simple_tesseract 0.0.1

data/ext/extconf.rb

@@ -0,0 +1,6 @@
+require 'mkmf'
+
+have_library('tiff')
+have_library('tesseract_full')
+
+create_makefile('simple_tesseract_ext')

data/ext/simple_tesseract_ext.cpp

@@ -0,0 +1,41 @@
+#include <stdio.h>
+#include <ruby/ruby.h>
+#include <tiffio.h>
+#include <tesseract/img.h>
+#include <tesseract/imgs.h>
+#include <tesseract/baseapi.h>
+
+void read_tiff_image(TIFF* tif, IMAGE* image);
+
+extern "C" {
+  VALUE rb_cTesseract;
+
+  VALUE
+  rb_cTesseract_get_text(VALUE self, VALUE language, VALUE file,
+      VALUE rx, VALUE ry, VALUE rwidth, VALUE rheight) {
+    const char *lang = (const char *)(language == Qnil ? NULL : RSTRING_PTR(language));
+    TIFF * tif = TIFFOpen(RSTRING_PTR(file), "r");
+    int x = NUM2INT(rx),
+        y = NUM2INT(ry),
+        width = NUM2INT(rwidth),
+        height = NUM2INT(rheight);
+    IMAGE img;
+    read_tiff_image(tif, &img);
+    int bytes_per_line = check_legal_image_size(img.get_xsize(),
+        img.get_ysize(), img.get_bpp());
+
+    TessBaseAPI::InitWithLanguage(NULL, NULL, lang, NULL, false, 0, NULL);
+    char *text = TessBaseAPI::TesseractRect(img.get_buffer(), img.get_bpp()/8,
+        bytes_per_line, x, y, width, height);
+    TessBaseAPI::End();
+    TIFFClose(tif);
+
+    return rb_str_new2(text);
+  }
+
+  void
+  Init_simple_tesseract_ext() {
+    rb_cTesseract = rb_define_class("Tesseract", rb_cObject);
+    rb_define_private_method(rb_cTesseract, "get_text", (VALUE (*)(...))rb_cTesseract_get_text, 6);
+  }
+}

data/lib/tesseract.rb

@@ -0,0 +1,72 @@
+require 'RMagick'
+require 'simple_tesseract_ext'
+require 'stringio'
+
+class Tesseract
+  attr_reader :src, :blob
+  attr_accessor :lang, :editor
+
+  def initialize (opts={})
+    @lang = opts.delete(:lang) || opts.delete(:language) || 'eng'
+    self.src = opts.delete(:src) || opts.delete(:source) || opts.delete(:image)
+    self.blob = opts.delete(:blob)
+    @editor = opts.delete(:editor) || lambda {|x|x}
+    self.strip = opts.delete(:strip)
+    @tmp = Tempfile.new(['rbtesseract', '.tiff']).tap {|x| x.close }.path
+
+    ObjectSpace.define_finalizer(self, method(:finalize))
+  end
+
+  def src= (file)
+    @blob = nil if file
+    @src = file
+  end
+
+  def blob= (string)
+    @src = nil if string
+    @blob = string
+  end
+
+  def strip= (bool)
+    @strip = !!bool
+  end
+
+  def strip?
+    @strip
+  end
+
+  alias language lang
+  alias language= lang=
+  alias source src
+  alias source= src=
+  alias image src
+  alias image= src=
+
+  def solve (x=0, y=0, width=nil, height=nil)
+    editor.call((@src ? Magick::Image.read(@src) : Magick::Image.from_blob(@blob)).first).write(@tmp)
+    img = Magick::Image.read(@tmp).first
+    x ||= 0
+    y ||= 0
+    width ||= img.columns
+    height ||= img.rows
+
+    get_text(@lang, @tmp, x, y, width, height).tap {|x|
+      x.strip! if strip?
+    }
+  end
+
+  def crops (*areas)
+    areas.map {|area|
+      solve(*area)
+    }.join
+  end
+
+  def to_s
+    solve
+  end
+
+  def finalize
+    File.unlink(@tmp) rescue nil
+  end
+  alias close finalize
+end

metadata

@@ -0,0 +1,78 @@
+--- !ruby/object:Gem::Specification 
+name: simple_tesseract
+version: !ruby/object:Gem::Version 
+  prerelease: false
+  segments: 
+  - 0
+  - 0
+  - 1
+  version: 0.0.1
+platform: ruby
+authors: 
+- shura
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2011-09-06 00:00:00 +02:00
+default_executable: 
+dependencies: 
+- !ruby/object:Gem::Dependency 
+  name: rmagick
+  prerelease: false
+  requirement: &id001 !ruby/object:Gem::Requirement 
+    none: false
+    requirements: 
+    - - ">="
+      - !ruby/object:Gem::Version 
+        segments: 
+        - 0
+        version: "0"
+  type: :runtime
+  version_requirements: *id001
+description: tesseract ruby bindings
+email: shura1991@gmail.com
+executables: []
+
+extensions: 
+- ext/extconf.rb
+extra_rdoc_files: []
+
+files: 
+- lib/tesseract.rb
+- ext/simple_tesseract_ext.cpp
+- ext/extconf.rb
+has_rdoc: true
+homepage: http://github.com/shurizzle/simple_tesseract
+licenses: []
+
+post_install_message: 
+rdoc_options: []
+
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      segments: 
+      - 0
+      version: "0"
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.7
+signing_key: 
+specification_version: 3
+summary: tesseract ruby bindings
+test_files: []
+