pdfbeads 1.1.1 → 1.1.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 577a47277a3e474a47b740ce93ca89041402c4fb
-  data.tar.gz: 88a1f950e31e41e47f79ef978d9e589e1b9724fb
+SHA256:
+  metadata.gz: 198ab9ffc035604ce4cfb3528dbebfb7746d746de1a429089028102003e35480
+  data.tar.gz: e32b5a1de30aeb1bb715f0ea6417478f16f9ba9f229268a55ad3fd4ae9bc67ab
 SHA512:
-  metadata.gz: 16ab45bd0c4d63d7c6f567f3df5e2dea2ecc1e9237ac9b5acfc9cd92e31a72a8df7ad184e5c7fd779cb990f58ff64dbb3027464c7579fd79d58fc270ac01c841
-  data.tar.gz: b946c901407f06aeaf509988341759ba4f4cfc217067b00fb72ad5bc9ae60e4d1dd0729bb7a3761a6c06f71816dcb9398e2cde4335ee8a5620c1686e208fe901
+  metadata.gz: 485725e99d06c9216e238626b35ec73d7ea2c64513f5112c24448191b6a09ccef40ef58a24bac291a975c8881779a6e62450e629ebee4c89ec34be9d976b4ccb
+  data.tar.gz: f0f53d06628a9433684d77e175cf95446abeebaa54e14e59bb2a637666d06eaa559734a2bac5e43b3d69182d6afb8ef6627d356bcd6c5b3991200a5cd9e8e858

data/ChangeLog

@@ -77,4 +77,24 @@
 
     * hpricot is no longer developed, so switch to Nokagiri for hOCR processing.
 
+    * use String#encode instead of Iconv, when available
+
+    * Got tired from deps not being automatically resolved, so add them to
+      the gemspeck (the preferred method to install on Debian-based distributions
+      is now converting pdfbeads*.gem to a deb package with gem2deb).
+
     + English HTML documentation added.
+
+2016 Dec 20 (Alexey Kryukov) Version 1.1.2
+    * Add a missing String#encode call
+
+2020 Jan 05 (Alexey Kryukov) Version 1.1.2
+    * Respect the photometric interpretation of Group4-encoded tiff images
+    * Some JPEG images were erroneously treated as LZW-encoded
+
+2020 Jan 24 (Alexey Kryukov) Version 1.1.2
+    * Fixed the /BaseState field in the optional content dictionary, which caused files to be rejected by some viewers
+
+2021 Nov 24 (Alexey Kryukov) Version 1.1.3
+    * Fixed some errors/warnings produced by newer rmagick versions
+    * Bumped the required rmagick version up to 3.2.0

data/bin/pdfbeads

@@ -32,7 +32,6 @@
 #######################################################################
 
 require 'optparse'
-require 'iconv'
 require 'time'
 
 require 'pdfbeads'

data/doc/pdfbeads.en.html

@@ -498,31 +498,27 @@ files (hOCR is a HTML language extension, allowing to store information
 about exact positioning of characters and markup elements on the page) or
 imported from another PDF file.</p>
 
