json_stream_trigger 0.1
Sign up to get free protection for your applications and to get access to all the features.
- checksums.yaml +7 -0
- data/README.md +76 -0
- data/lib/json_path.rb +22 -0
- data/lib/json_stream_trigger.rb +188 -0
- metadata +47 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA1:
|
3
|
+
metadata.gz: 11dbed3cd388bb5830c6956a4792bafb172d249a
|
4
|
+
data.tar.gz: 73ff8ca8f6301a079c9ebdddb531b292e683e641
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 78a5fcc5b1a352fe79cbfbc0218d5d5f41ae3d34a83a69eb4c0cc85c94fb6004e440d0705c90de0822591b1fc6eca00ee12cf59a082386f4517a99882cc9e23b
|
7
|
+
data.tar.gz: ffbcc34b8cdb0564293d548368f13b233804819d691746ca7266b3c84759cae38b1f75e17572f5e9d6e028bd219553ef4744c2e9ec85bd6f1e2b406f9cfe3d39
|
data/README.md
ADDED
@@ -0,0 +1,76 @@
|
|
1
|
+
# Ruby `json_stream_trigger` Gem
|
2
|
+
|
3
|
+
Instead of parsing a huge JSON files and loading it into memory,
|
4
|
+
this library will stream the bytes through
|
5
|
+
[json-stream](https://github.com/dgraham/json-stream) and only
|
6
|
+
creates a small buffer for objects whose JSONPath matches a pattern you specify.
|
7
|
+
When the object is completed, the specified block will be called.
|
8
|
+
|
9
|
+
Install with `gem "json_stream_trigger"` in your Gemfile.
|
10
|
+
|
11
|
+
## Example:
|
12
|
+
|
13
|
+
```ruby
|
14
|
+
f = File.open('really_big_file.json')
|
15
|
+
stream = JsonStreamTrigger.new()
|
16
|
+
|
17
|
+
# Match each array item. If you wanted to whole array use $.data
|
18
|
+
stream.on('$.data[*]') do |json_string|
|
19
|
+
import JSON.parse(json_string, :quirks_mode => true)
|
20
|
+
end
|
21
|
+
|
22
|
+
# Will match for $.any.sub[*].item.meta
|
23
|
+
stream.on('$..meta') do |json_string|
|
24
|
+
save_meta JSON.parse(json_string, :quirks_mode => true)
|
25
|
+
end
|
26
|
+
|
27
|
+
# read in 1MB chunks
|
28
|
+
while chunk = f.read(1024)
|
29
|
+
stream << chunk
|
30
|
+
end
|
31
|
+
|
32
|
+
```
|
33
|
+
|
34
|
+
The captured JSON strinb buffer will be passed to the block. Note, Ruby's JSON library expects
|
35
|
+
JSON documents to be passed to it - not primatives - this is why `:quirks_mode => true` has been added.
|
36
|
+
|
37
|
+
## Path Details
|
38
|
+
The JSONPaths are similar to XPath notation. `$` is the root,
|
39
|
+
single wild card keys can be done with `$.*.version`,
|
40
|
+
or you can do muli-level wildcard with `$.docs..name`.
|
41
|
+
[More info on JSONPath](http://goessner.net/articles/JsonPath/)
|
42
|
+
|
43
|
+
A few more examples:
|
44
|
+
|
45
|
+
```javascript
|
46
|
+
{
|
47
|
+
meta: {version: 0.1},
|
48
|
+
docs: [
|
49
|
+
{id: 1},
|
50
|
+
{id: 2},
|
51
|
+
{id: 3},
|
52
|
+
{id: 4},
|
53
|
+
{
|
54
|
+
id: 5,
|
55
|
+
user: {
|
56
|
+
name: "Tyler"
|
57
|
+
}
|
58
|
+
}
|
59
|
+
]
|
60
|
+
}
|
61
|
+
```
|
62
|
+
|
63
|
+
```ruby
|
64
|
+
on('$.docs[*].id') # triggers for id property of every item in docs array
|
65
|
+
on('$.docs') # returns full array of items
|
66
|
+
on('$.docs[*]') # triggers for each item in the array
|
67
|
+
on('$.docs[1].id') # returns value of ID 1
|
68
|
+
on('$.docs[*].*.name') # returns 'Tyler'
|
69
|
+
on('$..name') # matches any value who's key is 'name'
|
70
|
+
```
|
71
|
+
|
72
|
+
|
73
|
+
|
74
|
+
## Tests
|
75
|
+
`rake test`
|
76
|
+
|
data/lib/json_path.rb
ADDED
@@ -0,0 +1,22 @@
|
|
1
|
+
|
2
|
+
class JsonPath
|
3
|
+
|
4
|
+
def self.matches?(path, pattern)
|
5
|
+
re = self.convert_to_re(pattern)
|
6
|
+
!path.match(re).nil?
