Module: JSONP3::Path

Defined in:
lib/json_p3.rb,
lib/json_p3/errors.rb,
lib/json_p3/path/node.rb,
lib/json_p3/path/lexer.rb,
lib/json_p3/path/query.rb,
lib/json_p3/path/filter.rb,
lib/json_p3/path/parser.rb,
lib/json_p3/path/segment.rb,
lib/json_p3/path/function.rb,
lib/json_p3/path/selector.rb,
lib/json_p3/path/unescape.rb,
lib/json_p3/path/serialize.rb,
lib/json_p3/path/environment.rb,
lib/json_p3/path/function_extensions/count.rb,
lib/json_p3/path/function_extensions/match.rb,
lib/json_p3/path/function_extensions/value.rb,
lib/json_p3/path/function_extensions/length.rb,
lib/json_p3/path/function_extensions/search.rb,
lib/json_p3/path/function_extensions/pattern.rb

Overview

JSONPath query expressions.

Defined Under Namespace

Classes: AbsoluteQueryExpression, BooleanLiteral, ChildSegment, Count, DescendantSegment, Environment, EqExpression, Error, Expression, FilterContext, FilterExpression, FilterExpressionLiteral, FilterSelector, FloatLiteral, FunctionExpression, FunctionExtension, GeExpression, GtExpression, IndexSelector, InfixExpression, IntegerLiteral, LeExpression, Length, LogicalAndExpression, LogicalNotExpression, LogicalOrExpression, LtExpression, Match, NameError, NameSelector, NeExpression, Node, NodeList, NullLiteral, Parser, Query, QueryExpression, RecursionError, RelativeQueryExpression, Search, Segment, Selector, SliceSelector, StringLiteral, SymbolNameSelector, SyntaxError, TypeError, Value, WildcardSelector

Constant Summary collapse

DefaultEnvironment = 
JSONP3::Path::Environment.new
RE_FLOAT = 
/\G(?:-?\d+\.\d+(?:[eE][+-]?\d+)?)|(-?\d+[eE]-\d+)/
RE_INDEX = 
/\G-?\d+/
RE_INT = 
/\G-?\d+[eE]\+?\d+/
RE_SLASH_U = 
/\\u([0-9a-fA-F]{4})/
TRANS = 
{ "\\\"" => "\"", "'" => "\\'" }.freeze

Class Method Summary collapse

Class Method Details

.canonical_string(value) ⇒ Object

Return value formatted as a canonical string literal.

Parameters:

  • value (String) 


12
13
14
 
# File 'lib/json_p3/path/serialize.rb', line 12

def self.canonical_string(value)
  "'#{(JSON.dump(value)[1..-2] || raise).gsub(/('|\\")/, TRANS)}'"
end

.compile(path) ⇒ Object 



32
33
34
 
# File 'lib/json_p3.rb', line 32

def self.compile(path)
  DefaultEnvironment.compile(path)
end

.eq?(left, right) ⇒ Boolean

Returns:

  • (Boolean) 


423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
 
# File 'lib/json_p3/path/filter.rb', line 423

def self.eq?(left, right)
  left = left.first.value if left.is_a?(NodeList) && left.length == 1
  right = right.first.value if right.is_a?(NodeList) && right.length == 1

  right, left = left, right if right.is_a?(NodeList)

  if left.is_a? NodeList
    return left == right if right.is_a? NodeList
    return right == :nothing if left.empty?
    return left.first == right if left.length == 1

    return false
  end

  return true if left == :nothing && right == :nothing

  left == right
end

.find(path, data) ⇒ Object 



24
25
26
 
# File 'lib/json_p3.rb', line 24

def self.find(path, data)
  DefaultEnvironment.find(path, data)
end

.find_enum(path, data) ⇒ Object 



28
29
30
 
# File 'lib/json_p3.rb', line 28

def self.find_enum(path, data)
  DefaultEnvironment.find_enum(path, data)
end

.first(path, data) ⇒ Object 



44
45
46
 
# File 'lib/json_p3.rb', line 44

def self.first(path, data)
  DefaultEnvironment.first(path, data)
end

.get_token_value(token, query) ⇒ Object 



216
217
218
 
# File 'lib/json_p3/path/lexer.rb', line 216

def self.get_token_value(token, query)
  query.byteslice(token[1], token.last - token[1]) || raise
end

