rtesseract 3.0.3 → 3.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 8eb694ea4b37475f756451795c145b8ba9618a0e5e94b0774a90301bb1aa97a2
-  data.tar.gz: fe725c774fc39720ff830e47e4580f7335def9ece76f67fad77fe9722f415d6c
+  metadata.gz: 87f85a70ab24a03a719e7726d95debeb469d822c447acc39a8adc1579c43e6d1
+  data.tar.gz: a18cd83cd4632ed5adc096f9384659f0e47a4ba33e2f8a76e804cf8140e8798b
 SHA512:
-  metadata.gz: f40f8b53fc3c63e4968d9b1adab5153771730897fd681e354afea79f2007c28c4216dc61cad0d218dc7d76814360ae9169a0ffb8999ab6c8d15ef51ad712ec07
-  data.tar.gz: e2dfbf63b972c6e678d4bb79ec00064a958c9607f76cffe197bc26f555dff35bd7b0e1c263e376ede9598e2f429a344abb4b8d39b0e47ede6d1c09a32e6c44a4
+  metadata.gz: 4f6b37e1645d5f82c759e5feea081cfcf5384e2ad91423d2d04030a89ef007eb9469ecf398ba4cd2364bf2baeea4bad885866c4c41b19fa4da2c078b79cef4a6
+  data.tar.gz: bea642e1d7d2576dbdaeeb68ff730627104894d914a405a4ec2a37b8c65755f2f9c33a5091e16eaced51162725b6ea62f647b764ec838eefe3bfe4f58f461986

data/.github/workflows/ci.yml

@@ -0,0 +1,31 @@
+name: CI
+on: [push]
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        ruby:
+          - '2.5.x'
+          - '2.6.x'
+          - '2.7.x'
+    steps:
+      - uses: actions/checkout@v2
+      - name: Install tesseract-ocr
+        run: |
+          sudo add-apt-repository ppa:alex-p/tesseract-ocr -y
+          sudo apt-get update -q
+          sudo apt-get install tesseract-ocr tesseract-ocr-eng ghostscript -y
+      - name: Setup Ruby
+        uses: actions/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby }}
+      - name: Bundle
+        env:
+          MTSR_RAILS_VERSION: ${{ matrix.rails }}
+        run: |
+          gem uninstall -aIx bundler
+          gem install bundler
+          bundle install --jobs 4 --retry 3
+      - name: Test
+        run: bundle exec rake

data/.hound.yml

@@ -0,0 +1,3 @@
+rubocop:
+  config_file: .rubocop.yml
+  version: 0.80.0

data/.rubocop.yml

@@ -1,8 +1,20 @@
-Documentation:
-  Enabled: false
 
-Metrics/LineLength:
+Layout/LineLength:
   Max: 150
 
 Metrics/BlockLength:
   Max: 50
+
+Metrics/AbcSize:
+  Max: 30
+
+Style/Documentation:
+  Enabled: false
+Style/HashEachMethods:
+  Enabled: true
+
+Style/HashTransformKeys:
+  Enabled: true
+
+Style/HashTransformValues:
+  Enabled: true

data/CHANGELOG.md

@@ -1,3 +1,39 @@
+## v3.1.2
+
+#### Added
+
+* Added confidence for each word in box mode
+
+## v3.1.1
+
+#### Changed
+
+* Changed RTesseract::Command to receive a block when success run
+
+## v3.1.0
+
+#### Changed
+
+* Removed nokogi dependency.
+
+## v3.0.4
+
+#### Changed
+
+* Updated dependencies by security alerts.
+
+## v3.0.5
+
+#### Changed
+
+* Updated dependencies by security alerts.
+
+## v3.0.4
+
+#### Changed
+
+* Updated dependencies by security alerts.
+
 ## v3.0.3
 
 #### Changed

data/Gemfile

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 source 'https://rubygems.org'
 
 git_source(:github) { |repo_name| "https://github.com/#{repo_name}" }

data/Gemfile.lock

@@ -1,58 +1,55 @@
 PATH
   remote: .
   specs:
-    rtesseract (3.0.3)
-      nokogiri
+    rtesseract (3.1.2)
 