-<p>Для создания файлов в формате hOCR необходимо воспользоваться программой
-оптического распознавания символов, поддерживающей этот формат, например
-<a href="https://launchpad.net/cuneiform-linux/">Cuneiform</a> или
+<p>In order to create hOCR files one should use an OCR application which
+supports this output format, such as
+<a href="https://launchpad.net/cuneiform-linux/">Cuneiform</a> or
 <a href="http://code.google.com/p/tesseract-ocr/">Tesseract</a>.
-Распознанный текст следует сохранить в той же директории, что и остальные
-файлы, относящиеся к проекту. При этом каждой распознанной странице должен
-соответствовать отдельный файл с тем же базовым именем, что и у исходного
-изображения, при расширении HTM(L) или HOCR. Обработка файлов hOCR
-осуществляется автоматически при условии, что интерпретатору Ruby доступно
-расширение Nokogiri.</p>
-
-<p>Иное возможное решение заключается в том, чтобы импортировать текстовый
-слой из другого PDF-файла (естественно, последний должен быть получен путем
-распознавания тех же самых изображений, которые предполагается затем обработать
-с помощью pdfbeads). Имя полученного файла следует передать pdfbeads с помощью
-ключа <tt>-T</tt> (полная форма&nbsp;&mdash; <tt>-text-pdf</tt>). Эта
-возможность особенно важна в тех случаях, когда приходится использовать для
-распознавания текста коммерческое приложение (например,
-<a href="http://www.abbyy.ru/finereader/">ABBYY Finereader</a>), в котором
-не предусмотрена поддержка формата hOCR. <strong>Внимание:</strong> при
-создании промежуточного PDF-файла в ABBYY Finereader следует использовать
-настройки &laquo;текст под изображением&raquo; или &laquo;текст поверх
-изображения&raquo;, поскольку при иных настройках размещение символов на
-странице может оказаться не вполне соответствующим исходному графическому
-файлу.</p>
+The recognized text should be placed into the same directory as other project
+files by a such way that each hOCR file should have the same base name as its
+source image, while the extension should be either HTM(L) or HOCR. hOCR files
+are automatically processed by pdfbeads if only the Nokogiri extension is
+installed and accessible to the Ruby engine.</p>
+
+<p>Another possible solution is to import an OCR layer from another PDF file
+(naturally the later should be produced by recognizing the same set
+of images which is supposed to be processed with pdfbeads). The path to the
+file can be passed to pdfbeads via the <tt>-T</tt> (or <tt>-text-pdf</tt>)
+option. This feature is especially important in those cases where one has
+to use a commercial OCR application (such as
+<a href="http://www.abbyy.ru/finereader/">ABBYY Finereader</a>), which
+doesn't support the hOCR format. <strong>Warning:</strong> you may
+need to experiment with your OCR application options related with PDF output
+to get the best correspondance between the layout of the recognized
+text and the source image.</p>
 
 <h3>Processing files with the right-to-left text direction</h3>
 

data/lib/imageinspector.rb

@@ -49,7 +49,7 @@ module ImageInspector
 end
 
 # Parse image header and retrieve its basic properties. The code is inspired
-# by Sam Stephenson's snippet which demonstrates how to determine a JPEG 
+# by Sam Stephenson's snippet which demonstrates how to determine a JPEG
 # image size ( see http://snippets.dzone.com/posts/show/805) and Paul
 # Schreiber's code for TIFF (see
 # http://paulschreiber.com/blog/2010/06/10/tiff-file-dimensions-in-ruby/)
@@ -67,12 +67,12 @@ class ImageInspector::Image
   # Image format and compression method
   attr_reader :format, :compression
   # Return TIFF tags as a hash for TIFF images or JPEG images with EXIF
-  # data. Otherwise this property is nil. 
+  # data. Otherwise this property is nil.
   attr_reader :tags
 
   @@gc = (IO.method_defined? :getbyte) ? (:getbyte) : (:getc)
 
-  # Set all image attributes to nil and open an image if an optional 
+  # Set all image attributes to nil and open an image if an optional
   # argument is specified.
   def initialize( input=nil )
     clearData()
@@ -99,7 +99,7 @@ class ImageInspector::Image
     end
   end
 
-  # Return image data (possibly compressed) for a previously initialized 
+  # Return image data (possibly compressed) for a previously initialized
   # image as a sring. For JPEG and JPEG2000 this would be the whole image
   # as it is stored on the disk, while for TIFF and PNG all headers are
   # stripped and a raw data stream is returned.
@@ -169,7 +169,7 @@ class ImageInspector::Image
     sign = io.read( 2 )
     if sign.eql? "\xFF\xD8".to_binary
       @format = :JPEG
-      @compression = :DCTDecode 
+      @compression = :DCTDecode
       jpgExamine( io )
       return
     end
@@ -191,7 +191,7 @@ class ImageInspector::Image
     sign << io.read( 4 )
     if sign.eql? "\x00\x00\x00\x0CjP  \x0D\x0A\x87\x0A".to_binary
       @format = :JPEG2000
-      @compression = :JPXDecode 
+      @compression = :JPXDecode
       j2kParseBox( io )
       return
     end
@@ -288,8 +288,8 @@ class ImageInspector::Image
 
