (#{result.class.name})"
|
326
|
+
#puts "nodeset = #{nodeset.inspect}"
|
327
|
+
if result.kind_of? Array
|
328
|
+
nodeset = result.zip(ns).collect{|m,n| n if m}.compact
|
329
|
+
else
|
330
|
+
nodeset = result ? nodeset : []
|
331
|
+
end
|
332
|
+
#puts "Outgoing NS = #{nodeset.inspect}"
|
333
|
+
=end
|
334
|
+
|
335
|
+
when :descendant_or_self
|
336
|
+
rv = descendant_or_self( path_stack, nodeset )
|
337
|
+
path_stack.clear
|
338
|
+
nodeset = rv
|
339
|
+
node_types = ELEMENTS
|
340
|
+
|
341
|
+
when :descendant
|
342
|
+
results = []
|
343
|
+
nt = nil
|
344
|
+
for node in nodeset
|
345
|
+
nt = node.node_type
|
346
|
+
results += expr( path_stack.dclone.unshift( :descendant_or_self ),
|
347
|
+
node.children ) if nt == :element or nt == :document
|
348
|
+
end
|
349
|
+
nodeset = results
|
350
|
+
node_types = ELEMENTS
|
351
|
+
|
352
|
+
when :following_sibling
|
353
|
+
#puts "FOLLOWING_SIBLING 1: nodeset = #{nodeset}"
|
354
|
+
results = []
|
355
|
+
nodeset.each do |node|
|
356
|
+
next if node.parent.nil?
|
357
|
+
all_siblings = node.parent.children
|
358
|
+
current_index = all_siblings.index( node )
|
359
|
+
following_siblings = all_siblings[ current_index+1 .. -1 ]
|
360
|
+
results += expr( path_stack.dclone, following_siblings )
|
361
|
+
end
|
362
|
+
#puts "FOLLOWING_SIBLING 2: nodeset = #{nodeset}"
|
363
|
+
nodeset = results
|
364
|
+
|
365
|
+
when :preceding_sibling
|
366
|
+
results = []
|
367
|
+
nodeset.each do |node|
|
368
|
+
next if node.parent.nil?
|
369
|
+
all_siblings = node.parent.children
|
370
|
+
current_index = all_siblings.index( node )
|
371
|
+
preceding_siblings = all_siblings[ 0, current_index ].reverse
|
372
|
+
results += preceding_siblings
|
373
|
+
end
|
374
|
+
nodeset = results
|
375
|
+
node_types = ELEMENTS
|
376
|
+
|
377
|
+
when :preceding
|
378
|
+
new_nodeset = []
|
379
|
+
for node in nodeset
|
380
|
+
new_nodeset += preceding( node )
|
381
|
+
end
|
382
|
+
#puts "NEW NODESET => #{new_nodeset.inspect}"
|
383
|
+
nodeset = new_nodeset
|
384
|
+
node_types = ELEMENTS
|
385
|
+
|
386
|
+
when :following
|
387
|
+
new_nodeset = []
|
388
|
+
for node in nodeset
|
389
|
+
new_nodeset += following( node )
|
390
|
+
end
|
391
|
+
nodeset = new_nodeset
|
392
|
+
node_types = ELEMENTS
|
393
|
+
|
394
|
+
when :namespace
|
395
|
+
#puts "In :namespace"
|
396
|
+
new_nodeset = []
|
397
|
+
prefix = path_stack.shift
|
398
|
+
for node in nodeset
|
399
|
+
if (node.node_type == :element or node.node_type == :attribute)
|
400
|
+
if @namespaces
|
401
|
+
namespaces = @namespaces
|
402
|
+
elsif (node.node_type == :element)
|
403
|
+
namespaces = node.namespaces
|
404
|
+
else
|
405
|
+
namespaces = node.element.namesapces
|
406
|
+
end
|
407
|
+
#puts "Namespaces = #{namespaces.inspect}"
|
408
|
+
#puts "Prefix = #{prefix.inspect}"
|
409
|
+
#puts "Node.namespace = #{node.namespace}"
|
410
|
+
if (node.namespace == namespaces[prefix])
|
411
|
+
new_nodeset << node
|
412
|
+
end
|
413
|
+
end
|
414
|
+
end
|
415
|
+
nodeset = new_nodeset
|
416
|
+
|
417
|
+
when :variable
|
418
|
+
var_name = path_stack.shift
|
419
|
+
return @variables[ var_name ]
