tesseract-ocr 0.1.7 → 0.1.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 8ba69437f990d1dec349c582d119d1d8a5007477
-  data.tar.gz: 7e536618d411e1e568afb1a57958b10074d8c5db
+  metadata.gz: 15d5c5417af8a215178f247ebca4750d2a7cbbcb
+  data.tar.gz: 35e1a440418421cc2b22f2f7f8e095600911e04c
 SHA512:
-  metadata.gz: c6509845536b6acc5451974756bc006bbdea858968f3cd793d3d59f57973eb8f777856ab230e9935caccb902ec73a4444de47527132d877b495fa30538dfd77b
-  data.tar.gz: 3528e0185fbf1c8c57284baa753e9e75c886f12952ebc2fb69169d9b08ea89a99b0b0df915b5f18e5dc00958baadbaabe696146d3a33284e3186eee100bd5078
+  metadata.gz: 71064917075114639018f60de081afb1cf164677b1cd9bbec940fe1f34063b0ef0a823f3ee0f2dba434f03775d000e4e50b7a17f93285b79d9982fa2d563f475
+  data.tar.gz: 188d4e48122324899a2d83b97585a932938a9905b4de3960c6496748ee4b352707a9e80455905da446b469afba47c8e87c9337270130b8972309322e8bfd5576

data/lib/tesseract/api.rb

@@ -99,8 +99,8 @@ class API
 		end
 	end
 
-	def init (datapath = Tesseract.prefix || '.', language = 'eng', mode = :DEFAULT)
-		unless C::BaseAPI.init(to_ffi, datapath, language.to_s, mode).zero?
+	def init (datapath = nil, language = 'eng', mode = :DEFAULT)
+		unless C::BaseAPI.init(to_ffi, datapath || Tesseract.prefix || '/usr/share', language.to_s, mode).zero?
 			raise 'the API did not Init correctly'
 		end
 	end
@@ -166,6 +166,19 @@ class API
 		C.free_array_of_char(pointer) unless pointer.null?
 	end
 
+	def get_hocr(page = 0)
+		pointer = C::BaseAPI.get_hocr_text(to_ffi, page)
+
+		return if pointer.null?
+
+		result = pointer.read_string
+		result.force_encoding 'UTF-8'
+
+		result
+	ensure
+		C.free_array_of_char(pointer) unless pointer.null?
+	end
+
 	def get_box (page = 0)
 		pointer = C::BaseAPI.get_box_text(to_ffi, page)
 		result  = pointer.read_string

data/lib/tesseract/c/baseapi.rb

@@ -202,6 +202,12 @@ module BaseAPI
 			}
 		}, blocking: true
 
+		cpp.function %{
+			char* get_hocr_text (TessBaseAPI* api, int page_number) {
+				return api->GetHOCRText(page_number);
+			}
+		}, blocking: true
+
 		cpp.function %{
 			char* get_box_text (TessBaseAPI* api, int page_number) {
 				return api->GetBoxText(page_number);

data/lib/tesseract/engine.rb

@@ -32,9 +32,9 @@ class Engine
 	attr_reader :config
 
 	named :path, :language, :mode, :variables,
-		:optional => { :path => '.', :language => :eng, :mode => :DEFAULT, :variables => {}, :config => [] },
+		:optional => { :path => nil, :language => :eng, :mode => :DEFAULT, :variables => {}, :config => [] },
 		:alias    => { :data => :path, :lang => :language }
-	def initialize (path = '.', language = :eng, mode = :DEFAULT, variables = {}, config = [], &block) # :yields: self
+	def initialize (path = nil, language = :eng, mode = :DEFAULT, variables = {}, config = [], &block) # :yields: self
 		@api = API.new
 
