amazon-textract-parser-ruby 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +7 -0
- data/.DS_Store +0 -0
- data/.gitignore +8 -0
- data/.travis.yml +7 -0
- data/Gemfile +6 -0
- data/Gemfile.lock +45 -0
- data/LICENSE.txt +21 -0
- data/README.md +46 -0
- data/Rakefile +10 -0
- data/amazon-textract-parser-ruby.gemspec +31 -0
- data/bin/console +14 -0
- data/bin/setup +8 -0
- data/lib/.DS_Store +0 -0
- data/lib/amazon-textract-parser-ruby.rb +590 -0
- data/lib/amazon-textract-parser-ruby/.DS_Store +0 -0
- data/lib/amazon-textract-parser-ruby/version.rb +3 -0
- metadata +129 -0
checksums.yaml
ADDED
@@ -0,0 +1,7 @@
|
|
1
|
+
---
|
2
|
+
SHA256:
|
3
|
+
metadata.gz: 415a439783ef9a00d5d1caf1ddbc111e5311a5bdce51a064daf5b831a6615c34
|
4
|
+
data.tar.gz: 358f5d7b9f74dad6a5f7110188769be65eda89ea9d346072ae047d03c083373f
|
5
|
+
SHA512:
|
6
|
+
metadata.gz: 70cd4616be9e8047fac583c6e4db430654ae414934a19c4df9c6e832792a01cdda55d300d4f602714b2bb87f52ebc1c0f1311717d3446a72486bfe9cf521a12f
|
7
|
+
data.tar.gz: 91bf70e8869832452cb896243691090fcb7042e4fb77a3ee121bf64265921ac320ba5ec399322e17fc32b166bd2656b6108cda75aad1417d1dbd6005b2de2205
|
data/.DS_Store
ADDED
Binary file
|
data/.gitignore
ADDED
data/.travis.yml
ADDED
data/Gemfile
ADDED
data/Gemfile.lock
ADDED
@@ -0,0 +1,45 @@
|
|
1
|
+
PATH
|
2
|
+
remote: .
|
3
|
+
specs:
|
4
|
+
amazon-textract-parser-ruby (0.1.0)
|
5
|
+
|
6
|
+
GEM
|
7
|
+
remote: https://rubygems.org/
|
8
|
+
specs:
|
9
|
+
activesupport (6.0.3.2)
|
10
|
+
concurrent-ruby (~> 1.0, >= 1.0.2)
|
11
|
+
i18n (>= 0.7, < 2)
|
12
|
+
minitest (~> 5.1)
|
13
|
+
tzinfo (~> 1.1)
|
14
|
+
zeitwerk (~> 2.2, >= 2.2.2)
|
15
|
+
ansi (1.5.0)
|
16
|
+
builder (3.2.4)
|
17
|
+
concurrent-ruby (1.1.6)
|
18
|
+
i18n (1.8.3)
|
19
|
+
concurrent-ruby (~> 1.0)
|
20
|
+
minitest (5.14.1)
|
21
|
+
minitest-reporters (1.4.2)
|
22
|
+
ansi
|
23
|
+
builder
|
24
|
+
minitest (>= 5.0)
|
25
|
+
ruby-progressbar
|
26
|
+
rake (12.3.3)
|
27
|
+
ruby-progressbar (1.10.1)
|
28
|
+
thread_safe (0.3.6)
|
29
|
+
tzinfo (1.2.7)
|
30
|
+
thread_safe (~> 0.1)
|
31
|
+
zeitwerk (2.4.0)
|
32
|
+
|
33
|
+
PLATFORMS
|
34
|
+
ruby
|
35
|
+
|
36
|
+
DEPENDENCIES
|
37
|
+
activesupport (~> 6.0.3.2)
|
38
|
+
amazon-textract-parser-ruby!
|
39
|
+
bundler (~> 1.17)
|
40
|
+
minitest (~> 5.0)
|
41
|
+
minitest-reporters
|
42
|
+
rake (~> 12.3.3)
|
43
|
+
|
44
|
+
BUNDLED WITH
|
45
|
+
1.17.2
|
data/LICENSE.txt
ADDED
@@ -0,0 +1,21 @@
|
|
1
|
+
The MIT License (MIT)
|
2
|
+
|
3
|
+
Copyright (c) 2020 Niels Vanspauwen
|
4
|
+
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
7
|
+
in the Software without restriction, including without limitation the rights
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
10
|
+
furnished to do so, subject to the following conditions:
|
11
|
+
|
12
|
+
The above copyright notice and this permission notice shall be included in
|
13
|
+
all copies or substantial portions of the Software.
|
14
|
+
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
21
|
+
THE SOFTWARE.
|
data/README.md
ADDED
@@ -0,0 +1,46 @@
|
|
1
|
+
# Amazon Textract Results Parser
|
2
|
+
|
3
|
+
This is a quick Ruby port of [https://github.com/mludvig/amazon-textract-parser](https://github.com/mludvig/amazon-textract-parser)
|
4
|
+
|
5
|
+
It's useful for interpreting the result of Amazon Textract info.