|
420
|
+
|
421
|
+
# :and, :or, :eq, :neq, :lt, :lteq, :gt, :gteq
|
422
|
+
# TODO: Special case for :or and :and -- not evaluate the right
|
423
|
+
# operand if the left alone determines result (i.e. is true for
|
424
|
+
# :or and false for :and).
|
425
|
+
when :eq, :neq, :lt, :lteq, :gt, :gteq, :and, :or
|
426
|
+
left = expr( path_stack.shift, nodeset.dup, context )
|
427
|
+
#puts "LEFT => #{left.inspect} (#{left.class.name})"
|
428
|
+
right = expr( path_stack.shift, nodeset.dup, context )
|
429
|
+
#puts "RIGHT => #{right.inspect} (#{right.class.name})"
|
430
|
+
res = equality_relational_compare( left, op, right )
|
431
|
+
#puts "RES => #{res.inspect}"
|
432
|
+
return res
|
433
|
+
|
434
|
+
when :and
|
435
|
+
left = expr( path_stack.shift, nodeset.dup, context )
|
436
|
+
#puts "LEFT => #{left.inspect} (#{left.class.name})"
|
437
|
+
if left == false || left.nil? || !left.inject(false) {|a,b| a | b}
|
438
|
+
return []
|
439
|
+
end
|
440
|
+
right = expr( path_stack.shift, nodeset.dup, context )
|
441
|
+
#puts "RIGHT => #{right.inspect} (#{right.class.name})"
|
442
|
+
res = equality_relational_compare( left, op, right )
|
443
|
+
#puts "RES => #{res.inspect}"
|
444
|
+
return res
|
445
|
+
|
446
|
+
when :div
|
447
|
+
left = Functions::number(expr(path_stack.shift, nodeset, context)).to_f
|
448
|
+
right = Functions::number(expr(path_stack.shift, nodeset, context)).to_f
|
449
|
+
return (left / right)
|
450
|
+
|
451
|
+
when :mod
|
452
|
+
left = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
453
|
+
right = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
454
|
+
return (left % right)
|
455
|
+
|
456
|
+
when :mult
|
457
|
+
left = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
458
|
+
right = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
459
|
+
return (left * right)
|
460
|
+
|
461
|
+
when :plus
|
462
|
+
left = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
463
|
+
right = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
464
|
+
return (left + right)
|
465
|
+
|
466
|
+
when :minus
|
467
|
+
left = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
468
|
+
right = Functions::number(expr(path_stack.shift, nodeset, context )).to_f
|
469
|
+
return (left - right)
|
470
|
+
|
471
|
+
when :union
|
472
|
+
left = expr( path_stack.shift, nodeset, context )
|
473
|
+
right = expr( path_stack.shift, nodeset, context )
|
474
|
+
return (left | right)
|
475
|
+
|
476
|
+
when :neg
|
477
|
+
res = expr( path_stack, nodeset, context )
|
478
|
+
return -(res.to_f)
|
479
|
+
|
480
|
+
when :not
|
481
|
+
when :function
|
482
|
+
func_name = path_stack.shift.tr('-','_')
|
483
|
+
arguments = path_stack.shift
|
484
|
+
#puts "FUNCTION 0: #{func_name}(#{arguments.collect{|a|a.inspect}.join(', ')})"
|
485
|
+
subcontext = context ? nil : { :size => nodeset.size }
|
486
|
+
|
487
|
+
res = []
|
488
|
+
cont = context
|
489
|
+
nodeset.each_with_index { |n, i|
|
490
|
+
if subcontext
|
491
|
+
subcontext[:node] = n
|
492
|
+
subcontext[:index] = i
|
493
|
+
cont = subcontext
|
494
|
+
end
|
495
|
+
arg_clone = arguments.dclone
|
496
|
+
args = arg_clone.collect { |arg|
|
497
|
+
#puts "FUNCTION 1: Calling expr( #{arg.inspect}, [#{n.inspect}] )"
|
498
|
+
expr( arg, [n], cont )
|
499
|
+
}
|
500
|
+
#puts "FUNCTION 2: #{func_name}(#{args.collect{|a|a.inspect}.join(', ')})"
|
501
|
+
Functions.context = cont
|
502
|
+
res << Functions.send( func_name, *args )
|
503
|
+
#puts "FUNCTION 3: #{res[-1].inspect}"
|
504
|
+
}
|
505
|
+
return res
|
506
|
+
|
507
|
+
end
|
508
|
+
end # while
|
509
|
+
#puts "EXPR returning #{nodeset.inspect}"