|
7
|
+
end
|
8
|
+
|
9
|
+
def self.convert_to_re(pattern)
|
10
|
+
re = pattern.dup
|
11
|
+
re = re.gsub('[', '\[').gsub(']', '\]')# escape brackets
|
12
|
+
re.gsub!(/^\$/, '^\$') # escape $ and fix it to root
|
13
|
+
re.gsub!(/ \\\[ \* \\\] /x, '\[\d+\]') # change [*] to [\d+]
|
14
|
+
re.gsub!('..', '(?<=[\.\$\]]).*[\.\]]') # change .. to match a dot, $, or ] followed by anything, and ending in a . or ]
|
15
|
+
re.gsub!('.*.', '\.[^\.\[\]]+\.') # wild card will match any key
|
16
|
+
|
17
|
+
re += '(?=$)' #'(?=$|\.)
|
18
|
+
return Regexp.new(re)
|
19
|
+
end
|
20
|
+
|
21
|
+
end
|
22
|
+
|
@@ -0,0 +1,188 @@
|
|
1
|
+
require 'json'
|
2
|
+
require 'json/stream'
|
3
|
+
require_relative 'json_path'
|
4
|
+
|
5
|
+
class JsonStreamTrigger
|
6
|
+
attr_reader :key_path, :triggers, :full_buffer
|
7
|
+
DEBUG=false
|
8
|
+
|
9
|
+
def initialize()
|
10
|
+
@parser = JSON::Stream::Parser.new
|
11
|
+
|
12
|
+
@parser.start_object &method(:start_object)
|
13
|
+
@parser.end_object &method(:end_object)
|
14
|
+
@parser.start_array &method(:start_array)
|
15
|
+
@parser.end_array &method(:end_array)
|
16
|
+
@parser.key &method(:key)
|
17
|
+
@parser.value &method(:value)
|
18
|
+
|
19
|
+
@last_call = nil
|
20
|
+
@key_path = ''
|
21
|
+
@triggers = {}
|
22
|
+
@active_buffers = {}
|
23
|
+
@full_buffer = ''
|
24
|
+
end
|
25
|
+
|
26
|
+
def on(pattern, &block)
|
27
|
+
@triggers[pattern] = block
|
28
|
+
end
|
29
|
+
|
30
|
+
def <<(bytes)
|
31
|
+
debug "bytes: #{bytes.inspect}"
|
32
|
+
@parser << bytes
|
33
|
+
@full_buffer << bytes if DEBUG
|
34
|
+
end
|
35
|
+
|
36
|
+
def path_matches?(pattern)
|
37
|
+
JsonPath.matches?(@key_path, pattern)
|
38
|
+
end
|
39
|
+
|
40
|
+
protected
|
41
|
+
|
42
|
+
|
43
|
+
################################ PARSING TRIGGERS ###########################
|
44
|
+
#def start_document
|
45
|
+
#end
|
46
|
+
#def end_document
|
47
|
+
#end
|
48
|
+
|
49
|
+
def start_object
|
50
|
+
debug "start object"
|
51
|
+
@key_path << (@key_path.empty? ? "$" : '')
|
52
|
+
increment_path_array() do
|
53
|
+
activate_buffers_for_matching()
|
54
|
+
end
|
55
|
+
append_buffers ',{'
|
56
|
+
@last_call = :start_object
|
57
|
+
end
|
58
|
+
|
59
|
+
def end_object
|
60
|
+
debug "end object"
|
61
|
+
append_buffers '}' if @last_call == :start_object
|
62
|
+
trigger_block_for_matching()
|
63
|
+
append_buffers '}' if @last_call != :start_object
|
64
|
+
trim_segment(/[\.\$][^\.\[\[]+$/) # remove last .key
|
65
|
+
@last_call = :end_object
|
66
|
+
end
|
67
|
+
|
68
|
+
def start_array
|
69
|
+
debug "start array"
|
70
|
+
increment_path_array() do
|
71
|
+
activate_buffers_for_matching()
|
72
|
+
end
|
73
|
+
append_buffers ',['