 GEM
   remote: https://rubygems.org/
   specs:
-    coveralls (0.8.22)
+    coveralls (0.8.23)
       json (>= 1.8, < 3)
       simplecov (~> 0.16.1)
       term-ansicolor (~> 1.3)
-      thor (~> 0.19.4)
+      thor (>= 0.19.4, < 2.0)
       tins (~> 1.6)
     diff-lcs (1.3)
-    docile (1.3.1)
-    json (2.1.0)
-    mini_portile2 (2.4.0)
-    nokogiri (1.10.1)
-      mini_portile2 (~> 2.4.0)
-    rake (10.5.0)
-    rspec (3.8.0)
-      rspec-core (~> 3.8.0)
-      rspec-expectations (~> 3.8.0)
-      rspec-mocks (~> 3.8.0)
-    rspec-core (3.8.0)
-      rspec-support (~> 3.8.0)
-    rspec-expectations (3.8.2)
+    docile (1.3.2)
+    json (2.3.0)
+    rake (13.0.1)
+    rspec (3.9.0)
+      rspec-core (~> 3.9.0)
+      rspec-expectations (~> 3.9.0)
+      rspec-mocks (~> 3.9.0)
+    rspec-core (3.9.1)
+      rspec-support (~> 3.9.1)
+    rspec-expectations (3.9.1)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.8.0)
-    rspec-mocks (3.8.0)
+      rspec-support (~> 3.9.0)
+    rspec-mocks (3.9.1)
       diff-lcs (>= 1.2.0, < 2.0)
-      rspec-support (~> 3.8.0)
-    rspec-support (3.8.0)
+      rspec-support (~> 3.9.0)
+    rspec-support (3.9.2)
     simplecov (0.16.1)
       docile (~> 1.1)
       json (>= 1.8, < 3)
       simplecov-html (~> 0.10.0)
     simplecov-html (0.10.2)
-    term-ansicolor (1.7.0)
+    sync (0.5.0)
+    term-ansicolor (1.7.1)
       tins (~> 1.0)
-    thor (0.19.4)
-    tins (1.20.2)
+    thor (1.0.1)
+    tins (1.24.1)
+      sync
 
 PLATFORMS
   ruby
 
 DEPENDENCIES
-  bundler (~> 1.17)
+  bundler (~> 2)
   coveralls
-  rake (~> 10.0)
-  rspec (~> 3.0)
+  rake
+  rspec
   rtesseract!
-  simplecov
 