     # We should not expect to find required image properties (such as width
     # or height) in EXIF data of a JPEG image.
-    raise 'malformed TIFF: a required tag is missing' unless @format.eql? :JPEG or ( 
-      @tags.has_key? 0x0100 and @tags.has_key? 0x0101 and 
+    raise 'malformed TIFF: a required tag is missing' unless @format.eql? :JPEG or (
+      @tags.has_key? 0x0100 and @tags.has_key? 0x0101 and
       @tags.has_key? 0x0106 and @tags.has_key? 0x0111 and @tags.has_key? 0x0117 )
 
     unless @format.eql? :JPEG
@@ -325,16 +325,19 @@ class ImageInspector::Image
     end
     @tags.merge! tiffParseIFD( io,@tags[0x8769][0],intgr ) if @tags.has_key? 0x8769
 
-    if @tags.has_key? 0x0103
-      case @tags[0x0103][0]
-        when 1
-          @compression = :NoCompression
-        when 3, 4
-          @compression = :CCITTFaxDecode
-        when 5
-          @compression = :LZWDecode
-        when 8, 32946
-          @compression = :FlateDecode
+    # Strangely, got some JPEG's marked as LZW-decoded
+    unless @format.eql? :JPEG
+      if @tags.has_key? 0x0103
+        case @tags[0x0103][0]
+          when 1
+            @compression = :NoCompression
+          when 3, 4
+            @compression = :CCITTFaxDecode
+          when 5
+            @compression = :LZWDecode
+          when 8, 32946
+            @compression = :FlateDecode
+        end
       end
     end
 

data/lib/pdfbeads/pdfbuilder.rb

@@ -134,7 +134,7 @@ class PDFBeads::PDFBuilder
     })
     @doc.addObject(ocBack)
     cat.addToDict('OCProperties',
-      sprintf("<< /OCGs[%s %s] /D<< /Intent /View /BaseState (ON) /Order[%s %s] >>>>",
+      sprintf("<< /OCGs[%s %s] /D<< /Intent /View /BaseState /ON /Order[%s %s] >>>>",
         ref(ocFore.getID), ref(ocBack.getID), ref(ocFore.getID), ref(ocBack.getID)))
 
     page_objs = Array.new()
@@ -159,7 +159,11 @@ class PDFBeads::PDFBuilder
             nTree << "/P (#{ltitl.to_text}) "
           # Iconv::InvalidCharacter, Iconv::IllegalSequence, Encoding::UndefinedConversionError, Encoding::InvalidByteSequenceError
           rescue
-            ltitl = Iconv.iconv( "utf-16be", "utf-8", rng[:prefix] ).first
+            if rng[:prefix].respond_to? :encode
+              ltitl = rng[:prefix].encode( "utf-16be", "utf-8" )
+            else
+              ltitl = Iconv.iconv( "utf-16be", "utf-8", rng[:prefix] ).first
+            end
             # If there is no number (just prefix) then put a zero character after the prefix:
             # this makes acroread happy, but prevents displaying the number in evince
             unless rng.has_key? :style
@@ -399,11 +403,17 @@ class PDFBeads::PDFBuilder
           key = $1
           if keys.include? key
             begin
+              tmp_str = ''
               if $2.respond_to? :encode
-                ret[key] = $2.encode( "utf-16be", "utf-8" )
+                tmp_str = $2.encode( "utf-16be", "utf-8" )
               else
-                ret[key] = Iconv.iconv( "utf-16be", "utf-8", $2 ).first
+                tmp_str = Iconv.iconv( "utf-16be", "utf-8", $2 ).first
               end
+              # a parenthesis code in a formally correct utf-16 should nevertheless be escaped
+              ret[key] = tmp_str.to_binary
+              ret[key].gsub!( /\x5C/,"\x5C\x5C" )
+              ret[key].gsub!( /\x28/,"\x5C\x28" )
+              ret[key].gsub!( /\x29/,"\x5C\x29" )
             rescue
               $stderr.puts("Error: metadata should be specified in utf-8")
             end
@@ -651,7 +661,7 @@ class PDFBeads::PDFBuilder
   end
 
   def elementText( elem )
-    # used to put some Iconv stuff here, but nokogiri makes this conversion itself
+    # used to put some Iconv stuff here, but nokogiri performs this conversion itself
     return elem.inner_text.strip
   end
 