.high_surrogate?(code_point) ⇒ Boolean

Returns:

  • (Boolean) 


126
127
128
 
# File 'lib/json_p3/path/unescape.rb', line 126

def self.high_surrogate?(code_point)
  code_point.between?(0xD800, 0xDBFF)
end

.low_surrogate?(code_point) ⇒ Boolean

Returns:

  • (Boolean) 


130
131
132
 
# File 'lib/json_p3/path/unescape.rb', line 130

def self.low_surrogate?(code_point)
  code_point.between?(0xDC00, 0xDFFF)
end

.lt?(left, right) ⇒ Boolean

Returns:

  • (Boolean) 


442
443
444
445
446
447
448
449
450
 
# File 'lib/json_p3/path/filter.rb', line 442

def self.lt?(left, right)
  left = left.first.value if left.is_a?(NodeList) && left.length == 1
  right = right.first.value if right.is_a?(NodeList) && right.length == 1
  return left < right if left.is_a?(String) && right.is_a?(String)
  return left < right if (left.is_a?(Integer) || left.is_a?(Float)) &&
                         (right.is_a?(Integer) || right.is_a?(Float))

  false
end

.map_iregexp(pattern) ⇒ String

Map I-Regexp pattern to Ruby regex pattern.

Parameters:

  • pattern (String)

Returns:

  • (String) 


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
 
# File 'lib/json_p3/path/function_extensions/pattern.rb', line 9

def self.map_iregexp(pattern)
  escaped = false
  char_class = false
  mapped = String.new(encoding: "UTF-8")

  pattern.each_char do |c|
    if escaped
      mapped << c
      escaped = false
      next
    end

    case c
    when "."
      # mapped << (char_class ? c : "(?:(?![\\r\\n])\\P{Cs}|\\p{Cs}\\p{Cs})")
      mapped << (char_class ? c : "[^\\n\\r]")
    when "\\"
      escaped = true
      mapped << "\\"
    when "["
      char_class = true
      mapped << "["
    when "]"
      char_class = false
      mapped << "]"
    else
      mapped << c
    end
  end

  mapped
end

.match(path, data) ⇒ Object 



36
37
38
 
# File 'lib/json_p3.rb', line 36

def self.match(path, data)
  DefaultEnvironment.match(path, data)
end

.match?(path, data) ⇒ Boolean

Returns:

  • (Boolean) 


40
41
42
 
# File 'lib/json_p3.rb', line 40

def self.match?(path, data)
  DefaultEnvironment.match?(path, data)
end

.name_ch?(ch) ⇒ Boolean

Returns:

  • (Boolean) 


200
201
202
203
204
205
206
 
# File 'lib/json_p3/path/lexer.rb', line 200

def self.name_ch?(ch)
  (ch >= 48 && ch <= 57) ||
    (ch >= 65 && ch <= 90) ||
    (ch >= 97 && ch <= 122) ||
    ch == 95 ||
    (ch >= 0x80 && ch <= 0xffff)
end

.name_first?(ch) ⇒ Boolean

Returns:

  • (Boolean) 


196
197
198
 
# File 'lib/json_p3/path/lexer.rb', line 196

def self.name_first?(ch)
  (ch >= 65 && ch <= 90) || (ch >= 97 && ch <= 122) || ch == 95 || (ch >= 0x80 && ch <= 0xffff)
end

.number_ch?(ch) ⇒ Boolean

Returns:

  • (Boolean) 


208
209
210
 
# File 'lib/json_p3/path/lexer.rb', line 208

def self.number_ch?(ch)
  ch == 45 || (ch >= 48 && ch <= 57)
end

.scan_string_literal(query, byte, pos) ⇒ Object

Raises:



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
 
# File 'lib/json_p3/path/lexer.rb', line 145

def self.scan_string_literal(query, byte, pos)
  start = pos
  length = query.bytesize

