lazy-json 1.0.0
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/lib/lazy-json.rb +284 -0
- metadata +72 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 294fda2ead2eb4a2b288c6849dea7fd4f9360521
|
4
|
+
data.tar.gz: a71bab4930cd523c96cef883478df1fd9cef3cd4
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 2b94b57110452a1ff8d05678dbcd9d57423993e93059701d2055784f7d091de0a90a85871b9ff421058d20befe200be91c825e877d67723470d20389981e15bb
|
7
|
+
data.tar.gz: e9d42662275f2c1e9846d146fea127479b975aa714a5182d37ad7ee1cbf4c95ecd9b225cff9a299c07ce58d402a7bba9d94fb7c5f97e246975c295697a8d1b1e
|
data/lib/lazy-json.rb
ADDED
@@ -0,0 +1,284 @@
|
|
1
|
+
require 'oj'
|
2
|
+
|
3
|
+
module LazyJson
|
4
|
+
|
5
|
+
def self.attach(json)
|
6
|
+
LazyValue.new(Sequence.new(json))
|
7
|
+
end
|
8
|
+
|
9
|
+
# A sequence of json JSON characters
|
10
|
+
class Sequence
|
11
|
+
|
12
|
+
# See http://stackoverflow.com/questions/16042274/definition-of-whitespace-in-json
|
13
|
+
WHITESPACE = [
|
14
|
+
0x20, # Space
|
15
|
+
0x09, # Horizontal tab
|
16
|
+
0x0A, # Line feed or New line
|
17
|
+
0x0D # Carriage return
|
18
|
+
]
|
19
|
+
|
20
|
+
# Note positions are in bytes, not characters
|
21
|
+
def initialize(json, start_pos = 0, end_pos = json.bytesize)
|
22
|
+
raise "Sequence end ${ end_pos } is before start #{ start_pos }" if end_pos < start_pos
|
23
|
+
@json = json
|
24
|
+
@start_pos = start_pos
|
25
|
+
@end_pos = end_pos
|
26
|
+
end
|
27
|
+
|
28
|
+
attr_reader :json
|
29
|
+
attr_reader :start_pos
|
30
|
+
attr_reader :end_pos
|
31
|
+
|
32
|
+
def to_s
|
33
|
+
@json.byteslice(@start_pos...@end_pos)
|
34
|
+
end
|
35
|
+
|
36
|
+
def byte_at(i)
|
37
|
+
@json.getbyte(@start_pos + i)
|
38
|
+
end
|
39
|
+
|
40
|
+
def first
|
41
|
+
byte_at(0)
|
42
|
+
end
|
43
|
+
|
44
|
+
def empty?
|
45
|
+
@start_pos == @end_pos
|
46
|
+
end
|
47
|
+
|
48
|
+
def prefix(end_pos)
|
49
|
+
Sequence.new(@json, @start_pos, end_pos)
|
50
|
+
end
|
51
|
+
|
52
|
+
def suffix(start_pos)
|
53
|
+
Sequence.new(@json, start_pos, @end_pos)
|
54
|
+
end
|
55
|
+
|
56
|
+
def remainder(enclosing_seq)
|
57
|
+
Sequence.new(@json, @end_pos, enclosing_seq.end_pos)
|
58
|
+
end
|
59
|
+
|
60
|
+
def read_whitespace
|
61
|
+
prefix(skim_whitespace(@start_pos))
|
62
|
+
end
|
63
|
+
|
64
|
+
def skip_whitespace
|
65
|
+
suffix(skim_whitespace(@start_pos))
|
66
|
+
end
|
67
|
+
|
68
|
+
def read_byte(byte, required = true)
|
69
|
+
prefix(skim_byte(@start_pos, byte, required))
|
70
|
+
end
|
71
|
+
|
72
|
+
def skip_byte(byte, required = true)
|
73
|
+
suffix(skim_byte(@start_pos, byte, required))
|
74
|
+
end
|
75
|
+
|
76
|
+
def read_until(terminator, include_terminator)
|
77
|
+
prefix(skim_until(@start_pos, false, terminator, include_terminator))
|
78
|
+
end
|
79
|
+
|
80
|
+
def skip_until(terminator, include_terminator)
|
81
|
+
suffix(skim_until(@start_pos, false, terminator, include_terminator))
|
82
|
+
end
|
83
|
+
|
84
|
+
private
|
85
|
+
|
86
|
+
def skim_whitespace(start)
|
87
|
+
i = start
|
88
|
+
while i < @end_pos && WHITESPACE.include?(@json.getbyte(i))
|
89
|
+
i += 1
|
90
|
+
end
|
91
|
+
i
|
92
|
+
end
|
93
|
+
|
94
|
+
def skim_byte(at, byte, required)
|
95
|
+
byte = [ byte ] unless byte.is_a?(::Array)
|
96
|
+
if byte.include?(@json.getbyte(at))
|
97
|
+
at + 1
|
98
|
+
elsif required
|
99
|
+
raise "Expected #{ byte } but got '#{ @json.getbyte(at) }'"
|
100
|
+
else
|
101
|
+
at
|
102
|
+
end
|
103
|
+
end
|
104
|
+
|
105
|
+
def skim_until(start, in_string, terminator, include_terminator)
|
106
|
+
terminator = [ terminator ] unless terminator.is_a?(::Array)
|
107
|
+
i = start
|
108
|
+
while i < @end_pos
|
109
|
+
byte = @json.getbyte(i)