@@ -893,15 +903,18 @@ class PDFBeads::PDFBuilder
 
     unless stencil.compression.eql? :CCITTFaxDecode and rows_per_strip >= height
       img = ImageList.new( path )
-      imgdata = img.to_blob{
-        self.format = 'TIFF'
-        self.define( 'TIFF','rows-per-strip',height )
-        self.compression = Group4Compression
+      imgdata = img.to_blob { |imd|
+        imd.format = 'TIFF'
+        imd.define( 'TIFF','rows-per-strip',height )
+        imd.compression = Group4Compression
       }
       stencil = ImageInspector.new( StringIO.new(imgdata) )
       img.destroy!
     end
     body = stencil.getRawData
+    photometric = 0
+    photometric = stencil.tags[0x106][0] if
+      stencil.format.eql? :TIFF and stencil.tags.has_key? 0x106
 
     xobj = XObj.new(Hash[
       'Type'             => '/XObject',
@@ -913,8 +926,14 @@ class PDFBeads::PDFBuilder
       'ColorSpace'       => '/DeviceGray',
       'BitsPerComponent' => '1',
       'Filter'           => '/CCITTFaxDecode',
-      'DecodeParms'      => "<< /Columns #{width} /K -1 >>"
+      'DecodeParms'      => "<< /Columns #{width} /K -1 >>",
     ], body)
+    if photometric == 1 then
+      # As ImageMask is always on, BlackIs1 actually doesn't work, while
+      # the Decode array does.
+      xobj.addToDict( 'BlackIs1', 'true' )
+      xobj.addToDict( 'Decode', '[1 0]' )
+    end
 
     return [ xobj,width,height,xres,yres ]
   end
@@ -966,10 +985,10 @@ class PDFBeads::PDFBuilder
       ( [ :FlateDecode,:LZWDecode,:CCITTFaxDecode ].include? insp.compression and insp.tags[0x0116][0] >= insp.height )))
 
       img = ImageList.new( impath )
-      imgdata = img.to_blob{
-        self.format = 'PNG'
-        self.quality = 95
-        self.compression = ZipCompression
+      imgdata = img.to_blob { |imd|
+        imd.format = 'PNG'
+        imd.quality = 95
+        imd.compression = ZipCompression
       }
       insp = ImageInspector.new( StringIO.new(imgdata) )
       img.destroy!

data/lib/pdfbeads/pdfpage.rb

@@ -160,19 +160,19 @@ class PDFBeads::PageDataProvider < Array
 
     def writeImage( img,path,fmt )
       begin
-        img.write( path ) do
+        img.write( path ) do |curimg|
           case fmt
           when 'JP2'
-            self.define( 'JP2','mode','real' )
-            self.define( 'JP2','numrlvls',4 )
-            self.define( 'JP2','rate',0.015625 )
+            curimg.define( 'JP2','mode','real' )
+            curimg.define( 'JP2','numrlvls',4 )
+            curimg.define( 'JP2','rate',0.015625 )
           when 'JPG'
-            self.quality = 50
+            curimg.quality = 50
           else
-            self.compression = ZipCompression
-            self.quality = 95
+            curimg.compression = ZipCompression
+            curimg.quality = 95
           end
-          self.format = fmt
+          curimg.format = fmt
         end
         return true
       rescue
@@ -205,10 +205,10 @@ class PDFBeads::PageDataProvider < Array
                 bitonal.colormap( j,crepl )
               end
               bitonal.compress_colormap!
-              bitonal.write( cpath ) do
-                self.format = 'TIFF'
-                self.define( 'TIFF','rows-per-strip',img.rows )
-                self.compression = Group4Compression
+              bitonal.write( cpath ) do |curimg|
+                curimg.format = 'TIFF'
+                curimg.define( 'TIFF','rows-per-strip',img.rows )
+                curimg.compression = Group4Compression
               end
               bitonal.destroy!
               created = true
@@ -230,10 +230,10 @@ class PDFBeads::PageDataProvider < Array
       binpath = "#{@basename}.black.tiff"
       if not File.exists? binpath or force
         im_copy = img.copy; bitonal = im_copy.threshold(QuantumRange/255*treshold); im_copy.destroy!
-        bitonal.write( binpath ){
-          self.format = 'TIFF'
-          self.define( 'TIFF','rows-per-strip',img.rows )
-          self.compression = Group4Compression
+        bitonal.write( binpath ) { |curimg|
+          curimg.format = 'TIFF'
+          curimg.define( 'TIFF','rows-per-strip',img.rows )
+          curimg.compression = Group4Compression
         }
         bitonal.destroy!
         map[:created] = true
@@ -324,7 +324,7 @@ class PDFBeads::PageDataProvider < Array
       PageData.fixResolution( img )
       mask.resize!( imw,imh ) if mask.columns != imw or mask.rows != imh
 