|
6
|
+
|
7
|
+
## Installation
|
8
|
+
|
9
|
+
Add this line to your application's Gemfile:
|
10
|
+
|
11
|
+
```ruby
|
12
|
+
gem 'amazon-textract-parser-ruby'
|
13
|
+
```
|
14
|
+
|
15
|
+
And then execute:
|
16
|
+
|
17
|
+
$ bundle
|
18
|
+
|
19
|
+
Or install it yourself as:
|
20
|
+
|
21
|
+
$ gem install amazon-textract-parser-ruby
|
22
|
+
|
23
|
+
## Usage
|
24
|
+
|
25
|
+
```ruby
|
26
|
+
textract = Aws::Textract::Client.new
|
27
|
+
textract.start_document_analysis({...})
|
28
|
+
response = textract.get_document_analysis({...})
|
29
|
+
doc = AmazonTRP::Document.new(response.to_h)
|
30
|
+
```
|
31
|
+
|
32
|
+
## Development
|
33
|
+
|
34
|
+
After checking out the repo, run `bin/setup` to install dependencies. Then, run `rake test` to run the tests. You can also run `bin/console` for an interactive prompt that will allow you to experiment.
|
35
|
+
|
36
|
+
To install this gem onto your local machine, run `bundle exec rake install`. To release a new version, update the version number in `version.rb`, and then run `bundle exec rake release`, which will create a git tag for the version, push git commits and tags, and push the `.gem` file to [rubygems.org](https://rubygems.org).
|
37
|
+
|
38
|
+
For more info on creating and maintaining gems, check https://bundler.io/v2.0/guides/creating_gem.html
|
39
|
+
|
40
|
+
## Contributing
|
41
|
+
|
42
|
+
Bug reports and pull requests are welcome on GitHub at https://github.com/nielsvanspauwen/amazon-textract-parser-ruby.
|
43
|
+
|
44
|
+
## License
|
45
|
+
|
46
|
+
The gem is available as open source under the terms of the [MIT License](https://opensource.org/licenses/MIT).
|
data/Rakefile
ADDED
@@ -0,0 +1,31 @@
|
|
1
|
+
|
2
|
+
lib = File.expand_path("../lib", __FILE__)
|
3
|
+
$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
|
4
|
+
require "amazon-textract-parser-ruby/version"
|
5
|
+
|
6
|
+
Gem::Specification.new do |spec|
|
7
|
+
spec.name = "amazon-textract-parser-ruby"
|
8
|
+
spec.version = AmazonTRP::VERSION
|
9
|
+
spec.authors = ["Niels Vanspauwen"]
|
10
|
+
spec.email = ["niels.vanspauwen@gmail.com"]
|
11
|
+
|
12
|
+
spec.summary = %q{Amazon Textract Results Parser}
|
13
|
+
spec.description = %q{This is a quick Ruby port of https://github.com/mludvig/amazon-textract-parser\nIt's useful for interpreting the result of Amazon Textract info.}
|
14
|
+
spec.homepage = "https://github.com/nielsvanspauwen/amazon-textract-parser-ruby"
|
15
|
+
spec.license = "MIT"
|
16
|
+
|
17
|
+
# Specify which files should be added to the gem when it is released.
|
18
|
+
# The `git ls-files -z` loads the files in the RubyGem that have been added into git.
|
19
|
+
spec.files = Dir.chdir(File.expand_path('..', __FILE__)) do
|
20
|
+
`git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
|
21
|
+
end
|
22
|
+
spec.bindir = "exe"
|
23
|
+
spec.executables = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
|
24
|
+
spec.require_paths = ["lib"]
|
25
|
+
|
26
|
+
spec.add_development_dependency "bundler", "~> 1.17"
|
27
|
+
spec.add_development_dependency "rake", "~> 12.3.3"
|
28
|
+
spec.add_development_dependency "minitest", "~> 5.0"
|
29
|
+
spec.add_development_dependency "minitest-reporters"
|
30
|
+
spec.add_development_dependency "activesupport", "~> 6.0.3.2"
|
31
|
+
end
|
data/bin/console
ADDED
@@ -0,0 +1,14 @@
|
|
1
|
+
#!/usr/bin/env ruby
|
2
|
+
|
3
|
+
require "bundler/setup"
|
4
|
+
require "amazon-textract-parser-ruby"
