hexp 0.0.1 → 0.2.0
Sign up to get free protection for your applications and to get access to all the features.
- data/.travis.yml +12 -3
- data/Changelog.md +9 -0
- data/Gemfile +3 -5
- data/Gemfile.devtools +20 -18
- data/Gemfile.lock +97 -84
- data/Rakefile +16 -0
- data/config/flay.yml +2 -2
- data/config/flog.yml +1 -1
- data/config/reek.yml +42 -18
- data/config/rubocop.yml +31 -0
- data/config/yardstick.yml +39 -1
- data/examples/from_nokogiri.rb +77 -0
- data/examples/selector_rewriter_chaining.rb +14 -0
- data/examples/todo.rb +138 -0
- data/examples/widget.rb +64 -0
- data/hexp.gemspec +8 -3
- data/lib/hexp.rb +103 -2
- data/lib/hexp/builder.rb +256 -0
- data/lib/hexp/css_selector.rb +205 -0
- data/lib/hexp/css_selector/parser.rb +74 -0
- data/lib/hexp/css_selector/sass_parser.rb +22 -0
- data/lib/hexp/dom.rb +0 -2
- data/lib/hexp/dsl.rb +27 -0
- data/lib/hexp/errors.rb +21 -0
- data/lib/hexp/h.rb +5 -2
- data/lib/hexp/list.rb +67 -9
- data/lib/hexp/node.rb +197 -41
- data/lib/hexp/node/attributes.rb +176 -0
- data/lib/hexp/node/children.rb +44 -0
- data/lib/hexp/node/css_selection.rb +73 -0
- data/lib/hexp/node/domize.rb +52 -6
- data/lib/hexp/node/normalize.rb +19 -9
- data/lib/hexp/node/pp.rb +32 -0
- data/lib/hexp/node/rewriter.rb +52 -0
- data/lib/hexp/node/selector.rb +59 -0
- data/lib/hexp/nokogiri/equality.rb +61 -0
- data/lib/hexp/nokogiri/reader.rb +27 -0
- data/lib/hexp/sass/selector_parser.rb +4 -0
- data/lib/hexp/text_node.rb +129 -9
- data/lib/hexp/version.rb +1 -1
- data/notes +34 -0
- data/spec/shared_helper.rb +6 -0
- data/spec/spec_helper.rb +2 -6
- data/spec/unit/hexp/builder_spec.rb +101 -0
- data/spec/unit/hexp/css_selector/attribute_spec.rb +137 -0
- data/spec/unit/hexp/css_selector/class_spec.rb +15 -0
- data/spec/unit/hexp/css_selector/comma_sequence_spec.rb +20 -0
- data/spec/unit/hexp/css_selector/element_spec.rb +11 -0
- data/spec/unit/hexp/css_selector/parser_spec.rb +51 -0
- data/spec/unit/hexp/css_selector/simple_sequence_spec.rb +48 -0
- data/spec/unit/hexp/dsl_spec.rb +55 -0
- data/spec/unit/hexp/h_spec.rb +38 -0
- data/spec/unit/hexp/list_spec.rb +19 -0
- data/spec/unit/hexp/node/attr_spec.rb +55 -0
- data/spec/unit/hexp/node/attributes_spec.rb +125 -0
- data/spec/unit/hexp/node/children_spec.rb +33 -0
- data/spec/unit/hexp/node/class_spec.rb +37 -0
- data/spec/unit/hexp/node/css_selection_spec.rb +86 -0
- data/spec/unit/hexp/node/normalize_spec.rb +12 -6
- data/spec/unit/hexp/node/rewrite_spec.rb +67 -30
- data/spec/unit/hexp/node/selector_spec.rb +78 -0
- data/spec/unit/hexp/node/text_spec.rb +7 -0
- data/spec/unit/hexp/node/to_dom_spec.rb +1 -1
- data/spec/unit/hexp/nokogiri/reader_spec.rb +8 -0
- data/spec/unit/hexp/parse_spec.rb +23 -0
- data/spec/unit/hexp/text_node_spec.rb +25 -0
- data/spec/unit/hexp_spec.rb +33 -0
- metadata +129 -16
- data/lib/hexp/format_error.rb +0 -8
@@ -0,0 +1,176 @@
|
|
1
|
+
module Hexp
|
2
|
+
class Node
|
3
|
+
# Node API methods that deal with attributes
|
4
|
+
#
|
5
|
+
module Attributes
|
6
|
+
# Attribute getter/setter
|
7
|
+
#
|
8
|
+
# When called with one argument : return the attribute value with that name.
|
9
|
+
# When called with two arguments : return a new Node with the attribute set.
|
