1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
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
138
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
203
204
205
206
207
208
209
210
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
|
class Nokogiri::CSS::Parser
token FUNCTION INCLUDES DASHMATCH LBRACE HASH PLUS GREATER S STRING IDENT
token COMMA NUMBER PREFIXMATCH SUFFIXMATCH SUBSTRINGMATCH TILDE NOT_EQUAL
token SLASH DOUBLESLASH NOT EQUAL RPAREN LSQUARE RSQUARE HAS
rule
selector
: selector COMMA simple_selector_1toN {
result = [val.first, val.last].flatten
}
| prefixless_combinator_selector { result = val.flatten }
| optional_S simple_selector_1toN { result = [val.last].flatten }
;
combinator
: PLUS { result = :DIRECT_ADJACENT_SELECTOR }
| GREATER { result = :CHILD_SELECTOR }
| TILDE { result = :FOLLOWING_SELECTOR }
| DOUBLESLASH { result = :DESCENDANT_SELECTOR }
| SLASH { result = :CHILD_SELECTOR }
;
simple_selector
: element_name hcap_0toN {
result = if val[1].nil?
val.first
else
Node.new(:CONDITIONAL_SELECTOR, [val.first, val[1]])
end
}
| function
| function pseudo {
result = Node.new(:CONDITIONAL_SELECTOR, val)
}
| function attrib {
result = Node.new(:CONDITIONAL_SELECTOR, val)
}
| hcap_1toN {
result = Node.new(:CONDITIONAL_SELECTOR,
[Node.new(:ELEMENT_NAME, ['*']), val.first]
)
}
;
prefixless_combinator_selector
: combinator simple_selector_1toN {
result = Node.new(val.first, [nil, val.last])
}
;
simple_selector_1toN
: simple_selector combinator simple_selector_1toN {
result = Node.new(val[1], [val.first, val.last])
}
| simple_selector S simple_selector_1toN {
result = Node.new(:DESCENDANT_SELECTOR, [val.first, val.last])
}
| simple_selector
;
class
: '.' IDENT { result = Node.new(:CLASS_CONDITION, [val[1]]) }
;
element_name
: namespaced_ident
| '*' { result = Node.new(:ELEMENT_NAME, val) }
;
namespaced_ident
: namespace '|' IDENT {
result = Node.new(:ELEMENT_NAME,
[[val.first, val.last].compact.join(':')]
)
}
| IDENT {
name = @namespaces.key?('xmlns') ? "xmlns:#{val.first}" : val.first
result = Node.new(:ELEMENT_NAME, [name])
}
;
namespace
: IDENT { result = val[0] }
|
;
attrib
: LSQUARE attrib_name attrib_val_0or1 RSQUARE {
result = Node.new(:ATTRIBUTE_CONDITION,
[val[1]] + (val[2] || [])
)
}
| LSQUARE function attrib_val_0or1 RSQUARE {
result = Node.new(:ATTRIBUTE_CONDITION,
[val[1]] + (val[2] || [])
)
}
| LSQUARE NUMBER RSQUARE {
# Non standard, but hpricot supports it.
result = Node.new(:PSEUDO_CLASS,
[Node.new(:FUNCTION, ['nth-child(', val[1]])]
)
}
;
attrib_name
: namespace '|' IDENT {
result = Node.new(:ELEMENT_NAME,
[[val.first, val.last].compact.join(':')]
)
}
| IDENT {
# Default namespace is not applied to attributes.
# So we don't add prefix "xmlns:" as in namespaced_ident.
result = Node.new(:ELEMENT_NAME, [val.first])
}
;
function
: FUNCTION RPAREN {
result = Node.new(:FUNCTION, [val.first.strip])
}
| FUNCTION expr RPAREN {
result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
}
| FUNCTION nth RPAREN {
result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
}
| NOT expr RPAREN {
result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
}
| HAS selector RPAREN {
result = Node.new(:FUNCTION, [val.first.strip, val[1]].flatten)
}
;
expr
: NUMBER COMMA expr { result = [val.first, val.last] }
| STRING COMMA expr { result = [val.first, val.last] }
| IDENT COMMA expr { result = [val.first, val.last] }
| NUMBER
| STRING
| IDENT # even, odd
{
case val[0]
when 'even'
result = Node.new(:NTH, ['2','n','+','0'])
when 'odd'
result = Node.new(:NTH, ['2','n','+','1'])
when 'n'
result = Node.new(:NTH, ['1','n','+','0'])
else
# This is not CSS standard. It allows us to support this:
# assert_xpath("//a[foo(., @href)]", @parser.parse('a:foo(@href)'))
# assert_xpath("//a[foo(., @a, b)]", @parser.parse('a:foo(@a, b)'))
# assert_xpath("//a[foo(., a, 10)]", @parser.parse('a:foo(a, 10)'))
result = val
end
}
;
nth
: NUMBER IDENT PLUS NUMBER # 5n+3 -5n+3
{
if val[1] == 'n'
result = Node.new(:NTH, val)
else
raise Racc::ParseError, "parse error on IDENT '#{val[1]}'"
end
}
| IDENT PLUS NUMBER { # n+3, -n+3
if val[0] == 'n'
val.unshift("1")
result = Node.new(:NTH, val)
elsif val[0] == '-n'
val[0] = 'n'
val.unshift("-1")
result = Node.new(:NTH, val)
else
raise Racc::ParseError, "parse error on IDENT '#{val[1]}'"
end
}
| NUMBER IDENT { # 5n, -5n, 10n-1
n = val[1]
if n[0, 2] == 'n-'
val[1] = 'n'
val << "-"
# b is contained in n as n is the string "n-b"
val << n[2, n.size]
result = Node.new(:NTH, val)
elsif n == 'n'
val << "+"
val << "0"
result = Node.new(:NTH, val)
else
raise Racc::ParseError, "parse error on IDENT '#{val[1]}'"
end
}
;
pseudo
: ':' function {
result = Node.new(:PSEUDO_CLASS, [val[1]])
}
| ':' IDENT { result = Node.new(:PSEUDO_CLASS, [val[1]]) }
;
hcap_0toN
: hcap_1toN
|
;
hcap_1toN
: attribute_id hcap_1toN {
result = Node.new(:COMBINATOR, val)
}
| class hcap_1toN {
result = Node.new(:COMBINATOR, val)
}
| attrib hcap_1toN {
result = Node.new(:COMBINATOR, val)
}
| pseudo hcap_1toN {
result = Node.new(:COMBINATOR, val)
}
| negation hcap_1toN {
result = Node.new(:COMBINATOR, val)
}
| attribute_id
| class
| attrib
| pseudo
| negation
;
attribute_id
: HASH { result = Node.new(:ID, val) }
;
attrib_val_0or1
: eql_incl_dash IDENT { result = [val.first, val[1]] }
| eql_incl_dash STRING { result = [val.first, val[1]] }
|
;
eql_incl_dash
: EQUAL { result = :equal }
| PREFIXMATCH { result = :prefix_match }
| SUFFIXMATCH { result = :suffix_match }
| SUBSTRINGMATCH { result = :substring_match }
| NOT_EQUAL { result = :not_equal }
| INCLUDES { result = :includes }
| DASHMATCH { result = :dash_match }
;
negation
: NOT negation_arg RPAREN {
result = Node.new(:NOT, [val[1]])
}
;
negation_arg
: element_name
| element_name hcap_1toN
| hcap_1toN
;
optional_S
: S
|
;
end
---- header
require 'nokogiri/css/parser_extras'
|