|
5
|
+
|
6
|
+
# You can add fixtures and/or initialization code here to make experimenting
|
7
|
+
# with your gem easier. You can also use a different console, if you like.
|
8
|
+
|
9
|
+
# (If you use this, don't forget to add pry to your Gemfile!)
|
10
|
+
# require "pry"
|
11
|
+
# Pry.start
|
12
|
+
|
13
|
+
require "irb"
|
14
|
+
IRB.start(__FILE__)
|
data/bin/setup
ADDED
data/lib/.DS_Store
ADDED
Binary file
|
@@ -0,0 +1,590 @@
|
|
1
|
+
require "amazon-textract-parser-ruby/version"
|
2
|
+
|
3
|
+
module AmazonTRP
|
4
|
+
class Error < StandardError; end
|
5
|
+
|
6
|
+
def AmazonTRP.stable_sort_by(e)
|
7
|
+
e.sort_by.with_index { |x, idx| [yield(x), idx] }
|
8
|
+
end
|
9
|
+
|
10
|
+
|
11
|
+
class BoundingBox
|
12
|
+
attr_reader :width
|
13
|
+
attr_reader :height
|
14
|
+
attr_reader :left
|
15
|
+
attr_reader :top
|
16
|
+
|
17
|
+
def initialize(width, height, left, top)
|
18
|
+
@width = width
|
19
|
+
@height = height
|
20
|
+
@left = left
|
21
|
+
@top = top
|
22
|
+
end
|
23
|
+
|
24
|
+
def to_s
|
25
|
+
"width: #{@width}, height: #{@height}, left: #{@left}, top: #{@top}"
|
26
|
+
end
|
27
|
+
|
28
|
+
def right
|
29
|
+
@left + @width
|
30
|
+
end
|
31
|
+
|
32
|
+
def bottom
|
33
|
+
@top + @height
|
34
|
+
end
|
35
|
+
end
|
36
|
+
|
37
|
+
|
38
|
+
class Point
|
39
|
+
attr_reader :x
|
40
|
+
attr_reader :y
|
41
|
+
|
42
|
+
def initialize(x, y)
|
43
|
+
@x = x
|
44
|
+
@y = y
|
45
|
+
end
|
46
|
+
|
47
|
+
def to_s
|
48
|
+
"(#{@x}, #{@y})"
|
49
|
+
end
|
50
|
+
end
|
51
|
+
|
52
|
+
|
53
|
+
class Geometry
|
54
|
+
attr_reader :boundingBox
|
55
|
+
attr_reader :polygon
|
56
|
+
|
57
|
+
def initialize(geometry)
|
58
|
+
bbox = geometry[:bounding_box]
|
59
|
+
pg = geometry[:polygon]
|
60
|
+
@boundingBox = BoundingBox.new(bbox[:width], bbox[:height], bbox[:left], bbox[:top])
|
61
|
+
@polygon = pg.map{|p| Point.new(p[:x], p[:y])}
|
62
|
+
end
|
63
|
+
|
64
|
+
def to_s
|
65
|
+
"BoundingBox: #{@bounding_box}"
|
66
|
+
end
|
67
|
+
end
|
68
|
+
|
69
|
+
|
70
|
+
class Word
|
71
|
+
attr_reader :confidence
|
72
|
+
attr_reader :geometry
|
73
|
+
attr_reader :id
|
74
|
+
attr_reader :text
|
75
|
+
attr_reader :block
|
76
|
+
|
77
|
+
def initialize(block, blockMap)
|
78
|
+
@block = block
|
79
|
+
@confidence = block[:confidence]
|
80
|
+
@geometry = Geometry.new(block[:geometry])
|
81
|
+
@id = block[:id]
|
82
|
+
@text = block[:text] || ""
|
83
|
+
end
|
84
|
+
|
85
|
+
def to_s
|
86
|
+
@text
|
87
|
+
end
|
88
|
+
end
|
89
|
+
|
90
|
+
|
91
|
+
class Line
|
92
|
+
attr_reader :confidence
|
93
|
+
attr_reader :geometry
|
94
|
+
attr_reader :id
|
95
|
+
attr_reader :words
|
96
|
+
attr_reader :text
|
97
|
+
attr_reader :block
|
98
|
+
|
99
|
+
def initialize(block, blockMap)
|
100
|
+
@block = block
|
101
|
+
@confidence = block[:confidence]
|
102
|
+
@geometry = Geometry.new(block[:geometry])
|
103
|
+
@id = block[:id]
|
104
|
+
|
105
|
+
@text = block[:text] || ""
|
106
|
+
|
107
|
+
@words = []
|
108
|
+
if block[:relationships]
|
109
|
+
block[:relationships].each do |rs|
|
110
|
+
if rs[:type] == 'CHILD'
|
111
|
+
rs[:ids].each do |cid|
|
112
|
+
if blockMap[cid][:block_type] == "WORD"
|
113
|
+
@words.append(Word.new(blockMap[cid], blockMap))
|
114
|
+
end
|
115
|
+
end
|
116
|
+
end
|
117
|
+
end
|
118
|
+
end
|
119
|
+
end
|
120
|
+
|
121
|
+
def to_s
|
122
|
+
s = "Line: "
|
123
|
+
s = s + @text + "\n"