10
|
+
# When the second argument is nil : return a new Node with the attribute unset.
|
11
|
+
#
|
12
|
+
# @example
|
13
|
+
# H[:p, class: 'hello'].attr('class') # => "hello"
|
14
|
+
# H[:p, class: 'hello'].attr('id', 'para1') # => H[:p, {"class"=>"hello", "id"=>"para1"}]
|
15
|
+
# H[:p, class: 'hello'].attr('class', nil) # => H[:p]
|
16
|
+
#
|
17
|
+
# @return [String|Hexp::Node]
|
18
|
+
# @api private
|
19
|
+
#
|
20
|
+
def attr(*args)
|
21
|
+
arity = args.count
|
22
|
+
attr_name = args[0].to_s
|
23
|
+
|
24
|
+
case arity
|
25
|
+
when 1
|
26
|
+
attributes[attr_name]
|
27
|
+
when 2
|
28
|
+
set_attr(*args)
|
29
|
+
else
|
30
|
+
raise ArgumentError, "wrong number of arguments(#{arity} for 1..2)"
|
31
|
+
end
|
32
|
+
end
|
33
|
+
|
34
|
+
# Is an attribute present
|
35
|
+
#
|
36
|
+
# This will also return true if the attribute is present but empty.
|
37
|
+
#
|
38
|
+
# @example
|
39
|
+
# H[:option].has_attr?('selected') #=> false
|
40
|
+
#
|
41
|
+
# @param name [String|Symbol] the name of the attribute
|
42
|
+
# @return [Boolean]
|
43
|
+
# @api public
|
44
|
+
#
|
45
|
+
def has_attr?(name)
|
46
|
+
attributes.has_key? name.to_s
|
47
|
+
end
|
48
|
+
|
49
|
+
# Check for the presence of a class
|
50
|
+
#
|
51
|
+
# @example
|
52
|
+
# H[:span, class: "banner strong"].class?("strong") #=> true
|
53
|
+
#
|
54
|
+
# @param klass [String] the name of the class to check for
|
55
|
+
# @return [Boolean] true if the class is present, false otherwise
|
56
|
+
# @api public
|
57
|
+
#
|
58
|
+
def class?(klass)
|
59
|
+
attr('class') && attr('class').split(' ').include?(klass.to_s)
|
60
|
+
end
|
61
|
+
|
62
|
+
# Add a CSS class to the element
|
63
|
+
#
|
64
|
+
# @example
|
65
|
+
# H[:div].add_class('foo') #=> H[:div, class: 'foo']
|
66
|
+
#
|
67
|
+
# @param klass [#to_s] The class to add
|
68
|
+
# @return [Hexp::Node]
|
69
|
+
# @api public
|
70
|
+
#
|
71
|
+
def add_class(klass)
|
72
|
+
attr('class', [attr('class'), klass].compact.join(' '))
|
73
|
+
end
|
74
|
+
|
75
|
+
# The CSS classes of this element as an array
|
76
|
+
#
|
77
|
+
# Convenience method so you don't have to split the class list yourself.
|
78
|
+
#
|
79
|
+
# @return [Array<String>]
|
80
|
+
# @api public
|
81
|
+
#
|
82
|
+
def class_list
|
83
|
+
@class_list ||= (attr('class') || '').split(' ').freeze
|
84
|
+
end
|
85
|
+
|
86
|
+
# Remove a CSS class from this element
|
87
|
+
#
|
88
|
+
# If the resulting class list is empty, the class attribute will be
|
89
|
+
# removed. If the class is present several times all instances will
|
90
|
+
# be removed. If it's not present at all, the class list will be
|
91
|
+
# unmodified.
|
92
|
+
#
|
93
|
+
# Calling this on a node with a class attribute that is equal to an
|
94
|
+
# empty string will result in the class attribute being removed.
|
95
|
+
#
|
96
|
+
# @param klass [#to_s] The class to be removed
|
97
|
+
# @return [Hexp::Node] A node that is identical to this one, but with
|
98
|
+
# the given class removed
|
99
|
+
# @api public
|
100
|
+
#
|
101
|
+
def remove_class(klass)
|
102
|
+
return self unless has_attr?('class')
|
103
|
+
new_list = class_list - [klass.to_s]
|
104
|
+
return remove_attr('class') if new_list.empty?