|
110
|
+
|
111
|
+
# Skip unicode characters. See table at https://en.wikipedia.org/wiki/UTF-8.
|
112
|
+
if byte & 0b11100000 == 0b11000000
|
113
|
+
i += 2
|
114
|
+
elsif byte & 0b11110000 == 0b11100000
|
115
|
+
i += 3
|
116
|
+
elsif byte & 0b11111000 == 0b11110000
|
117
|
+
i += 4
|
118
|
+
|
119
|
+
elsif in_string && byte == 92 # '\\'.ord
|
120
|
+
i += escape_sequence_length(i) # String escape sequence
|
121
|
+
elsif terminator.include?(byte)
|
122
|
+
i += 1 if include_terminator
|
123
|
+
break
|
124
|
+
else
|
125
|
+
i += 1
|
126
|
+
if ! in_string
|
127
|
+
if byte == 34 # '"'.ord
|
128
|
+
i = skim_until(i, true, 34, true) # '"'.ord - String start
|
129
|
+
elsif byte == 91 # '['.ord
|
130
|
+
i = skim_until(i, false, 93, true) # ']'.ord - Array start
|
131
|
+
elsif byte == 123 # '{'.ord
|
132
|
+
i = skim_until(i, false, 125, true) # '}'.ord - Object start
|
133
|
+
end
|
134
|
+
end
|
135
|
+
end
|
136
|
+
end
|
137
|
+
i
|
138
|
+
end
|
139
|
+
|
140
|
+
def escape_sequence_length(start)
|
141
|
+
raise 'Escape sequence must start with \\' if @json.getbyte(start) != 92 # '\\'.ord
|
142
|
+
byte = @json.getbyte(start + 1)
|
143
|
+
if byte == 120 # 'x'.ord - \x followed by 2 hex digits
|
144
|
+
4
|
145
|
+
elsif byte == 117 # 'u'.ord - \u followed by 4 hex digits
|
146
|
+
6
|
147
|
+
elsif byte >= 48 && byte <= 57 # '0'.ord, '9'.ord - \ followed by 3 octal digits
|
148
|
+
4
|
149
|
+
else # \ followed by single escaped character
|
150
|
+
2
|
151
|
+
end
|
152
|
+
end
|
153
|
+
|
154
|
+
end
|
155
|
+
|
156
|
+
class Value
|
157
|
+
|
158
|
+
def initialize(seq)
|
159
|
+
@seq = seq
|
160
|
+
end
|
161
|
+
|
162
|
+
def parse
|
163
|
+
Oj.load(@seq.to_s) # Note JSON.parse fails on primitives since they're invalid as documents
|
164
|
+
end
|
165
|
+
|
166
|
+
end
|
167
|
+
|
168
|
+
class LazyValue < Value
|
169
|
+
|
170
|
+
def initialize(seq)
|
171
|
+
super(seq)
|
172
|
+
@parsed = false
|
173
|
+
@value = nil
|
174
|
+
end
|
175
|
+
|
176
|
+
def value
|
177
|
+
if ! @parsed
|
178
|
+
byte = @seq.skip_whitespace.first
|
179
|
+
if byte == 123 # '{'.ord
|
180
|
+
@value = Object.new(@seq)
|
181
|
+
elsif byte == 91 # '['.ord
|
182
|
+
@value = Array.new(@seq)
|
183
|
+
else
|
184
|
+
@value = Primitive.new(@seq)
|
185
|
+
end
|
186
|
+
@parsed = true
|
187
|
+
end
|
188
|
+
@value
|
189
|
+
end
|
190
|
+
|
191
|
+
def [](key_or_index)
|
192
|
+
value[key_or_index]
|
193
|
+
end
|
194
|
+
|
195
|
+
end
|
196
|
+
|
197
|
+
class Object < Value
|
198
|
+
|
199
|
+
def initialize(seq)
|
200
|
+
super(seq)
|
201
|
+
@fields = {}
|
202
|
+
@fseq = @seq.skip_whitespace.skip_byte(123) # '{'.ord
|
203
|
+
end
|
204
|
+
|
205
|
+
# Access a field, lazily parsing if not yet parsed
|
206
|
+
def [](key)
|
207
|
+
if ! @fields.has_key?(key) && ! @fseq.empty?