|
124
|
+
s = s + "Words: "
|
125
|
+
@words.each do |word|
|
126
|
+
s = s + "[#{word}]"
|
127
|
+
end
|
128
|
+
return s
|
129
|
+
end
|
130
|
+
end
|
131
|
+
|
132
|
+
|
133
|
+
class SelectionElement
|
134
|
+
attr_reader :confidence
|
135
|
+
attr_reader :geometry
|
136
|
+
attr_reader :id
|
137
|
+
attr_reader :selectionStatus
|
138
|
+
|
139
|
+
def initialize(block, blockMap)
|
140
|
+
@confidence = block[:confidence]
|
141
|
+
@geometry = Geometry.new(block[:geometry])
|
142
|
+
@id = block[:id]
|
143
|
+
@selectionStatus = block[:selection_status]
|
144
|
+
end
|
145
|
+
end
|
146
|
+
|
147
|
+
|
148
|
+
class FieldKey
|
149
|
+
attr_reader :confidence
|
150
|
+
attr_reader :geometry
|
151
|
+
attr_reader :id
|
152
|
+
attr_reader :content
|
153
|
+
attr_reader :text
|
154
|
+
attr_reader :block
|
155
|
+
|
156
|
+
def initialize(block, children, blockMap)
|
157
|
+
@block = block
|
158
|
+
@confidence = block[:confidence]
|
159
|
+
@geometry = Geometry.new(block[:geometry])
|
160
|
+
@id = block[:id]
|
161
|
+
@text = ""
|
162
|
+
@content = []
|
163
|
+
|
164
|
+
t = []
|
165
|
+
children.each do |eid|
|
166
|
+
wb = blockMap[eid]
|
167
|
+
if wb[:block_type] == "WORD"
|
168
|
+
w = Word.new(wb, blockMap)
|
169
|
+
@content.append(w)
|
170
|
+
t.append(w.text)
|
171
|
+
end
|
172
|
+
end
|
173
|
+
@text = t.join(' ') if t
|
174
|
+
end
|
175
|
+
|
176
|
+
def to_s
|
177
|
+
@text
|
178
|
+
end
|
179
|
+
end
|
180
|
+
|
181
|
+
|
182
|
+
class FieldValue
|
183
|
+
attr_reader :confidence
|
184
|
+
attr_reader :geometry
|
185
|
+
attr_reader :id
|
186
|
+
attr_reader :content
|
187
|
+
attr_reader :text
|
188
|
+
attr_reader :block
|
189
|
+
|
190
|
+
def initialize(block, children, blockMap)
|
191
|
+
@block = block
|
192
|
+
@confidence = block[:confidence]
|
193
|
+
@geometry = Geometry.new(block[:geometry])
|
194
|
+
@id = block[:id]
|
195
|
+
@text = ""
|
196
|
+
@content = []
|
197
|
+
|
198
|
+
t = []
|
199
|
+
children.each do |eid|
|
200
|
+
wb = blockMap[eid]
|
201
|
+
if wb[:block_type] == "WORD"
|
202
|
+
w = Word.new(wb, blockMap)
|
203
|
+
@content.append(w)
|
204
|
+
t.append(w.text)
|
205
|
+
elsif wb[:block_type] == "SELECTION_ELEMENT"
|
206
|
+
se = SelectionElement.new(wb, blockMap)
|
207
|
+
@content.append(se)
|
208
|
+
t.append(se.selectionStatus)
|
209
|
+
end
|
210
|
+
end
|
211
|
+
|
212
|
+
@text = t.join(' ') if t
|
213
|
+
end
|
214
|
+
|
215
|
+
def to_s
|
216
|
+
@text
|
217
|
+
end
|
218
|
+
end
|
219
|
+
|
220
|
+
|
221
|
+
class Field
|
222
|
+
attr_reader :key
|
223
|
+
attr_reader :value
|
224
|
+
|
225
|
+
def initialize(block, blockMap)
|
226
|
+
@key = nil
|
227
|
+
@value = nil
|
228
|
+
|
229
|
+
block[:relationships].each do |item|
|
230
|
+
if item[:type] == "CHILD"
|
231
|
+
@key = FieldKey.new(block, item[:ids], blockMap)
|
232
|
+
elsif item[:type] == "VALUE"
|
233
|
+
item[:ids].each do |eid|
|
234
|
+
vkvs = blockMap[eid]
|
235
|
+
if vkvs[:entity_types].include?('VALUE')
|
236
|
+
if vkvs.has_key?(:relationships)
|
237
|
+
vkvs[:relationships].each do |vitem|
|
238
|
+
@value = FieldValue.new(vkvs, vitem[:ids], blockMap) if vitem[:type] == "CHILD"
|
239
|
+
end
|
240
|
+
end
|
241
|
+
end
|
242
|
+
end
|
243
|
+
end
|
244
|
+
end
|
245
|
+
end
|
246
|
+
|
247
|
+
def to_s
|
248
|
+
k = ""
|
249
|
+
v = ""
|
250
|
+
|
251
|
+
k = @key.to_s if @key
|
252
|
+
v = @value.to_s if @value