|
510
|
+
return nodeset
|
511
|
+
end
|
512
|
+
|
513
|
+
|
514
|
+
##########################################################
|
515
|
+
# FIXME
|
516
|
+
# The next two methods are BAD MOJO!
|
517
|
+
# This is my achilles heel. If anybody thinks of a better
|
518
|
+
# way of doing this, be my guest. This really sucks, but
|
519
|
+
# it is a wonder it works at all.
|
520
|
+
# ########################################################
|
521
|
+
|
522
|
+
def descendant_or_self( path_stack, nodeset )
|
523
|
+
rs = []
|
524
|
+
#puts "#"*80
|
525
|
+
#puts "PATH_STACK = #{path_stack.inspect}"
|
526
|
+
#puts "NODESET = #{nodeset.collect{|n|n.inspect}.inspect}"
|
527
|
+
d_o_s( path_stack, nodeset, rs )
|
528
|
+
#puts "RS = #{rs.collect{|n|n.inspect}.inspect}"
|
529
|
+
document_order(rs.flatten.compact)
|
530
|
+
#rs.flatten.compact
|
531
|
+
end
|
532
|
+
|
533
|
+
def d_o_s( p, ns, r )
|
534
|
+
#puts "IN DOS with #{ns.inspect}; ALREADY HAVE #{r.inspect}"
|
535
|
+
nt = nil
|
536
|
+
ns.each_index do |i|
|
537
|
+
n = ns[i]
|
538
|
+
#puts "P => #{p.inspect}"
|
539
|
+
x = expr( p.dclone, [ n ] )
|
540
|
+
nt = n.node_type
|
541
|
+
d_o_s( p, n.children, x ) if nt == :element or nt == :document and n.children.size > 0
|
542
|
+
r.concat(x) if x.size > 0
|
543
|
+
end
|
544
|
+
end
|
545
|
+
|
546
|
+
|
547
|
+
# Reorders an array of nodes so that they are in document order
|
548
|
+
# It tries to do this efficiently.
|
549
|
+
#
|
550
|
+
# FIXME: I need to get rid of this, but the issue is that most of the XPath
|
551
|
+
# interpreter functions as a filter, which means that we lose context going
|
552
|
+
# in and out of function calls. If I knew what the index of the nodes was,
|
553
|
+
# I wouldn't have to do this. Maybe add a document IDX for each node?
|
554
|
+
# Problems with mutable documents. Or, rewrite everything.
|
555
|
+
def document_order( array_of_nodes )
|
556
|
+
new_arry = []
|
557
|
+
array_of_nodes.each { |node|
|
558
|
+
node_idx = []
|
559
|
+
np = node.node_type == :attribute ? node.element : node
|
560
|
+
while np.parent and np.parent.node_type == :element
|
561
|
+
node_idx << np.parent.index( np )
|
562
|
+
np = np.parent
|
563
|
+
end
|
564
|
+
new_arry << [ node_idx.reverse, node ]
|
565
|
+
}
|
566
|
+
#puts "new_arry = #{new_arry.inspect}"
|
567
|
+
new_arry.sort{ |s1, s2| s1[0] <=> s2[0] }.collect{ |s| s[1] }
|
568
|
+
end
|
569
|
+
|
570
|
+
|
571
|
+
def recurse( nodeset, &block )
|
572
|
+
for node in nodeset
|
573
|
+
yield node
|
574
|
+
recurse( node, &block ) if node.node_type == :element
|
575
|
+
end
|
576
|
+
end
|
577
|
+
|
578
|
+
|
579
|
+
|
580
|
+
# Builds a nodeset of all of the preceding nodes of the supplied node,
|
581
|
+
# in reverse document order
|
582
|
+
# preceding:: includes every element in the document that precedes this node,
|
583
|
+
# except for ancestors
|
584
|
+
def preceding( node )
|
585
|
+
#puts "IN PRECEDING"
|
586
|
+
ancestors = []
|
587
|
+
p = node.parent
|
588
|
+
while p
|
589
|
+
ancestors << p
|
590
|
+
p = p.parent
|
591
|
+
end
|
592
|
+
|
593
|
+
acc = []
|
594
|
+
p = preceding_node_of( node )
|
595
|
+
#puts "P = #{p.inspect}"
|
596
|
+
while p
|
597
|
+
if ancestors.include? p
|
598
|
+
ancestors.delete(p)
|
599
|
+
else
|
600
|
+
acc << p
|
601
|
+
end
|
602
|
+
p = preceding_node_of( p )
|
603
|
+
#puts "P = #{p.inspect}"
|
604
|
+
end
|
605
|
+
acc
|
606
|
+
end
|
607
|
+
|
608
|
+
def preceding_node_of( node )
|
609
|
+
#puts "NODE: #{node.inspect}"
|
610
|
+
#puts "PREVIOUS NODE: #{node.previous_sibling_node.inspect}"
|
611
|
+
#puts "PARENT NODE: #{node.parent}"
|
612
|
+
psn = node.previous_sibling_node
|
613
|
+
if psn.nil?