|
74
|
+
@key_path << (@key_path.empty? ? "$[]" : "[]")
|
75
|
+
activate_buffers_for_matching()
|
76
|
+
@last_call = :start_array
|
77
|
+
end
|
78
|
+
|
79
|
+
def end_array
|
80
|
+
debug "end array"
|
81
|
+
append_buffers ']'
|
82
|
+
|
83
|
+
trim_segment(/\[\d*\]+$/) # remove last [\d] and check triggers to match .my-array
|
84
|
+
trigger_block_for_matching()
|
85
|
+
trim_segment(/[\.$][^\.\[\]]+$/) # remove last .my-array
|
86
|
+
@last_call = :end_array
|
87
|
+
end
|
88
|
+
|
89
|
+
def key(k)
|
90
|
+
debug "new key '#{k}'"
|
91
|
+
trim_segment(/\.[^\.\[\]]+$/) unless @last_call == :start_object# remove last .key[\d]
|
92
|
+
@key_path << ".#{k}"
|
93
|
+
append_buffers ",\"#{k}\":"
|
94
|
+
activate_buffers_for_matching()
|
95
|
+
@last_call = :key
|
96
|
+
end
|
97
|
+
|
98
|
+
def value(v)
|
99
|
+
debug "value '#{v}'"
|
100
|
+
increment_path_array() do
|
101
|
+
activate_buffers_for_matching()
|
102
|
+
end
|
103
|
+
|
104
|
+
append_buffers ','
|
105
|
+
append_buffers JSON.dump(v)
|
106
|
+
|
107
|
+
trigger_block_for_matching()
|
108
|
+
@last_call = :value
|
109
|
+
end
|
110
|
+
|
111
|
+
################################ BUFFER STUFF ###########################
|
112
|
+
|
113
|
+
# Called when we know the name of the object/array we are workign with
|
114
|
+
def activate_buffers_for_matching
|
115
|
+
@triggers.keys.each do |pattern|
|
116
|
+
debug "checking #{@key_path} matches #{pattern}"
|
117
|
+
if JsonPath.matches?(@key_path, pattern) && !@active_buffers.keys.include?(pattern)
|
118
|
+
debug ">> Activating buffer for #{pattern.inspect}"
|
119
|
+
@active_buffers[pattern] = ''
|
120
|
+
end
|
121
|
+
end
|
122
|
+
end
|
123
|
+
|
124
|
+
# To be called when exiting an object or array so the buffer is completed
|
125
|
+
def trigger_block_for_matching
|
126
|
+
active_patterns = @active_buffers.keys
|
127
|
+
active_patterns.each do |pattern|
|
128
|
+
if JsonPath.matches?(@key_path, pattern)
|
129
|
+
debug "<< Calling trigger for '#{pattern}'"
|
130
|
+
@triggers[pattern].call @active_buffers[pattern]
|
131
|
+
if pattern[-3..3] == '[*]'
|
132
|
+
@active_buffers[pattern] = ''
|
133
|
+
debug "Clearing buffer for '#{pattern}'"
|
134
|
+
else
|
135
|
+
@active_buffers.delete(pattern)
|
136
|
+
debug "Stopping buffer for '#{pattern}'"
|
137
|
+
end
|
138
|
+
end
|
139
|
+
end
|
140
|
+
end
|
141
|
+
|
142
|
+
def append_buffers(bytes)
|
143
|
+
@active_buffers.keys.each do |k|
|
144
|
+
# remove comma if it's not needed
|
145
|
+
if bytes[0] == ',' && [nil, '[', '{', ':'].include?(@active_buffers[k][-1, 1])
|
146
|
+
bytes = bytes[1..-1]
|
147
|
+
end
|
148
|
+
@active_buffers[k] += "#{bytes}"
|
149
|
+