|
253
|
+
|
254
|
+
return "Field: #{k} = #{v}"
|
255
|
+
end
|
256
|
+
end
|
257
|
+
|
258
|
+
|
259
|
+
class Form
|
260
|
+
attr_reader :fields
|
261
|
+
|
262
|
+
def initialize
|
263
|
+
@fields = []
|
264
|
+
@fieldsMap = {}
|
265
|
+
end
|
266
|
+
|
267
|
+
def addField(field)
|
268
|
+
@fields.append(field)
|
269
|
+
@fieldsMap[field.key.text] = field
|
270
|
+
end
|
271
|
+
|
272
|
+
def to_s
|
273
|
+
s = "Form fields:\n"
|
274
|
+
@fields.each do |field|
|
275
|
+
s = s + field.to_s + "\n"
|
276
|
+
end
|
277
|
+
return s
|
278
|
+
end
|
279
|
+
|
280
|
+
def getFieldByKey(key)
|
281
|
+
@fieldsMap[key]
|
282
|
+
end
|
283
|
+
|
284
|
+
def findFieldsByKey(key)
|
285
|
+
searchKey = key.downcase()
|
286
|
+
results = []
|
287
|
+
@fields.each do |field|
|
288
|
+
if field.key && (field.key.text.downcase.include?(searchKey))
|
289
|
+
results.append(field)
|
290
|
+
end
|
291
|
+
end
|
292
|
+
return results
|
293
|
+
end
|
294
|
+
|
295
|
+
def findFieldByKey(key)
|
296
|
+
fields = findFieldsByKey(key)
|
297
|
+
# Choose the shortest match
|
298
|
+
match = nil
|
299
|
+
matchLength = 0
|
300
|
+
fields.each do |f|
|
301
|
+
if match.nil? || f.key.text.length < matchLength
|
302
|
+
match = f
|
303
|
+
matchLength = f.key.text.length
|
304
|
+
end
|
305
|
+
end
|
306
|
+
return match
|
307
|
+
end
|
308
|
+
end
|
309
|
+
|
310
|
+
|
311
|
+
class Cell
|
312
|
+
attr_reader :confidence
|
313
|
+
attr_reader :rowIndex
|
314
|
+
attr_reader :columnIndex
|
315
|
+
attr_reader :rowSpan
|
316
|
+
attr_reader :columnSpan
|
317
|
+
attr_reader :geometry
|
318
|
+
attr_reader :id
|
319
|
+
attr_reader :content
|
320
|
+
attr_reader :text
|
321
|
+
attr_reader :block
|
322
|
+
|
323
|
+
def initialize(block, blockMap)
|
324
|
+
@block = block
|
325
|
+
@confidence = block[:confidence]
|
326
|
+
@rowIndex = block[:row_index]
|
327
|
+
@columnIndex = block[:column_index]
|
328
|
+
@rowSpan = block[:row_span]
|
329
|
+
@columnSpan = block[:column_span]
|
330
|
+
@geometry = Geometry.new(block[:geometry])
|
331
|
+
@id = block[:id]
|
332
|
+
@content = []
|
333
|
+
@text = ""
|
334
|
+
if block[:relationships]
|
335
|
+
block[:relationships].each do |rs|
|
336
|
+
if rs[:type] == 'CHILD'
|
337
|
+
for cid in rs[:ids]
|
338
|
+
blockType = blockMap[cid][:block_type]
|
339
|
+
if blockType == "WORD"
|
340
|
+
w = Word.new(blockMap[cid], blockMap)
|
341
|
+
@content.append(w)
|
342
|
+
@text = @text + w.text + ' '
|
343
|
+
elsif blockType == "SELECTION_ELEMENT"
|
344
|
+
se = SelectionElement.new(blockMap[cid], blockMap)
|
345
|
+
@content.append(se)
|
346
|
+
@text = @text + se.selectionStatus + ', '
|
347
|
+
end
|
348
|
+
end
|
349
|
+
end
|
350
|
+
end
|
351
|
+
end
|
352
|
+
@text = @text.strip
|
353
|
+
end
|
354
|
+
|
355
|
+
def to_s
|
356
|
+
@text
|
357
|
+
end
|
358
|
+
end
|
359
|
+
|
360
|
+
|
361
|
+
class Row
|
362
|
+
attr_reader :cells
|
363
|
+
|
364
|
+
def initialize
|
365
|
+
@cells = []
|
366
|
+
end
|
367
|
+
|
368
|
+
def to_s
|
369
|
+
s = ""
|
370
|
+
@cells.each do |cell|
|
371
|
+
s = s + "[#{cell}]"
|
372
|
+
end
|
373
|
+
return s
|
374
|
+
end
|
375
|
+
end
|
376
|
+
|
377
|
+
|
378
|
+
class Table
|
379
|
+
attr_reader :confidence
|
380
|
+
attr_reader :geometry
|
381
|
+
attr_reader :id
|
382
|
+
attr_reader :rows
|
383
|
+
attr_reader :block
|
384
|
+
|
385
|
+
def initialize(block, blockMap)