|
105
|
+
attr('class', new_list.join(' '))
|
106
|
+
end
|
107
|
+
|
108
|
+
# Set or override multiple attributes using a hash syntax
|
109
|
+
#
|
110
|
+
# @param attrs[Hash]
|
111
|
+
# @return [Hexp::Node]
|
112
|
+
# @api public
|
113
|
+
#
|
114
|
+
def set_attrs(attrs)
|
115
|
+
H[
|
116
|
+
self.tag,
|
117
|
+
self.attributes.merge(Hash[*attrs.flat_map{|k,v| [k.to_s, v]}]),
|
118
|
+
self.children
|
119
|
+
]
|
120
|
+
end
|
121
|
+
alias :% :set_attrs
|
122
|
+
alias :add_attributes :set_attrs
|
123
|
+
|
124
|
+
# Remove an attribute by name
|
125
|
+
#
|
126
|
+
# @param name [#to_s] The attribute to be removed
|
127
|
+
# @return [Hexp::Node] a new node with the attribute removed
|
128
|
+
# @api public
|
129
|
+
#
|
130
|
+
def remove_attr(name)
|
131
|
+
H[
|
132
|
+
self.tag,
|
133
|
+
self.attributes.reject {|key,_| key == name.to_s},
|
134
|
+
self.children
|
135
|
+
]
|
136
|
+
end
|
137
|
+
|
138
|
+
# Attribute accessor
|
139
|
+
#
|
140
|
+
# @param attribute_name [#to_s] The name of the attribute
|
141
|
+
# @return [String] The value of the attribute
|
142
|
+
# @api public
|
143
|
+
#
|
144
|
+
def [](attr_name)
|
145
|
+
self.attributes[attr_name.to_s]
|
146
|
+
end
|
147
|
+
|
148
|
+
# Merge attributes into this Hexp
|
149
|
+
#
|
150
|
+
# Class attributes are treated special : the class lists are merged, rather
|
151
|
+
# than being overwritten. See {set_attrs} for a more basic version.
|
152
|
+
#
|
153
|
+
# This method is analoguous with {Hash#merge}. As argument it can take a
|
154
|
+
# Hash, or another Hexp element, in which case that element's attributes
|
155
|
+
# are used.
|
156
|
+
#
|
157
|
+
# @param node_or_hash [#to_hexp|Hash]
|
158
|
+
# @return [Hexp::Node]
|
159
|
+
# @api public
|
160
|
+
#
|
161
|
+
def merge_attrs(node_or_hash)
|
162
|
+
hash = node_or_hash.respond_to?(:to_hexp) ?
|
163
|
+
node_or_hash.to_hexp.attributes : node_or_hash
|
164
|
+
result = self
|
165
|
+
hash.each do |key,value|
|
166
|
+
result = if key.to_s == 'class'
|
167
|
+
result.add_class(value)
|
168
|
+
else
|
169
|
+
result.attr(key, value)
|
170
|
+
end
|
171
|
+
end
|
172
|
+
result
|
173
|
+
end
|
174
|
+
end
|
175
|
+
end
|
176
|
+
end
|
@@ -0,0 +1,44 @@
|
|
1
|
+
module Hexp
|
2
|
+
class Node
|
3
|
+
# Node API methods that deal with child_nodes
|
4
|
+
#
|
5
|
+
module Children
|
6
|
+
# Is this node an empty node
|
7
|
+
#
|
8
|
+
# H[:p, class: 'foo'].empty? #=> true
|
9
|
+
# H[:p, [H[:span]].empty? #=> false
|
10
|
+
#
|
11
|
+
# @return [Boolean] true if this node has no children
|
12
|
+
# @api public
|
13
|
+
#
|
14
|
+
def empty?