-      no_fg = img.composite( mask,CenterGravity,CopyOpacityCompositeOp )
+      no_fg = img.composite( mask,CenterGravity,CopyAlphaCompositeOp )
       bg = no_fg.blur_channel( 0,6,AllChannels )
       bg.alpha( DeactivateAlphaChannel )
 
@@ -346,7 +346,7 @@ class PDFBeads::PageDataProvider < Array
         ksam = mask.negate
         mask.destroy!
 
-        no_bg = img.composite( ksam,CenterGravity,CopyOpacityCompositeOp )
+        no_bg = img.composite( ksam,CenterGravity,CopyAlphaCompositeOp )
         fg = no_bg.clone
 
         # Resize the image to a tiny size and then back to the original size
@@ -403,8 +403,8 @@ class PDFBeads::PageDataProvider < Array
 
     # A hack for some Windows versions of RMagick, which throw an error the
     # first time when Magick.formats is accessed
+    retries = 2
     begin
-      retries = 2
       mfmts = Magick.formats
     rescue
       retry if (retries -= 1 ) > 0

data/lib/pdfbeads.rb

@@ -32,7 +32,7 @@
 
 require 'zlib'
 
-require 'RMagick'
+require 'rmagick'
 include Magick
 
 begin

metadata

@@ -1,55 +1,55 @@
 --- !ruby/object:Gem::Specification
 name: pdfbeads
 version: !ruby/object:Gem::Version
-  version: 1.1.1
+  version: 1.1.3
 platform: ruby
 authors:
 - Alexey Kryukov
-autorequire: 
+autorequire:
 bindir: bin
 cert_chain: []
-date: 2014-01-30 00:00:00.000000000 Z
+date: 2021-11-24 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rmagick
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 2.13.0
+        version: 3.2.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
-        version: 2.13.0
+        version: 3.2.0
 - !ruby/object:Gem::Dependency
   name: nokogiri
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: 1.5.10
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: 1.5.10
 - !ruby/object:Gem::Dependency
   name: pdf-reader
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: 1.0.0
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - '>='
+    - - ">="
       - !ruby/object:Gem::Version
         version: 1.0.0
 description: |2
@@ -70,45 +70,44 @@ extra_rdoc_files:
 - COPYING
 - ChangeLog
 files:
+- COPYING
+- ChangeLog
+- README
+- bin/pdfbeads
+- doc/pdfbeads.en.html
+- doc/pdfbeads.ru.html
+- lib/imageinspector.rb
+- lib/pdfbeads.rb
 - lib/pdfbeads/pdfbuilder.rb
-- lib/pdfbeads/pdfpage.rb
-- lib/pdfbeads/pdftoc.rb
+- lib/pdfbeads/pdfdoc.rb
 - lib/pdfbeads/pdffont.rb
 - lib/pdfbeads/pdflabels.rb
-- lib/pdfbeads/pdfdoc.rb
-- lib/pdfbeads.rb
-- lib/imageinspector.rb
-- bin/pdfbeads
-- doc/pdfbeads.ru.html
-- doc/pdfbeads.en.html
-- README
-- COPYING
-- ChangeLog
+- lib/pdfbeads/pdfpage.rb
+- lib/pdfbeads/pdftoc.rb
 homepage: http://pdfbeads.rubyforge.org
 licenses:
-- GPL
+- GPL-3.0+
 metadata: {}
-post_install_message: 
+post_install_message:
 rdoc_options: []
 require_paths:
 - lib
 required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
-  - - '>='
+  - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements:
-- RMagick, v2.13.0 or greater
+- RMagick, v3.2.0 or greater
 - nokogiri, v1.5.10 or greater
 - PDF::Reader, v1.0.0 or greater
-rubyforge_project: PDFBeads
-rubygems_version: 2.0.3
-signing_key: 
+rubygems_version: 3.2.29
+signing_key:
 specification_version: 4
 summary: PDFBeads -- convert scanned images to a single PDF file.
 test_files: []