|
386
|
+
@block = block
|
387
|
+
|
388
|
+
@confidence = block[:confidence]
|
389
|
+
@geometry = Geometry.new(block[:geometry])
|
390
|
+
|
391
|
+
@id = block[:id]
|
392
|
+
@rows = []
|
393
|
+
|
394
|
+
ri = 1
|
395
|
+
row = Row.new()
|
396
|
+
cell = nil
|
397
|
+
if block[:relationships]
|
398
|
+
block[:relationships].each do |rs|
|
399
|
+
if rs[:type] == 'CHILD'
|
400
|
+
for cid in rs[:ids]
|
401
|
+
cell = Cell.new(blockMap[cid], blockMap)
|
402
|
+
if cell.rowIndex > ri
|
403
|
+
@rows.append(row)
|
404
|
+
row = Row.new()
|
405
|
+
ri = cell.rowIndex
|
406
|
+
end
|
407
|
+
row.cells.append(cell)
|
408
|
+
end
|
409
|
+
@rows.append(row) if row && row.cells
|
410
|
+
end
|
411
|
+
end
|
412
|
+
end
|
413
|
+
end
|
414
|
+
|
415
|
+
def to_s
|
416
|
+
s = "Table:\n"
|
417
|
+
@rows.each do |row|
|
418
|
+
s = s + row.to_s + "\n"
|
419
|
+
end
|
420
|
+
return s
|
421
|
+
end
|
422
|
+
end
|
423
|
+
|
424
|
+
|
425
|
+
class Page
|
426
|
+
attr_reader :blocks
|
427
|
+
attr_reader :text
|
428
|
+
attr_reader :lines
|
429
|
+
attr_reader :form
|
430
|
+
attr_reader :tables
|
431
|
+
attr_reader :content
|
432
|
+
attr_reader :geometry
|
433
|
+
attr_reader :id
|
434
|
+
|
435
|
+
def initialize(blocks, blockMap)
|
436
|
+
@blocks = blocks
|
437
|
+
@text = ""
|
438
|
+
@lines = []
|
439
|
+
@form = Form.new()
|
440
|
+
@tables = []
|
441
|
+
@content = []
|
442
|
+
|
443
|
+
_parse(blockMap)
|
444
|
+
end
|
445
|
+
|
446
|
+
def to_s
|
447
|
+
s = "Page:\n"
|
448
|
+
@content.each do |item|
|
449
|
+
s = s + item.to_s + "\n"
|
450
|
+
end
|
451
|
+
return s
|
452
|
+
end
|
453
|
+
|
454
|
+
def _parse(blockMap)
|
455
|
+
@blocks.each do |item|
|
456
|
+
if item[:block_type] == "PAGE"
|
457
|
+
@geometry = Geometry.new(item[:geometry])
|
458
|
+
@id = item[:id]
|
459
|
+
elsif item[:block_type] == "LINE"
|
460
|
+
l = Line.new(item, blockMap)
|
461
|
+
@lines.append(l)
|
462
|
+
@content.append(l)
|
463
|
+
@text = @text + l.text + '\n'
|
464
|
+
elsif item[:block_type] == "TABLE"
|
465
|
+
t = Table.new(item, blockMap)
|
466
|
+
@tables.append(t)
|
467
|
+
@content.append(t)
|
468
|
+
elsif item[:block_type] == "KEY_VALUE_SET"
|
469
|
+
if item[:entity_types].include?('KEY')
|
470
|
+
f = Field.new(item, blockMap)
|
471
|
+
if f.key
|
472
|
+
@form.addField(f)
|
473
|
+
@content.append(f)
|
474
|
+
end
|
475
|
+
end
|
476
|
+
end
|
477
|
+
end
|
478
|
+
end
|
479
|
+
|
480
|
+
def getLinesInReadingOrder
|
481
|
+
columns = []
|
482
|
+
lines = []
|
483
|
+
@lines.each do |item|
|
484
|
+
column_found = false
|
485
|
+
columns.each_with_index do |column, index|
|
486
|
+
bbox_left = item.geometry.boundingBox.left
|
487
|
+
bbox_right = item.geometry.boundingBox.right
|
488
|
+
bbox_centre = item.geometry.boundingBox.left + item.geometry.boundingBox.width/2
|
489
|
+
column_centre = column[:left] + ((column[:right] - column[:left]) / 2)
|
490
|
+
if (bbox_centre > column[:left] && bbox_centre < column[:right]) || (column_centre > bbox_left && column_centre < bbox_right)
|
491
|
+
# Bbox appears inside the column
|
492
|
+
lines.append({:column => index, :text => item.text})
|
493
|
+
column_found = true
|
494
|
+
break
|
495
|
+
end
|
496
|
+
end
|
497
|
+
if !column_found
|
498
|
+
columns.append({:left => item.geometry.boundingBox.left, :right => item.geometry.boundingBox.right})
|
499
|
+