|
15
|
+
children.empty?
|
16
|
+
end
|
17
|
+
|
18
|
+
def add_child(child)
|
19
|
+
H[
|
20
|
+
self.tag,
|
21
|
+
self.attributes,
|
22
|
+
self.children + [child]
|
23
|
+
]
|
24
|
+
end
|
25
|
+
alias :add :add_child
|
26
|
+
alias :<< :add_child
|
27
|
+
|
28
|
+
def text
|
29
|
+
children.map do |node|
|
30
|
+
node.text? ? node : node.text
|
31
|
+
end.join
|
32
|
+
end
|
33
|
+
|
34
|
+
def set_children(new_children)
|
35
|
+
H[tag, attributes, new_children]
|
36
|
+
end
|
37
|
+
|
38
|
+
def map_children(&blk)
|
39
|
+
return to_enum(:map_children) unless block_given?
|
40
|
+
H[tag, attributes, children.map(&blk)]
|
41
|
+
end
|
42
|
+
end
|
43
|
+
end
|
44
|
+
end
|
@@ -0,0 +1,73 @@
|
|
1
|
+
module Hexp
|
2
|
+
class Node
|
3
|
+
# Select nodes using CSS selectors
|
4
|
+
#
|
5
|
+
class CssSelection < Selector
|
6
|
+
def initialize(node, css_selector)
|
7
|
+
@node, @css_selector = node, css_selector
|
8
|
+
end
|
9
|
+
|
10
|
+
def inspect
|
11
|
+
"#<#{self.class} @node=#{@node.inspect} @css_selector=#{@css_selector.inspect} matches=#{node_matches?}>"
|
12
|
+
end
|
13
|
+
|
14
|
+
def each(&block)
|
15
|
+
return to_enum(:each) unless block_given?
|
16
|
+
|
17
|
+
@node.children.each do |child|
|
18
|
+
next_selection_for(child).each(&block)
|
19
|
+
end
|
20
|
+
yield @node if node_matches?
|
21
|
+
end
|
22
|
+
|
23
|
+
def rewrite(&block)
|
24
|
+
return @node if @node.text?
|
25
|
+
|
26
|
+
new_node = H[
|
27
|
+
@node.tag,
|
28
|
+
@node.attributes,
|
29
|
+
@node.children.flat_map do |child|
|
30
|
+
next_selection_for(child).rewrite(&block)
|
31
|
+
end
|
32
|
+
]
|
33
|
+
node_matches? ? block.call(new_node) : new_node
|
34
|
+
end
|
35
|
+
|
36
|
+
private
|
37
|
+
|
38
|
+
def comma_sequence
|
39
|
+
@comma_sequence ||= coerce_to_comma_sequence(@css_selector)
|
40
|
+
end
|
41
|
+
|
42
|
+
def coerce_to_comma_sequence(css_selector)
|
43
|
+
return css_selector if css_selector.is_a? CssSelector::CommaSequence
|
44
|
+
CssSelector::Parser.call(@css_selector)
|
45
|
+
end
|
46
|
+
|
47
|
+
def node_matches?
|
48
|
+
comma_sequence.matches?(@node)
|
49
|
+
end
|
50
|
+
|
51
|
+
# returns a new commasequence with the parts removed that have been consumed by matching
|
52
|
+
# against this node. If no part matches, return nil
|
53
|
+
def next_comma_sequence
|
54
|
+
@next_comma_sequence ||= CssSelector::CommaSequence.new(consume_matching_heads)
|
55
|
+
end
|
56
|
+
|
57
|
+
def next_selection_for(child)
|
58
|
+
self.class.new(child, next_comma_sequence)
|
59
|
+
end
|
60
|
+
|
61
|
+
def consume_matching_heads
|
62
|
+
comma_sequence.members.flat_map do |sequence|
|
63
|
+
if sequence.head_matches? @node
|
64
|
+
[sequence, sequence.drop_head]
|
65
|
+
else
|
66
|
+
[sequence]
|
67
|
+
end
|
68
|
+
end.reject(&:empty?)