debug "Appended to #{k} => '#{bytes}' to buffer '#{@active_buffers[k]}'"
|
150
|
+
end
|
151
|
+
end
|
152
|
+
|
153
|
+
############################### PATH STUFF #######################
|
154
|
+
|
155
|
+
# trim off the last segment of the key_path
|
156
|
+
def trim_segment(re)
|
157
|
+
@key_path.sub!(re, '')
|
158
|
+
@key_path << '$' if @key_path == '' # add back the $ if we trimmed it off
|
159
|
+
debug " trimmed off #{re}"
|
160
|
+
end
|
161
|
+
|
162
|
+
def increment_path_array(&block)
|
163
|
+
# Increment the array index if we are in an array
|
164
|
+
# Note: Xpath indexes start at 1
|
165
|
+
# pull out the [\d] from the last array index
|
166
|
+
did_update = false
|
167
|
+
@key_path.sub!(/\[(\d*)\]$/) do |m, x|
|
168
|
+
debug "incrementing path array: #{@key_path}"
|
169
|
+
new_i = m.match(/\[(\d*)\]/)[1].to_i + 1
|
170
|
+
did_update = true
|
171
|
+
debug " new array i = #{new_i}"
|
172
|
+
'[' + new_i.to_s + ']'
|
173
|
+
end
|
174
|
+
|
175
|
+
if did_update
|
176
|
+
block.call()
|
177
|
+
end
|
178
|
+
|
179
|
+
end
|
180
|
+
|
181
|
+
|
182
|
+
|
183
|
+
def debug(msg)
|
184
|
+
indent = 60
|
185
|
+
puts msg + (" " * [0, (indent - msg.length)].max ) + "PATH: #{@key_path}" if DEBUG
|
186
|
+
end
|
187
|
+
|
188
|
+
end
|
metadata
ADDED
@@ -0,0 +1,47 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: json_stream_trigger
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: '0.1'
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Tyler Roberts
|
8
|
+
autorequire:
|
9
|
+
bindir: bin
|
10
|
+
cert_chain: []
|
11
|
+
date: 2016-01-12 00:00:00.000000000 Z
|
12
|
+
dependencies: []
|
13
|
+
description: ''
|
14
|
+
email: code@polar-concepts.com
|
15
|
+
executables: []
|
16
|
+
extensions: []
|
17
|
+
extra_rdoc_files: []
|
18
|
+
files:
|
19
|
+
- README.md
|
20
|
+
- lib/json_path.rb
|
21
|
+
- lib/json_stream_trigger.rb
|
22
|
+
homepage: http://github.com/bdevel/json_stream_trigger
|
23
|
+
licenses:
|
24
|
+
- MIT
|
25
|
+
metadata: {}
|
26
|
+
post_install_message:
|
27
|
+
rdoc_options: []
|
28
|
+
require_paths:
|
29
|
+
- lib
|
30
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
31
|
+
requirements:
|
32
|
+
- - '>='
|
33
|
+
- !ruby/object:Gem::Version
|
34
|
+
version: '0'
|
35
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
36
|
+
requirements:
|
37
|
+
- - '>='
|
38
|
+
- !ruby/object:Gem::Version
|
39
|
+
version: '0'
|
40
|
+
requirements: []
|
41
|
+
rubyforge_project:
|
42
|
+
rubygems_version: 2.2.2
|
43
|
+
signing_key:
|
44
|
+
specification_version: 4
|
45
|
+
summary: Parse large JSON files as a stream and trigger events upon key matching.
|
46
|
+
test_files: []
|
47
|
+
has_rdoc:
|