-
-
Notifications
You must be signed in to change notification settings - Fork 141
/
semantic_tokenizer.cr
171 lines (143 loc) · 4.42 KB
/
semantic_tokenizer.cr
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
module Mint
class SemanticTokenizer
# This is a subset of the LSPs SemanticTokenTypes enum.
enum TokenType
TypeParameter
Type
Namespace
Property
Keyword
Comment
Variable
Operator
String
Number
Regexp
end
TOKEN_TYPES = TokenType.names.map!(&.camelcase(lower: true))
# This represents which token types are used for which node.
TOKEN_MAP = {
Ast::TypeVariable => TokenType::TypeParameter,
Ast::Comment => TokenType::Comment,
Ast::StringLiteral => TokenType::String,
Ast::RegexpLiteral => TokenType::Regexp,
Ast::NumberLiteral => TokenType::Number,
Ast::Id => TokenType::Type,
}
# Represents a semantic token using the positions of the token instead
# of line / column (for the LSP it is converted to line /column).
record Token,
type : TokenType,
from : Int64,
to : Int64
# We keep a cache of all tokenized nodes to avoid duplications
getter cache = Set(Ast::Node).new
# This is where the resulting tokens are stored.
getter tokens = [] of Token
def self.tokenize(ast : Ast)
tokenizer = self.new
tokenizer.tokenize(ast)
parts = [] of String | Tuple(SemanticTokenizer::TokenType, String)
contents = ast.nodes.first.file.contents
position = 0
tokenizer.tokens.sort_by(&.from).each do |token|
if token.from > position
parts << contents[position, token.from - position]
end
parts << {token.type, contents[token.from, token.to - token.from]}
position = token.to
end
if position < contents.size
parts << contents[position, contents.size]
end
parts
end
def self.highlight(path : String, html : Bool = false)
ast =
Parser.parse(path)
parts =
tokenize(ast)
parts.join do |item|
case item
in String
html ? HTML.escape(item) : item
in Tuple(SemanticTokenizer::TokenType, String)
if html
next "<span class=\"#{item[0].to_s.underscore}\">#{HTML.escape(item[1])}</span>"
end
case item[0]
in .type?
item[1].colorize(:yellow)
in .type_parameter?
item[1].colorize(:light_yellow)
in .variable?
item[1].colorize(:dark_gray)
in .namespace?
item[1].colorize(:light_blue)
in .keyword?
item[1].colorize(:magenta)
in .property?
item[1].colorize(:dark_gray).mode(:underline)
in .comment?
item[1].colorize(:light_gray)
in .string?
item[1].colorize(:green)
in .number?
item[1].colorize(:red)
in .regexp?
item[1].colorize(:light_red)
in .operator?
item[1].colorize(:light_magenta)
end.to_s
end
end
end
def tokenize(ast : Ast)
# We add the operators and keywords directly from the AST
ast.operators.each { |(from, to)| add(from, to, :operator) }
ast.keywords.each { |(from, to)| add(from, to, :keyword) }
tokenize(ast.nodes)
end
def tokenize(nodes : Array(Ast::Node))
nodes.each { |node| tokenize(node) }
end
def tokenize(node : Ast::Node?)
if type = TOKEN_MAP[node.class]?
add(node, type)
end
end
def tokenize(node : Ast::Variable)
if node.value[0].ascii_lowercase?
add(node, TokenType::Variable)
else
add(node, TokenType::Type)
end
end
def tokenize(node : Ast::CssDefinition)
add(node.from, node.from + node.name.size, :property)
end
def tokenize(node : Ast::HtmlElement)
# The closing tag is not saved only the position to it.
node.closing_tag_position.try do |position|
add(position, position + node.tag.value.size, :namespace)
end
add(node.tag, TokenType::Namespace)
end
def tokenize(node : Ast::HtmlComponent)
node.closing_tag_position.try do |position|
add(position, position + node.component.value.size, :type)
end
end
def add(from : Int64, to : Int64, type : TokenType)
tokens << Token.new(
type: type,
from: from,
to: to)
end
def add(node : Ast::Node, type : TokenType)
return if cache.includes?(node)
add(node.from, node.to, type)
cache.add(node)
end
end
end