|
69
|
+
end
|
70
|
+
|
71
|
+
end
|
72
|
+
end
|
73
|
+
end
|
data/lib/hexp/node/domize.rb
CHANGED
@@ -2,22 +2,54 @@ module Hexp
|
|
2
2
|
class Node
|
3
3
|
# Turn nodes into DOM objects
|
4
4
|
class Domize
|
5
|
+
DEFAULT_OPTIONS = {
|
6
|
+
:include_doctype => true
|
7
|
+
}.freeze
|
8
|
+
|
9
|
+
# The resulting DOM Document
|
10
|
+
#
|
11
|
+
# @return [Nokogiri::HTML::Document]
|
12
|
+
# @api private
|
13
|
+
#
|
5
14
|
attr_reader :dom
|
6
15
|
|
7
|
-
|
8
|
-
|
9
|
-
|
16
|
+
# Instanitiate a Domizer
|
17
|
+
#
|
18
|
+
# @param hexp [Hexp::Node]
|
19
|
+
# @param options [Hash] :include_doctype defaults to true
|
20
|
+
# @api private
|
21
|
+
#
|
22
|
+
def initialize(hexp, options = {})
|
23
|
+
@dom = Hexp::DOM
|
24
|
+
@raw = hexp
|
25
|
+
@options = DEFAULT_OPTIONS.merge(options).freeze
|
10
26
|
end
|
11
27
|
|
28
|
+
# Turn the hexp into a DOM
|
29
|
+
#
|
30
|
+
# @return [Nokogiri::HTML::Document]
|
31
|
+
# @api private
|
32
|
+
#
|
12
33
|
def call
|
13
|
-
dom::Document.new
|
14
|
-
|
15
|
-
|
34
|
+
@doc = dom::Document.new
|
35
|
+
@root = domize(@raw)
|
36
|
+
@doc << @root
|
37
|
+
|
38
|
+
if @options[:include_doctype]
|
39
|
+
@doc
|
40
|
+
else
|
41
|
+
@root
|
16
42
|
end
|
17
43
|
end
|
18
44
|
|
19
45
|
private
|
20
46
|
|
47
|
+
# Turn a Hexp::Node into a Document
|
48
|
+
#
|
49
|
+
# @param hexp [Hexp::Node]
|
50
|
+
# @return [Nokogiri::HTML::Document]
|
51
|
+
# @api private
|
52
|
+
#
|
21
53
|
def domize(hexp)
|
22
54
|
dom::Node.new(hexp.tag.to_s, @doc).tap do |node|
|
23
55
|
set_attributes(node, hexp.attributes)
|
@@ -25,12 +57,26 @@ module Hexp
|
|
25
57
|
end
|
26
58
|
end
|
27
59
|
|
60
|
+
# Set attributes on a DOM node
|
61
|
+
#
|
62
|
+
# @param node [Nokogiri::XML::Element]
|
63
|
+
# @param attributes [Hash]
|
64
|
+
# @return [void]
|
65
|
+
# @api private
|
66
|
+
#
|
28
67
|
def set_attributes(node, attributes)
|
29
68
|
attributes.each do |key,value|
|
30
69
|
node[key] = value
|
31
70
|
end
|
32
71
|
end
|
33
72
|
|
73
|
+
# Set children on the DOM node
|
74
|
+
#
|
75
|
+
# @param node [Nokogiri::XML::Element]
|
76
|
+
# @param children [Hexp::List]
|
77
|
+
# @return [void]
|
78
|
+
# @api private
|
79
|
+
#
|
34
80
|
def set_children(node, children)
|
35
81
|
children.each do |child|
|
36
82
|
if child.instance_of?(TextNode)
|
data/lib/hexp/node/normalize.rb
CHANGED
@@ -40,6 +40,12 @@ module Hexp
|
|
40
40
|
{}
|
41
41
|
end
|
42
42
|
|
43
|
+
# Returns the attributes hash with key and value converted to strings
|
44
|
+
#
|
45
|
+
# @return [Hash]
|
46
|
+
#
|
47
|
+
# @api private
|
48
|
+
#
|
43
49
|
def normalized_attributes
|
44
50
|
Hash[*
|
45
51
|
attributes.flat_map do |key, value|
|
@@ -55,10 +61,14 @@ module Hexp
|
|
55
61
|
# @api private
|
56
62
|
#
|
57
63
|
def children
|
58
|
-
@raw
|
59
|
-
|
64
|
+
last = @raw.last
|
65
|
+