  # @type var token: t_token
  # @type var kind: t_token_kind
  # @type var esc_kind: t_token_kind

  kind = byte == 39 ? :token_single_quoted_string : :token_double_quoted_string
  esc_kind = byte == 39 ? :token_single_quoted_esc_string : :token_double_quoted_esc_string

  while pos < length
    ch = query.getbyte(pos)
    case ch
    when 92 # \
      kind = esc_kind
      pos += 2
    when nil
      break
    when 39 # '
      token = [kind, start, pos]
      return [token, pos + 1] if esc_kind == :token_single_quoted_esc_string

      pos += 1
    when 34 # "
      token = [kind, start, pos]
      return [token, pos + 1] if esc_kind == :token_double_quoted_esc_string

      pos += 1
    else
      # Escaped strings get scanned by the parser, where invalid characters will be caught.
      if ch <= 0x1f
        token = [:token_error, start, pos]
        raise JSONP3::Path::SyntaxError.new(
          "invalid character",
          token,
          query
        )
      end
      pos += 1
    end
  end

  token = [:token_error, start, pos]
  raise JSONP3::Path::SyntaxError.new(
    "unclosed string literal",
    token,
    query
  )
end

.tokenize(query) ⇒ Object 



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
 
# File 'lib/json_p3/path/lexer.rb', line 10

def self.tokenize(query)
  tokens = [] #: Array[t_token]
  length = query.bytesize
  start = 0
  pos = 0

  while pos < length
    byte = query.getbyte(pos)

    case byte
    when nil
      break
    when 42 # *
      pos += 1
      tokens << [:token_asterisk, start, pos]
    when 64 # @
      pos += 1
      tokens << [:token_at, start, pos]
    when 58 # :
      pos += 1
      tokens << [:token_colon, start, pos]
    when 44 # ,
      pos += 1
      tokens << [:token_comma, start, pos]
    when 36 # $
      pos += 1
      tokens << [:token_dollar, start, pos]
    when 40 # (
      pos += 1
      tokens << [:token_lparen, start, pos]
    when 91 # [
      pos += 1
      tokens << [:token_lbracket, start, pos]
    when 41 # )
      pos += 1
      tokens << [:token_rparen, start, pos]
    when 93 # ]
      pos += 1
      tokens << [:token_rbracket, start, pos]
    when 63 # ?
      pos += 1
      tokens << [:token_question, start, pos]
    when 38 # &
      tokens << if query.getbyte(pos + 1) == 38
                  pos += 2
                  [:token_and, start, pos]
                else
                  pos += 1
                  [:token_error, start, pos]
                end
    when 124 # |
      tokens << if query.getbyte(pos + 1) == 124
                  pos += 2
                  [:token_or, start, pos]
                else
                  pos += 1
                  [:token_error, start, pos]
                end
    when 46 # .
      tokens << if query.getbyte(pos + 1) == 46
                  pos += 2
                  [:token_double_dot, start, pos]
                else
                  pos += 1
                  [:token_dot, start, pos]
                end
    when 61 # =
      tokens << if query.getbyte(pos + 1) == 61
                  pos += 2
                  [:token_eq, start, pos]
                else
                  pos += 1
                  [:token_error, start, pos]
                end
    when 33 # !
      tokens << if query.getbyte(pos + 1) == 61
                  pos += 2
                  [:token_ne, start, pos]
                else
                  pos += 1
                  [:token_not, start, pos]
                end
    when 62 # >
      tokens << if query.getbyte(pos + 1) == 61
                  pos += 2
                  [:token_ge, start, pos]
                else
                  pos += 1
                  [:token_gt, start, pos]
                end
    when 60 # <
      tokens << if query.getbyte(pos + 1) == 61
                  pos += 2
                  [:token_le, start, pos]
                else
                  pos += 1
                  [:token_lt, start, pos]
                end
    when 39, 34 # ' or "
      pos += 1
      token, pos = scan_string_literal(query, byte, pos)
      tokens << token
    else
      if name_first?(byte)
        pos += 1 while name_ch?(query.getbyte(pos) || 0)
        tokens << [:token_name, start, pos]
      elsif trivia?(byte)
        pos += 1 while trivia?(query.getbyte(pos) || 0)
        tokens << [:token_trivia, start, pos]
      elsif number_ch?(byte)
        if (match = query.match(RE_FLOAT, pos))
          pos = match.end(0) || raise
          tokens << [:token_float, start, pos]
        elsif (match = query.match(RE_INT, pos))
          pos = match.end(0) || raise
          tokens << [:token_int, start, pos]
        elsif (match = query.match(RE_INDEX, pos))
          pos = match.end(0) || raise
          tokens << [:token_index, start, pos]
        else
          pos += 1
          tokens << [:token_error, start, pos]
        end
      else
        pos += 1
        tokens << [:token_error, start, pos]
      end
    end