 BUNDLED WITH
-   1.17.2
+   2.1.4

data/README.md

@@ -3,8 +3,8 @@
 <a href='http://badge.fury.io/rb/rtesseract'>
     <img src="https://badge.fury.io/rb/rtesseract.png" alt="Gem Version" />
 </a>
-<a href='https://travis-ci.org/dannnylo/rtesseract'>
-    <img src="https://travis-ci.org/dannnylo/rtesseract.png?branch=master" alt="Build Status" />
+<a href='https://github.com/dannnylo/rtesseract/workflows/CI/badge.svg'>
+  <img src="https://github.com/dannnylo/rtesseract/workflows/CI/badge.svg" alt="Build Status" />
 </a>
 <a href='https://coveralls.io/r/dannnylo/rtesseract?branch=master'>
     <img src="https://coveralls.io/repos/dannnylo/rtesseract/badge.png?branch=master" alt="Coverage Status" />
@@ -100,20 +100,20 @@ This will preserve the image colors, pictures and structure in the generated pdf
   ```ruby
     RTesseract.new('test_words.png').to_box
     => [
-      {:word => 'If',  :x_start=>52, :y_start=>13, :x_end=>63, :y_end=>27},
-      {:word => 'you', :x_start=>69, :y_start=>17, :x_end=>100, :y_end=>31},
-      {:word => 'are', :x_start=>108, :y_start=>17, :x_end=>136, :y_end=>27},
-      {:word => 'a', :x_start=>143, :y_start=>17, :x_end=>151, :y_end=>27},
-      {:word => 'friend,', :x_start=>158, :y_start=>13, :x_end=>214, :y_end=>29},
-      {:word => 'you', :x_start=>51, :y_start=>39, :x_end=>82, :y_end=>53},
-      {:word => 'speak', :x_start=>90, :y_start=>35, :x_end=>140, :y_end=>53},
-      {:word => 'the', :x_start=>146, :y_start=>35, :x_end=>174, :y_end=>49},
-      {:word => 'password,', :x_start=>182, :y_start=>35, :x_end=>267, :y_end=>53},
-      {:word => 'and', :x_start=>51, :y_start=>57, :x_end=>81, :y_end=>71},
-      {:word => 'the', :x_start=>89, :y_start=>57, :x_end=>117, :y_end=>71},
-      {:word => 'doors', :x_start=>124, :y_start=>57, :x_end=>172, :y_end=>71},
-      {:word => 'will', :x_start=>180, :y_start=>57, :x_end=>208, :y_end=>71},
-      {:word => 'open.', :x_start=>216, :y_start=>61, :x_end=>263, :y_end=>75}
+      { :word => 'If', :confidence=>89, :x_start=>52, :y_start=>13, :x_end=>63, :y_end=>27},
+      { :word => 'you', :confidence=>96, :x_start=>69, :y_start=>17, :x_end=>100, :y_end=>31},
+      { :word => 'are', :confidence=>92, :x_start=>108, :y_start=>17, :x_end=>136, :y_end=>27},
+      { :word => 'a', :confidence=>92, :x_start=>133, :y_start=>8, :x_end=>147, :y_end=>35},
+      { :word => 'friend,', :confidence=>95, :x_start=>158, :y_start=>13, :x_end=>214, :y_end=>29},
+      { :word => 'you', :confidence=>96, :x_start=>51, :y_start=>39, :x_end=>82, :y_end=>53},
+      { :word => 'speak', :confidence=>96, :x_start=>90, :y_start=>35, :x_end=>140, :y_end=>53},
+      { :word => 'the', :confidence=>96, :x_start=>146, :y_start=>35, :x_end=>174, :y_end=>49},
+      { :word => 'password,', :confidence=>96, :x_start=>182, :y_start=>35, :x_end=>267, :y_end=>53},
+      { :word => 'and', :confidence=>96, :x_start=>51, :y_start=>57, :x_end=>81, :y_end=>71},
+      { :word => 'the', :confidence=>96, :x_start=>89, :y_start=>57, :x_end=>117, :y_end=>71},
+      { :word => 'doors', :confidence=>96, :x_start=>124, :y_start=>57, :x_end=>172, :y_end=>71},
+      { :word => 'will', :confidence=>96, :x_start=>180, :y_start=>57, :x_end=>208, :y_end=>71},
+      { :word => 'open.', :confidence=>96, :x_start=>216, :y_start=>61, :x_end=>263, :y_end=>75}
     ]
   ```
 

data/Rakefile

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'bundler/gem_tasks'
 require 'rspec/core/rake_task'
 

data/bin/console

@@ -1,4 +1,5 @@
 #!/usr/bin/env ruby
+# frozen_string_literal: true
 
 require 'bundler/setup'
 require 'rtesseract'

data/lib/rtesseract.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'rtesseract/check'
 require 'rtesseract/configuration'
 require 'rtesseract/command'

data/lib/rtesseract/base.rb

@@ -1,13 +1,13 @@
+# frozen_string_literal: true
+
 require 'tmpdir'
 require 'securerandom'
 require 'pathname'
 
 class RTesseract
   module Base
-    def temp_file(ext = '')
-      @rand_file ||= "rtesseract_#{SecureRandom.uuid}"
-
-      Pathname.new(Dir.tmpdir).join("#{@rand_file}#{ext}").to_s
+    def temp_file_path
+      Pathname.new(Dir.tmpdir).join("rtesseract_#{SecureRandom.uuid}").to_s
     end
   end
 end

data/lib/rtesseract/box.rb

@@ -1,33 +1,50 @@
-require 'nokogiri'
+# frozen_string_literal: true
 
 class RTesseract
   module Box
     extend RTesseract::Base
 
-    def self.run(source, errors, options)
-      options.tessedit_create_hocr = 1
+    class << self
+      def run(source, errors, options)
+        options.tessedit_create_hocr = 1
 