if last.respond_to? :to_ary
|
66
|
+
last.to_ary
|
67
|
+
elsif @raw.count < 2 || last.instance_of?(Hash)
|
68
|
+
[]
|
69
|
+
else
|
70
|
+
[last]
|
60
71
|
end
|
61
|
-
[]
|
62
72
|
end
|
63
73
|
|
64
74
|
# Normalize the third element of a hexp node, the list of children
|
@@ -75,14 +85,14 @@ module Hexp
|
|
75
85
|
child
|
76
86
|
when String, TextNode
|
77
87
|
Hexp::TextNode.new(child)
|
88
|
+
when ->(ch) { ch.respond_to? :to_hexp }
|
89
|
+
response = child.to_hexp
|
90
|
+
raise FormatError, "to_hexp must return a Hexp::Node, got #{response.inspect}" unless response.instance_of?(Hexp::Node)
|
91
|
+
response
|
78
92
|
when Array
|
79
|
-
Hexp::Node[*child]
|
93
|
+
Hexp::Node[*child.map(&:freeze)]
|
80
94
|
else
|
81
|
-
|
82
|
-
response = child.to_hexp
|
83
|
-
raise FormatError, "to_hexp must return a Hexp::Node, got #{response.inspect}" unless response.instance_of?(Hexp::Node)
|
84
|
-
response
|
85
|
-
end
|
95
|
+
raise FormatError, "Invalid value in Hexp literal : #{child.inspect} (#{child.class}) does not implement #to_hexp ; #{children.inspect}"
|
86
96
|
end
|
87
97
|
end
|
88
98
|
]
|
data/lib/hexp/node/pp.rb
CHANGED
@@ -2,10 +2,20 @@ module Hexp
|
|
2
2
|
class Node
|
3
3
|
# Pretty-print a node and its contents
|
4
4
|
class PP
|
5
|
+
# Create a new pretty-printer
|
6
|
+
#
|
7
|
+
# @param node [Hexp::Node] The node to represent
|
8
|
+
# @api private
|
9
|
+
#
|
5
10
|
def initialize(node)
|
6
11
|
@node = node
|
7
12
|
end
|
8
13
|
|
14
|
+
# Perform the pretty-printing
|
15
|
+
#
|
16
|
+
# @return [String] The pp output
|
17
|
+
# @api private
|
18
|
+
#
|
9
19
|
def call
|
10
20
|
[
|
11
21
|
@node.class.inspect_name,
|
@@ -14,22 +24,44 @@ module Hexp
|
|
14
24
|
].join
|
15
25
|
end
|
16
26
|
|
27
|
+
# Format the node tag
|
28
|
+
#
|
29
|
+
# @return [String]
|
30
|
+
# @api private
|
31
|
+
#
|
17
32
|
def pp_tag
|
18
33
|
"[#{@node.tag.inspect}"
|
19
34
|
end
|
20
35
|
|
36
|
+
# Format the node attributes
|
37
|
+
#
|
38
|
+
# @return [String]
|
39
|
+
# @api private
|
40
|
+
#
|
21
41
|
def pp_attributes
|
22
42
|
attrs = @node.attributes
|
23
43
|
return '' if attrs.empty?
|
24
44
|
', ' + attrs.inspect
|
25
45
|
end
|
26
46
|
|
47
|
+
# Format the node children
|
48
|
+
#
|
49
|
+
# @return [String]
|
50
|
+
# @api private
|
51
|
+
#
|
27
52
|
def pp_children
|
28
53
|
children = @node.children
|
29
54
|
return ']' if children.empty?
|
30
55
|
", [\n#{ children.map(&:pp).join(",\n") }]]"
|
31
56
|
end
|
32
57
|
|
58
|
+
# Indent a multiline string with a number of spaces
|
59
|
+
#
|
60
|
+
# @param string [String] The string to indent
|
61
|
+
# @param indent [Integer] The number of spaces to use for indentation
|
62
|
+
# @return [String]
|
63
|
+
# @api private
|
64
|
+
#
|
33
65
|
def self.indent(string, indent = 2)
|
34
66
|
string.lines.map {|line| " "*indent + line}.join
|
35
67
|
end
|