    start = pos
  end

  tokens
end

.trivia?(ch) ⇒ Boolean

Returns:

  • (Boolean) 


212
213
214
 
# File 'lib/json_p3/path/lexer.rb', line 212

def self.trivia?(ch)
  ch == 32 || ch == 9 || ch == 10 || ch == 13 # rubocop: disable Style/MultipleComparison
end

.truthy?(obj) ⇒ Boolean

Returns:

  • (Boolean) 


416
417
418
419
420
421
 
# File 'lib/json_p3/path/filter.rb', line 416

def self.truthy?(obj)
  return !obj.empty? if obj.is_a?(NodeList)
  return false if obj == :nothing

  obj != false
end

.unescape(value, token, query) ⇒ Object

Replace escape sequences with their equivalent Unicode code point.



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
 
# File 'lib/json_p3/path/unescape.rb', line 11

def self.unescape(value, token, query)
  unescaped = String.new(encoding: "UTF-8")
  scanner = StringScanner.new(value)

  until scanner.eos?
    if scanner.scan(RE_SLASH_U)
      code_point = (scanner.captures&.first || raise).to_i(16)

      if low_surrogate?(code_point)
        raise JSONP3::Path::SyntaxError.new(
          "unexpected low surrogate",
          token,
          query
        )
      end

      if high_surrogate?(code_point)
        unless scanner.scan(RE_SLASH_U)
          raise JSONP3::Path::SyntaxError.new(
            "expected a low surrogate",
            token,
            query
          )
        end

        low_surrogate = (scanner.captures&.first || raise).to_i(16)

        unless low_surrogate?(low_surrogate)
          raise JSONP3::Path::SyntaxError.new(
            "expected a low surrogate",
            token,
            query
          )
        end

        code_point = 0x10000 + (
          ((code_point & 0x03FF) << 10) | (low_surrogate & 0x03FF)
        )
      end

      if code_point <= 0x1f
        raise JSONP3::Path::SyntaxError.new(
          "invalid character #{code_point}",
          token,
          query
        )
      end

      unescaped << code_point.chr(Encoding::UTF_8)
      next
    end

    ch = scanner.getch

    break if ch.nil?

    unless ch == "\\"
      if ch.ord <= 0x1f
        raise JSONP3::Path::SyntaxError.new(
          "invalid character #{ch.ord}",
          token,
          query
        )
      end
      unescaped << ch
      next
    end

    ch = scanner.getch

    case ch
    when "\""
      if token.first == :token_single_quoted_esc_string
        raise JSONP3::Path::SyntaxError.new(
          "unexpected \\\" escape in single quoted string",
          token,
          query
        )
      end
      unescaped << "\""
    when "'"
      if token.first == :token_double_quoted_esc_string
        raise JSONP3::Path::SyntaxError.new(
          "unexpected \\' escape in double quoted string",
          token,
          query
        )
      end
      unescaped << "'"
    when "\\"
      unescaped << "\\"
    when "/"
      unescaped << "/"
    when "b"
      unescaped << "\x08"
    when "f"
      unescaped << "\x0c"
    when "n"
      unescaped << "\n"
    when "r"
      unescaped << "\r"
    when "t"
      unescaped << "\t"
    when "u"
      raise JSONP3::Path::SyntaxError.new("unexpected \\u escape sequence", token, query)
    when nil
      raise JSONP3::Path::SyntaxError.new("incomplete escape sequence", token, query)
    else
      raise JSONP3::Path::SyntaxError.new("unknown escape sequence", token, query)
    end
  end

  unescaped
end