|
614
|
+
if node.parent.nil? or node.parent.class == Document
|
615
|
+
return nil
|
616
|
+
end
|
617
|
+
return node.parent
|
618
|
+
#psn = preceding_node_of( node.parent )
|
619
|
+
end
|
620
|
+
while psn and psn.kind_of? Element and psn.children.size > 0
|
621
|
+
psn = psn.children[-1]
|
622
|
+
end
|
623
|
+
psn
|
624
|
+
end
|
625
|
+
|
626
|
+
def following( node )
|
627
|
+
#puts "IN PRECEDING"
|
628
|
+
acc = []
|
629
|
+
p = next_sibling_node( node )
|
630
|
+
#puts "P = #{p.inspect}"
|
631
|
+
while p
|
632
|
+
acc << p
|
633
|
+
p = following_node_of( p )
|
634
|
+
#puts "P = #{p.inspect}"
|
635
|
+
end
|
636
|
+
acc
|
637
|
+
end
|
638
|
+
|
639
|
+
def following_node_of( node )
|
640
|
+
#puts "NODE: #{node.inspect}"
|
641
|
+
#puts "PREVIOUS NODE: #{node.previous_sibling_node.inspect}"
|
642
|
+
#puts "PARENT NODE: #{node.parent}"
|
643
|
+
if node.kind_of? Element and node.children.size > 0
|
644
|
+
return node.children[0]
|
645
|
+
end
|
646
|
+
return next_sibling_node(node)
|
647
|
+
end
|
648
|
+
|
649
|
+
def next_sibling_node(node)
|
650
|
+
psn = node.next_sibling_node
|
651
|
+
while psn.nil?
|
652
|
+
if node.parent.nil? or node.parent.class == Document
|
653
|
+
return nil
|
654
|
+
end
|
655
|
+
node = node.parent
|
656
|
+
psn = node.next_sibling_node
|
657
|
+
#puts "psn = #{psn.inspect}"
|
658
|
+
end
|
659
|
+
return psn
|
660
|
+
end
|
661
|
+
|
662
|
+
def norm b
|
663
|
+
case b
|
664
|
+
when true, false
|
665
|
+
return b
|
666
|
+
when 'true', 'false'
|
667
|
+
return Functions::boolean( b )
|
668
|
+
when /^\d+(\.\d+)?$/
|
669
|
+
return Functions::number( b )
|
670
|
+
else
|
671
|
+
return Functions::string( b )
|
672
|
+
end
|
673
|
+
end
|
674
|
+
|
675
|
+
def equality_relational_compare( set1, op, set2 )
|
676
|
+
#puts "EQ_REL_COMP(#{set1.inspect} #{op.inspect} #{set2.inspect})"
|
677
|
+
if set1.kind_of? Array and set2.kind_of? Array
|
678
|
+
#puts "#{set1.size} & #{set2.size}"
|
679
|
+
if set1.size == 1 and set2.size == 1
|
680
|
+
set1 = set1[0]
|
681
|
+
set2 = set2[0]
|
682
|
+
elsif set1.size == 0 or set2.size == 0
|
683
|
+
nd = set1.size==0 ? set2 : set1
|
684
|
+
rv = nd.collect { |il| compare( il, op, nil ) }
|
685
|
+
#puts "RV = #{rv.inspect}"
|
686
|
+
return rv
|
687
|
+
else
|
688
|
+
res = []
|
689
|
+
enum = SyncEnumerator.new( set1, set2 ).each { |i1, i2|
|
690
|
+
#puts "i1 = #{i1.inspect} (#{i1.class.name})"
|
691
|
+
#puts "i2 = #{i2.inspect} (#{i2.class.name})"
|
692
|
+
i1 = norm( i1 )
|
693
|
+
i2 = norm( i2 )
|
694
|
+
res << compare( i1, op, i2 )
|
695
|
+
}
|
696
|
+
return res
|
697
|
+
end
|
698
|
+
end
|
699
|
+
#puts "EQ_REL_COMP: #{set1.inspect} (#{set1.class.name}), #{op}, #{set2.inspect} (#{set2.class.name})"
|
700
|
+
#puts "COMPARING VALUES"
|
701
|
+
# If one is nodeset and other is number, compare number to each item
|
702
|
+
# in nodeset s.t. number op number(string(item))
|
703
|
+