 		@initializing = true
@@ -171,6 +171,26 @@ class Engine
 		text_at
 	end
 
+	named :image, :x, :y, :width, :height,
+		:optional => 0 .. -1,
+		:alias    => { :w => :width, :h => :height }
+	def hocr_for (image = nil, x = nil, y = nil, width = nil, height = nil, page = nil)
+		_setup(image, x, y, width, height)
+
+		@api.get_hocr(page || 0)
+	end
+
+	named :x, :y, :width, :height,
+		:optional => 0 .. -1,
+		:alias    => { :w => :width, :h => :height }
+	def hocr_at (x = nil, y = nil, width = nil, height = nil, page = nil)
+		hocr_for(nil, x, y, width, height, page)
+	end
+
+	def hocr
+		hocr_at
+	end
+
 	%w(block paragraph line word symbol).each {|level|
 		define_method "each_#{level}" do |&block|
 			raise ArgumentError, 'you have to pass a block' unless block
@@ -231,7 +251,7 @@ protected
 	def _init
 		@api.end
 
-		@api.init(File.expand_path(@path), API.to_language_code(@language), @mode)
+		@api.init(@path, API.to_language_code(@language), @mode)
 
 		@variables.each {|name, value|
 			@api.set_variable(name.to_s, value.to_s)

data/lib/tesseract/version.rb

@@ -24,6 +24,6 @@
 
 module Tesseract
 	def self.version
-		'0.1.7'
+		'0.1.8'
 	end
 end

data/test/tesseract_spec.rb

@@ -9,36 +9,38 @@ describe Tesseract::Engine do
 
 	describe '#text_for' do
 		it 'can read the first test image' do
-			engine.text_for('first.png').strip.should == 'ABC'
+			expect(engine.text_for('first.png').strip).to eq('ABC')
 		end
 
 		it 'can read the second test image' do
-			engine.text_for('second.png').strip.should == "#{Tesseract::API.new.version == '3.01' ? ?| : ?I}'m 12 and what is this.\nINSTALL GENTOO\nOH HAI 1234"
+			expect(engine.text_for('second.png').strip).to eq("#{Tesseract::API.new.version == '3.01' ? ?| : ?I}'m 12 and what is this.\nINSTALL GENTOO\nOH HAI 1234")
 		end
 
 		it 'raises when going out of the image boundaries' do
 			expect {
 				engine.text_for('second.png', 0, 0, 1000, 1000)
-			}.should raise_error
+			}.to raise_error IndexError
 		end
 	end
 
 	describe '#text_at' do
 		it 'can read the first test image' do
 			engine.image = 'first.png'
-			engine.text_at(2, 2, 2, 2).strip.should == ''
+
+			expect(engine.text_at(2, 2, 2, 2).strip).to eq('')
 		end
 
 		it 'can read the second test image' do
 			engine.image = 'second.png'
-			engine.text_at(242, 191, 129, 31).strip.should == 'OH HAI 1234'
+
+			expect(engine.text_at(242, 191, 129, 31).strip).to eq('OH HAI 1234')
 		end
 
 		it 'raises when going out of the image boundaries' do
 			expect {
 				engine.image = 'second.png'
 				engine.text_at(10, 20, 1000, 1000)
-			}.should raise_error
+			}.to raise_error IndexError
 		end
 	end
 
@@ -47,13 +49,14 @@ describe Tesseract::Engine do
 			engine.image = 'first.png'
 			engine.select 2, 2, 2, 2
 
-			engine.text.strip.should == ''
+			expect(engine.text.strip).to eq('')
 		end
 
 		it 'can read the second test image' do
 			engine.image = 'second.png'
 			engine.select 242, 191, 129, 31
-			engine.text.strip.should == 'OH HAI 1234'
+
+			expect(engine.text.strip).to eq('OH HAI 1234')
 		end
 