-      RTesseract::Command.new(source, temp_file, errors, options).run
+        RTesseract::Command.new(source, temp_file_path, errors, options).run do |output_path|
+          parse(File.read("#{output_path}.hocr"))
+        end
+      end
 
-      parse(File.read(temp_file('.hocr')))
-    end
+      def parse(content)
+        content.lines.map { |line| parse_line(line) }.compact
+      end
+
+      def parse_line(line)
+        return unless line.match?(/oc(rx|r)_word/)
+
+        word = line.match(/(?<=>)(.*?)(?=<)/).to_s
+
+        return if word.strip == ''
 
-    def self.parse(content)
-      html = Nokogiri::HTML(content)
-      html.css('span.ocrx_word, span.ocr_word').map do |word|
-        attributes = word.attributes['title'].value.to_s.delete(';').split(' ')
-        word_info(word, attributes)
+        word_info(word, parse_position(line), parse_confidence(line))
+      end
+
+      def word_info(word, positions, confidence)
+        {
+          word: word,
+          confidence: confidence[-1].to_i,
+          x_start: positions[1].to_i,
+          y_start: positions[2].to_i,
+          x_end: positions[3].to_i,
+          y_end: positions[4].to_i
+        }
       end
-    end
 
-    def self.word_info(word, data)
-      {
-        word: word.text,
-        x_start: data[1].to_i,
-        y_start: data[2].to_i,
-        x_end: data[3].to_i,
-        y_end: data[4].to_i
-      }
+      def parse_position(line)
+        line.match(/(?<=title)(.*?)(?=;)/).to_s.split(' ')
+      end
+
+      def parse_confidence(line)
+        line.match(/(?<=;)(.*?)(?=')/).to_s.split(' ')
+      end
     end
   end
 end

data/lib/rtesseract/check.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class RTesseract
   class << self
     def tesseract_version

data/lib/rtesseract/command.rb

@@ -1,24 +1,26 @@
+# frozen_string_literal: true
+
 class RTesseract
   class Command
     FIXED = %i[command psm oem lang tessdata_dir user_words user_patterns config_file].freeze
 
     attr_reader :options
 
-    def initialize(source, output, errors, options)
+    def initialize(source, output_path, errors, options)
       @source = source
-      @output = output
+      @output_path = output_path
       @options = options
       @errors = errors
-      @full_command = [options.command, @source, @output]
+      @full_command = [options.command, @source, @output_path]
     end
 
     def full_command
       add_option('--psm', options.psm)
       add_option('--oem', options.oem)
       add_option('-l', options.lang)
-      add_option('--tessdata_dir', options.tessdata_dir)
-      add_option('--user_words', options.user_words)
-      add_option('--user_patterns', options.user_patterns)
+      add_option('--tessdata-dir', options.tessdata_dir)
+      add_option('--user-words', options.user_words)
+      add_option('--user-patterns', options.user_patterns)
 
       other_configs
 
@@ -46,7 +48,11 @@ class RTesseract
 
       @errors.push(error)
 
-      return output if status.success?
+      if status.success?
+        return yield(@output_path) if block_given?
+
+        return output
+      end
 
       raise RTesseract::Error, error
     end

data/lib/rtesseract/configuration.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'ostruct'
 
 class RTesseract

data/lib/rtesseract/pdf.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class RTesseract
   module Pdf
     extend Base
@@ -5,9 +7,9 @@ class RTesseract
     def self.run(source, errors, options)
       options.tessedit_create_pdf = 1
 
-      RTesseract::Command.new(source, temp_file, errors, options).run
-
-      File.open(temp_file('.pdf'), 'r')
+      RTesseract::Command.new(source, temp_file_path, errors, options).run do |output_path|
+        File.open("#{output_path}.pdf", 'r')
+      end
     end
   end
 end

data/lib/rtesseract/text.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'open3'
 
 class RTesseract

data/lib/rtesseract/tsv.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class RTesseract
   module Tsv
     extend Base
@@ -5,9 +7,9 @@ class RTesseract
     def self.run(source, errors, options)
       options.tessedit_create_tsv = 1
 