# If one is nodeset and other is string, compare string to each item
|
704
|
+
# in nodeset s.t. string op string(item)
|
705
|
+
# If one is nodeset and other is boolean, compare boolean to each item
|
706
|
+
# in nodeset s.t. boolean op boolean(item)
|
707
|
+
if set1.kind_of? Array or set2.kind_of? Array
|
708
|
+
#puts "ISA ARRAY"
|
709
|
+
if set1.kind_of? Array
|
710
|
+
a = set1
|
711
|
+
b = set2
|
712
|
+
else
|
713
|
+
a = set2
|
714
|
+
b = set1
|
715
|
+
end
|
716
|
+
|
717
|
+
case b
|
718
|
+
when true, false
|
719
|
+
return a.collect {|v| compare( Functions::boolean(v), op, b ) }
|
720
|
+
when Numeric
|
721
|
+
return a.collect {|v| compare( Functions::number(v), op, b )}
|
722
|
+
when /^\d+(\.\d+)?$/
|
723
|
+
b = Functions::number( b )
|
724
|
+
#puts "B = #{b.inspect}"
|
725
|
+
return a.collect {|v| compare( Functions::number(v), op, b )}
|
726
|
+
else
|
727
|
+
#puts "Functions::string( #{b}(#{b.class.name}) ) = #{Functions::string(b)}"
|
728
|
+
b = Functions::string( b )
|
729
|
+
return a.collect { |v| compare( Functions::string(v), op, b ) }
|
730
|
+
end
|
731
|
+
else
|
732
|
+
# If neither is nodeset,
|
733
|
+
# If op is = or !=
|
734
|
+
# If either boolean, convert to boolean
|
735
|
+
# If either number, convert to number
|
736
|
+
# Else, convert to string
|
737
|
+
# Else
|
738
|
+
# Convert both to numbers and compare
|
739
|
+
s1 = set1.to_s
|
740
|
+
s2 = set2.to_s
|
741
|
+
#puts "EQ_REL_COMP: #{set1}=>#{s1}, #{set2}=>#{s2}"
|
742
|
+
if s1 == 'true' or s1 == 'false' or s2 == 'true' or s2 == 'false'
|
743
|
+
#puts "Functions::boolean(#{set1})=>#{Functions::boolean(set1)}"
|
744
|
+
#puts "Functions::boolean(#{set2})=>#{Functions::boolean(set2)}"
|
745
|
+
set1 = Functions::boolean( set1 )
|
746
|
+
set2 = Functions::boolean( set2 )
|
747
|
+
else
|
748
|
+
if op == :eq or op == :neq
|
749
|
+
if s1 =~ /^\d+(\.\d+)?$/ or s2 =~ /^\d+(\.\d+)?$/
|
750
|
+
set1 = Functions::number( s1 )
|
751
|
+
set2 = Functions::number( s2 )
|
752
|
+
else
|
753
|
+
set1 = Functions::string( set1 )
|
754
|
+
set2 = Functions::string( set2 )
|
755
|
+
end
|
756
|
+
else
|
757
|
+
set1 = Functions::number( set1 )
|
758
|
+
set2 = Functions::number( set2 )
|
759
|
+
end
|
760
|
+
end
|
761
|
+
#puts "EQ_REL_COMP: #{set1} #{op} #{set2}"
|
762
|
+
#puts ">>> #{compare( set1, op, set2 )}"
|
763
|
+
return compare( set1, op, set2 )
|
764
|
+
end
|
765
|
+
return false
|
766
|
+
end
|
767
|
+
|
768
|
+
def compare a, op, b
|
769
|
+
#puts "COMPARE #{a.inspect}(#{a.class.name}) #{op} #{b.inspect}(#{b.class.name})"
|
770
|
+
case op
|
771
|
+
when :eq
|
772
|
+
a == b
|
773
|
+
when :neq
|
774
|
+
a != b
|
775
|
+
when :lt
|
776
|
+
a < b
|
777
|
+
when :lteq
|
778
|
+
a <= b
|
779
|
+
when :gt
|
780
|
+
a > b
|
781
|
+
when :gteq
|
782
|
+
a >= b
|
783
|
+
when :and
|
784
|
+
a and b
|
785
|
+
when :or
|
786
|
+
a or b
|
787
|
+
else
|
788
|
+
false
|
789
|
+
end
|
790
|
+
end
|
791
|
+
end
|
792
|
+
end
|