|
208
|
+
while true
|
209
|
+
@fseq = @fseq.skip_whitespace
|
210
|
+
if @fseq.first == 125 # '}'.ord
|
211
|
+
@fseq = @fseq.skip_byte(125).skip_whitespace # '}'.ord
|
212
|
+
break
|
213
|
+
end
|
214
|
+
new_key, new_value = read_field_and_consume
|
215
|
+
@fields[new_key] = new_value
|
216
|
+
break if new_key == key
|
217
|
+
end
|
218
|
+
end
|
219
|
+
@fields[key]
|
220
|
+
end
|
221
|
+
|
222
|
+
private
|
223
|
+
|
224
|
+
def read_field_and_consume
|
225
|
+
key_seq = @fseq.read_until(58, false) # ':'.ord
|
226
|
+
key = Value.new(key_seq).parse
|
227
|
+
raise "Non-string object key #{ key }" unless key.is_a?(String)
|
228
|
+
@fseq = key_seq.remainder(@fseq)
|
229
|
+
@fseq = @fseq.skip_byte(58) # ':'.ord
|
230
|
+
value_seq = @fseq.read_until([ 44, 125 ], false) # ','.ord, '}'.ord
|
231
|
+
@fseq = value_seq.remainder(@fseq)
|
232
|
+
sep_seq = @fseq.read_byte([ 44, 125 ]) # ','.ord, '}'.ord
|
233
|
+
@fseq = sep_seq.remainder(@fseq) if sep_seq.first == 44 # ','.ord - Consume , but not }
|
234
|
+
[ key, LazyValue.new(value_seq) ]
|
235
|
+
end
|
236
|
+
|
237
|
+
end
|
238
|
+
|
239
|
+
class Array < Value
|
240
|
+
|
241
|
+
def initialize(seq)
|
242
|
+
super(seq)
|
243
|
+
@elements = []
|
244
|
+
@eseq = @seq.skip_whitespace.skip_byte(91) # '['.ord
|
245
|
+
end
|
246
|
+
|
247
|
+
# Access an element, lazily parsing if not yet parsed
|
248
|
+
def [](i)
|
249
|
+
if @elements.size <= i && ! @eseq.empty?
|
250
|
+
while true
|
251
|
+
@eseq = @eseq.skip_whitespace
|
252
|
+
if @eseq.first == 93 # ']'.ord
|
253
|
+
@eseq = @eseq.skip_byte(93).skip_whitespace # ']'.ord
|
254
|
+
break
|
255
|
+
end
|
256
|
+
new_value = read_value_and_consume
|
257
|
+
@elements << new_value
|
258
|
+
break if @elements.size > i
|
259
|
+
end
|
260
|
+
end
|
261
|
+
@elements[i]
|
262
|
+
end
|
263
|
+
|
264
|
+
private
|
265
|
+
|
266
|
+
def read_value_and_consume
|
267
|
+
value_seq = @eseq.read_until([ 44, 93 ], false) # ','.ord, ']'.ord
|
268
|
+
@eseq = value_seq.remainder(@eseq)
|
269
|
+
sep_seq = @eseq.read_byte([ 44, 93 ]) # ','.ord, ']'.ord
|
270
|
+
@eseq = sep_seq.remainder(@eseq) if sep_seq.first == 44 # ','.ord - Consume , but not ]
|
271
|
+
LazyValue.new(value_seq)
|
272
|
+
end
|
273
|
+
|
274
|
+
end
|
275
|
+
|
276
|
+
class Primitive < Value
|
277
|
+
|
278
|
+
def initialize(seq)
|
279
|
+
super(seq)
|
280
|
+
end
|
281
|
+
|
282
|
+
end
|
283
|
+
|
284
|
+
end
|
metadata
ADDED
@@ -0,0 +1,72 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: lazy-json
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 1.0.0
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Paul McReynolds
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2015-09-03 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: oj
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - "~>"
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '2'
|
20
|
+
type: :runtime
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '2'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: rspec
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: '3'
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: '3'
|
41
|
+
description: Lazy JSON skimmer-parser
|
42
|
+
email: paul@conspire.com
|
43
|
+
executables: []
|
44
|
+
extensions: []
|
45
|
+
extra_rdoc_files: []
|
46
|
+
files:
|
47
|
+
- lib/lazy-json.rb
|
48
|
+
homepage: https://github.com/conspire-org/lazy-json
|
49
|
+
licenses:
|
50
|
+
- MIT
|
51
|
+
metadata: {}
|
52
|
+
post_install_message:
|
53
|
+
rdoc_options: []
|
54
|
+
require_paths:
|
55
|
+
- lib
|
56
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
57
|
+
requirements:
|
58
|
+
- - ">="
|
59
|
+
- !ruby/object:Gem::Version
|
60
|
+
version: '0'
|
61
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
62
|
+
requirements:
|
63
|
+
- - ">="
|
64
|
+
- !ruby/object:Gem::Version
|
65
|
+
version: '0'
|
66
|
+
requirements: []
|
67
|
+
rubyforge_project:
|
68
|
+
rubygems_version: 2.4.5.1
|
69
|
+
signing_key:
|
70
|
+
specification_version: 4
|
71
|
+
summary: Lazy JSON
|
72
|
+
test_files: []
|