lines.append({:column => columns.count - 1, :text => item.text})
|
500
|
+
end
|
501
|
+
end
|
502
|
+
|
503
|
+
return AmazonTRP::stable_sort_by(lines) {|x| x[:column]}
|
504
|
+
end
|
505
|
+
|
506
|
+
def getTextInReadingOrder
|
507
|
+
lines = getLinesInReadingOrder()
|
508
|
+
text = ""
|
509
|
+
lines.each do |line|
|
510
|
+
text = text + line[:text] + "\n"
|
511
|
+
end
|
512
|
+
return text
|
513
|
+
end
|
514
|
+
|
515
|
+
def getLinesInBoundingBox(boundingBox)
|
516
|
+
lines = []
|
517
|
+
@lines.each do |line|
|
518
|
+
line_bbox = line.geometry.boundingBox
|
519
|
+
if (line_bbox.left >= boundingBox.left &&
|
520
|
+
line_bbox.left <= boundingBox.right &&
|
521
|
+
line_bbox.top >= boundingBox.top &&
|
522
|
+
line_bbox.top <= boundingBox.bottom)
|
523
|
+
lines.append(line)
|
524
|
+
end
|
525
|
+
end
|
526
|
+
return lines
|
527
|
+
end
|
528
|
+
end
|
529
|
+
|
530
|
+
|
531
|
+
class Document
|
532
|
+
attr_reader :blocks
|
533
|
+
attr_reader :pageBlocks
|
534
|
+
attr_reader :pages
|
535
|
+
|
536
|
+
def initialize(responsePages)
|
537
|
+
@responsePages = responsePages.is_a?(Array) ? responsePages : [responsePages]
|
538
|
+
@pages = []
|
539
|
+
_parse()
|
540
|
+
end
|
541
|
+
|
542
|
+
def to_s
|
543
|
+
s = "\nDocument:\n"
|
544
|
+
@pages.each do |p|
|
545
|
+
s = s + p.to_s + "\n\n"
|
546
|
+
end
|
547
|
+
return s
|
548
|
+
end
|
549
|
+
|
550
|
+
def _parseDocumentPagesAndBlockMap
|
551
|
+
blockMap = {}
|
552
|
+
|
553
|
+
documentPages = []
|
554
|
+
documentPage = nil
|
555
|
+
@responsePages.each do |page|
|
556
|
+
unless page[:blocks].nil?
|
557
|
+
page[:blocks].each do |block|
|
558
|
+
if block.has_key?(:block_type) && block.has_key?(:id)
|
559
|
+
blockMap[block[:id]] = block
|
560
|
+
end
|
561
|
+
|
562
|
+
if block[:block_type] == 'PAGE'
|
563
|
+
documentPages.append({:blocks => documentPage}) if documentPage
|
564
|
+
documentPage = []
|
565
|
+
documentPage.append(block)
|
566
|
+
else
|
567
|
+
documentPage.append(block)
|
568
|
+
end
|
569
|
+
end
|
570
|
+
end
|
571
|
+
end
|
572
|
+
documentPages.append({:blocks => documentPage}) if documentPage
|
573
|
+
return documentPages, blockMap
|
574
|
+
end
|
575
|
+
|
576
|
+
def _parse
|
577
|
+
@responseDocumentPages, @blockMap = _parseDocumentPagesAndBlockMap()
|
578
|
+
@responseDocumentPages.each do |documentPage|
|
579
|
+
page = Page.new(documentPage[:blocks], @blockMap)
|
580
|
+
@pages.append(page)
|
581
|
+
end
|
582
|
+
end
|
583
|
+
|
584
|
+
def getBlockById(blockId)
|
585
|
+
return @blockMap[blockId] if @blockMap && @blockMap.has_key?(blockId)
|
586
|
+
return nil
|
587
|
+
end
|
588
|
+
end
|
589
|
+
|
590
|
+
end
|
Binary file
|
metadata
ADDED
@@ -0,0 +1,129 @@
|
|
1
|
+
--- !ruby/object:Gem::Specification
|
2
|
+
name: amazon-textract-parser-ruby
|
3
|
+
version: !ruby/object:Gem::Version
|
4
|
+
version: 0.1.0
|
5
|
+
platform: ruby
|
6
|
+
authors:
|
7
|
+
- Niels Vanspauwen
|
8
|
+
autorequire:
|
9
|
+
bindir: exe
|
10
|
+
cert_chain: []
|
11
|
+
date: 2020-07-17 00:00:00.000000000 Z
|
12
|
+
dependencies:
|
13
|
+
- !ruby/object:Gem::Dependency
|
14
|
+
name: bundler
|
15
|
+
requirement: !ruby/object:Gem::Requirement
|
16
|
+
requirements:
|
17
|
+
- - "~>"
|
18
|
+
- !ruby/object:Gem::Version
|
19
|
+
version: '1.17'
|
20
|
+
type: :development
|
21
|
+
prerelease: false
|
22
|
+
version_requirements: !ruby/object:Gem::Requirement
|
23
|
+
requirements:
|
24
|
+
- - "~>"
|
25
|
+
- !ruby/object:Gem::Version
|
26
|
+
version: '1.17'
|
27
|
+
- !ruby/object:Gem::Dependency
|
28
|
+
name: rake
|
29
|
+
requirement: !ruby/object:Gem::Requirement
|
30
|
+
requirements:
|
31
|
+
- - "~>"
|
32
|
+
- !ruby/object:Gem::Version
|
33
|
+
version: 12.3.3
|
34
|
+
type: :development
|
35
|
+
prerelease: false
|
36
|
+
version_requirements: !ruby/object:Gem::Requirement
|
37
|
+
requirements:
|
38
|
+
- - "~>"
|
39
|
+
- !ruby/object:Gem::Version
|
40
|
+
version: 12.3.3
|
41
|
+
- !ruby/object:Gem::Dependency
|
42
|
+
name: minitest
|
43
|
+
requirement: !ruby/object:Gem::Requirement
|
44
|
+
requirements:
|
45
|
+
- - "~>"
|
46
|
+
- !ruby/object:Gem::Version
|
47
|
+
version: '5.0'
|
48
|
+
type: :development
|
49
|
+
prerelease: false
|
50
|
+
version_requirements: !ruby/object:Gem::Requirement
|
51
|
+
requirements:
|
52
|
+
- - "~>"
|
53
|
+
- !ruby/object:Gem::Version
|
54
|
+
version: '5.0'
|
55
|
+
- !ruby/object:Gem::Dependency
|
56
|
+
name: minitest-reporters
|
57
|
+
requirement: !ruby/object:Gem::Requirement
|
58
|
+
requirements:
|
59
|
+
- - ">="
|
60
|
+
- !ruby/object:Gem::Version
|
61
|
+
version: '0'
|
62
|
+
type: :development
|
63
|
+
prerelease: false
|
64
|
+
version_requirements: !ruby/object:Gem::Requirement
|
65
|
+
requirements:
|
66
|
+
- - ">="
|
67
|
+
- !ruby/object:Gem::Version
|
68
|
+
version: '0'
|
69
|
+
- !ruby/object:Gem::Dependency
|
70
|
+
name: activesupport
|
71
|
+
requirement: !ruby/object:Gem::Requirement
|
72
|
+
requirements:
|
73
|
+
- - "~>"
|
74
|
+
- !ruby/object:Gem::Version
|
75
|
+
version: 6.0.3.2
|
76
|
+
type: :development
|
77
|
+
prerelease: false
|
78
|
+
version_requirements: !ruby/object:Gem::Requirement
|
79
|
+
requirements:
|
80
|
+
- - "~>"
|
81
|
+
- !ruby/object:Gem::Version
|
82
|
+
version: 6.0.3.2
|
83
|
+
description: This is a quick Ruby port of https://github.com/mludvig/amazon-textract-parser\nIt's
|
84
|
+
useful for interpreting the result of Amazon Textract info.
|
85
|
+
email:
|
86
|
+
- niels.vanspauwen@gmail.com
|
87
|
+
executables: []
|
88
|
+
extensions: []
|
89
|
+
extra_rdoc_files: []
|
90
|
+
files:
|
91
|
+
- ".DS_Store"
|
92
|
+
- ".gitignore"
|
93
|
+
- ".travis.yml"
|
94
|
+
- Gemfile
|
95
|
+
- Gemfile.lock
|
96
|
+
- LICENSE.txt
|
97
|
+
- README.md
|
98
|
+
- Rakefile
|
99
|
+
- amazon-textract-parser-ruby.gemspec
|
100
|
+
- bin/console
|
101
|
+
- bin/setup
|
102
|
+
- lib/.DS_Store
|
103
|
+
- lib/amazon-textract-parser-ruby.rb
|
104
|
+
- lib/amazon-textract-parser-ruby/.DS_Store
|
105
|
+
- lib/amazon-textract-parser-ruby/version.rb
|
106
|
+
homepage: https://github.com/nielsvanspauwen/amazon-textract-parser-ruby
|
107
|
+
licenses:
|
108
|
+
- MIT
|
109
|
+
metadata: {}
|
110
|
+
post_install_message:
|
111
|
+
rdoc_options: []
|
112
|
+
require_paths:
|
113
|
+
- lib
|
114
|
+
required_ruby_version: !ruby/object:Gem::Requirement
|
115
|
+
requirements:
|
116
|
+
- - ">="
|
117
|
+
- !ruby/object:Gem::Version
|
118
|
+
version: '0'
|
119
|
+
required_rubygems_version: !ruby/object:Gem::Requirement
|
120
|
+
requirements:
|
121
|
+
- - ">="
|
122
|
+
- !ruby/object:Gem::Version
|
123
|
+
version: '0'
|
124
|
+
requirements: []
|
125
|
+
rubygems_version: 3.0.1
|
126
|
+
signing_key:
|
127
|
+
specification_version: 4
|
128
|
+
summary: Amazon Textract Results Parser
|
129
|
+
test_files: []
|