-      RTesseract::Command.new(source, temp_file, errors, options).run
-
-      File.open(temp_file('.tsv'), 'r')
+      RTesseract::Command.new(source, temp_file_path, errors, options).run do |output_path|
+        File.open("#{output_path}.tsv", 'r')
+      end
     end
   end
 end

data/lib/rtesseract/version.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class RTesseract
-  VERSION = '3.0.3'.freeze
+  VERSION = '3.1.2'
 end

data/rtesseract.gemspec

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 lib = File.expand_path('lib', __dir__)
 $LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
 require 'rtesseract/version'
@@ -8,9 +10,9 @@ Gem::Specification.new do |spec|
   spec.authors       = ['Danilo Jeremias da Silva']
   spec.email         = ['dannnylo@gmail.com']
 
-  spec.summary       = 'Ruby library for working with the Tesseract OCR.'.freeze
-  spec.description   = 'Ruby library for working with the Tesseract OCR.'.freeze
-  spec.homepage      = 'http://github.com/dannnylo/rtesseract'.freeze
+  spec.summary       = 'Ruby library for working with the Tesseract OCR.'
+  spec.description   = 'Ruby library for working with the Tesseract OCR.'
+  spec.homepage      = 'http://github.com/dannnylo/rtesseract'
   spec.license       = 'MIT'
 
   # Specify which files should be added to the gem when it is released.
@@ -22,11 +24,8 @@ Gem::Specification.new do |spec|
   spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
   spec.require_paths = ['lib']
 
-  spec.add_development_dependency 'bundler', '~> 1.17'
+  spec.add_development_dependency 'bundler', '~> 2'
   spec.add_development_dependency 'coveralls'
-  spec.add_development_dependency 'rake', '~> 10.0'
-  spec.add_development_dependency 'rspec', '~> 3.0'
-  spec.add_development_dependency 'simplecov'
-
-  spec.add_dependency 'nokogiri'
+  spec.add_development_dependency 'rake'
+  spec.add_development_dependency 'rspec'
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rtesseract
 version: !ruby/object:Gem::Version
-  version: 3.0.3
+  version: 3.1.2
 platform: ruby
 authors:
 - Danilo Jeremias da Silva
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2019-03-19 00:00:00.000000000 Z
+date: 2020-08-23 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -16,14 +16,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.17'
+        version: '2'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.17'
+        version: '2'
 - !ruby/object:Gem::Dependency
   name: coveralls
   requirement: !ruby/object:Gem::Requirement
@@ -40,34 +40,6 @@ dependencies:
         version: '0'
 - !ruby/object:Gem::Dependency
   name: rake
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '10.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '10.0'
-- !ruby/object:Gem::Dependency
-  name: rspec
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.0'
-  type: :development
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - "~>"
-      - !ruby/object:Gem::Version
-        version: '3.0'
-- !ruby/object:Gem::Dependency
-  name: simplecov
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
@@ -81,13 +53,13 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '0'
 - !ruby/object:Gem::Dependency
-  name: nokogiri
+  name: rspec
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ">="
       - !ruby/object:Gem::Version
         version: '0'
-  type: :runtime
+  type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
@@ -102,10 +74,11 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".document"
+- ".github/workflows/ci.yml"
 - ".gitignore"
+- ".hound.yml"
 - ".rspec"
 - ".rubocop.yml"
-- ".travis.yml"
 - CHANGELOG.md
 - CODE_OF_CONDUCT.md
 - Gemfile
@@ -145,8 +118,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubyforge_project: 
-rubygems_version: 2.7.8
+rubygems_version: 3.1.2
 signing_key: 
 specification_version: 4
 summary: Ruby library for working with the Tesseract OCR.

data/.travis.yml

@@ -1,17 +0,0 @@
----
-
-sudo: false
-dist: bionic
-language: ruby
-cache: bundler
-
-before_install:
-  - sudo add-apt-repository ppa:alex-p/tesseract-ocr -y
-  - sudo apt-get update -q
-  - sudo apt-get install tesseract-ocr tesseract-ocr-eng ghostscript -y
-  - gem install bundler -v 1.17.1
-
-rvm:
-  - 2.4.5
-  - 2.5.3
-  - 2.6.0