 		it 'raises when going out of the image boundaries' do
@@ -61,59 +64,83 @@ describe Tesseract::Engine do
 				engine.image = 'second.png'
 				engine.select 10, 20, 1000, 1000
 				engine.text
-			}.should raise_error
+			}.to raise_error IndexError
+		end
+	end
+
+	describe '#hocr' do
+		it 'can read the first test image' do
+			engine.image = 'first.png'
+			engine.select 2, 2, 2, 2
+
+			expect(engine.hocr).to eq("  <div class='ocr_page' id='page_1' title='image \"\"; bbox 2 2 2 2; ppageno 0'>\n  </div>\n")
+		end
+
+		it 'can read the second test image' do
+			engine.image = 'second.png'
+			engine.select 242, 191, 129, 31
+
+			expect(engine.hocr).to eq("  <div class='ocr_page' id='page_1' title='image \"\"; bbox 242 191 129 31; ppageno 0'>\n   <div class='ocr_carea' id='block_1_1' title=\"bbox 242 191 371 222\">\n    <p class='ocr_par' dir='ltr' id='par_1_1' title=\"bbox 250 200 365 213\">\n     <span class='ocr_line' id='line_1_1' title=\"bbox 250 200 365 213; baseline 0 0\"><span class='ocrx_word' id='word_1_1' title='bbox 250 200 275 213; x_wconf 94' lang='eng' dir='ltr'>OH</span> <span class='ocrx_word' id='word_1_2' title='bbox 285 200 313 213; x_wconf 90' lang='eng' dir='ltr'>HAI</span> <span class='ocrx_word' id='word_1_3' title='bbox 323 200 365 213; x_wconf 86' lang='eng'>1234</span> \n     </span>\n    </p>\n   </div>\n  </div>\n")
 		end
 
+		it 'raises when going out of the image boundaries' do
+			expect {
+				engine.image = 'second.png'
+				engine.select 10, 20, 1000, 1000
+
+				engine.hocr
+			}.to raise_error IndexError
+		end
 	end
 
 	describe '#blacklist' do
 		it 'works with removing weird signs' do
-			engine.with { |e| e.blacklist = '|' }.text_for('second.png').strip.should == "I'm 12 and what is this.\nINSTALL GENTOO\nOH HAI 1234"
+			expect(engine.with { |e| e.blacklist = '|' }.text_for('second.png').strip).to eq("I'm 12 and what is this.\nINSTALL GENTOO\nOH HAI 1234")
 		end
 	end
 
 	describe '#whitelist' do
 		it 'makes everything into a number' do
-			engine.with { |e| e.whitelist = '1234567890' }.text_for('second.png').strip.should match(/^[\d\s]*$/)
+			expect(engine.with { |e| e.whitelist = '1234567890' }.text_for('second.png').strip).to match(/^[\d\s]*$/)
 		end
 	end
 
 	describe '#page_segmentation_mode' do
 		it 'sets it correctly' do
-			engine.with {|e|
+			expect(engine.with {|e|
 				e.page_segmentation_mode = :single_line
 				e.whitelist              = [*'a'..'z', *'A'..'Z', *0..9, " ."].join
-			}.text_for('jsmj.png').strip.should == 'JSmj'
+			}.text_for('jsmj.png').strip).to eq('Jsmj')
 		end
 	end
 
 	describe '#blocks' do
 		it 'works properly with first image' do
-			engine.blocks_for('first.png').first.to_s.strip.should == 'ABC'
+			expect(engine.blocks_for('first.png').first.to_s.strip).to eq('ABC')
 		end
 	end
 
 	describe '#paragraphs' do
 		it 'works properly with first image' do
-			engine.paragraphs_for('first.png').first.to_s.strip.should == 'ABC'
+			expect(engine.paragraphs_for('first.png').first.to_s.strip).to eq('ABC')
 		end
 	end
 
 	describe '#lines' do
 		it 'works properly with first image' do
-			engine.lines_for('first.png').first.to_s.strip.should == 'ABC'
+			expect(engine.lines_for('first.png').first.to_s.strip).to eq('ABC')
 		end
 	end
 
 	describe '#words' do
 		it 'works properly with first image' do
-			engine.words_for('first.png').first.to_s.should == 'ABC'
+			expect(engine.words_for('first.png').first.to_s).to eq('ABC')
 		end
 	end
 
 	describe '#symbols' do
 		it 'works properly with first image' do
-			engine.symbols_for('first.png').first.to_s.should == 'A'
+			expect(engine.symbols_for('first.png').first.to_s).to eq('A')
 		end
 	end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tesseract-ocr
 version: !ruby/object:Gem::Version
-  version: 0.1.7
+  version: 0.1.8
 platform: ruby
 authors:
 - meh.
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-02-11 00:00:00.000000000 Z
+date: 2015-06-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: call-me
@@ -243,4 +243,3 @@ test_files:
 - test/tesseract_spec.rb
 - test/test-european.jpg
 - test/test.png
-has_rdoc: