pdfbeads 1.0.3 → 1.0.5

data/ChangeLog

@@ -18,3 +18,11 @@
 
     * TIFF files with an embedded EXIF block were incorrectly processed
       by ImageInspector.
+
+2012 February 4 (Alexey Kryukov) Version 1.0.5
+
+    * If JPEG2000 compression was requested but not available,
+      JPEG files were still written with the 'JP2' extension.
+
+    * Some tweaks to minimize the effect of page labels being inconsistently handled
+      in various PDF viewers (prefer ISO-8859-1 strings if possible).

data/README

@@ -51,3 +51,8 @@ Note that PDFBeads is intended for creating PDF files from previously
 processed images, and so it can't done some operations (e. g. converting
 color or grayscale scans to B&W) which should be typically performed with
 a special scan processing application, such as ScanTailor.
+
+PDFBeads requires RMagick (the Ruby bindings for the popular Magick++ image
+processing library). The hpricot extension is not required, but highly
+recommended, as without it PDFBeads would not be able to read data from hOCR
+files.

data/bin/pdfbeads

@@ -1,4 +1,4 @@
-#!/usr/bin/env ruby1.9.1
+#!/usr/bin/env ruby
 # encoding: UTF-8
 
 ######################################################################
@@ -9,7 +9,7 @@
 # Unlike other PDF creation tools, this utility attempts to implement
 # the approach typically used for DjVu books. Its key feature is
 # separating scanned text (typically black, but indexed images with
-# a small number of colors are also accepted) from halftone images 
+# a small number of colors are also accepted) from halftone images
 # placed into a background layer.
 #
 # Copyright (C) 2010 Alexey Kryukov (amkryukov@gmail.com).

data/lib/pdfbeads/pdfbuilder.rb

@@ -145,10 +145,25 @@ class PDFBeads::PDFBuilder
     if labels != nil and labels.length > 0
       nTree = "<</Nums[\n"
       labels.each do |rng|
-        ltitl = Iconv.iconv( "utf-16be", "utf-8", rng[:prefix] ).first.to_text
-
         nTree << "#{rng[:first]} << "
-        nTree << "/P (\xFE\xFF#{ltitl.to_text}) " if rng.has_key? :prefix
+        if rng.has_key? :prefix
+          begin
+            # If possible, use iso8859-1 (aka PDFDocEncoding) for page labels:
+            # it is at least guaranteed to be safe
+            ltitl = Iconv.iconv( "iso8859-1", "utf-8", rng[:prefix] ).first
+            nTree << "/P (#{ltitl.to_text}) " 
+          rescue Iconv::InvalidCharacter, Iconv::IllegalSequence
+            ltitl = Iconv.iconv( "utf-16be", "utf-8", rng[:prefix] ).first
+            # 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
+              nTree << "/P (\xFE\xFF#{ltitl.to_text}\x00\x00) "
+            # Otherwise put a formally correct Unicode string, which, however, may stumble acroread
+            else
+              nTree << "/P (\xFE\xFF#{ltitl.to_text}) "
+            end
+          end
+        end
         nTree << "/S /#{rng[:style]} " if rng.has_key? :style
         nTree << "/St #{rng[:start]}" if rng.has_key? :start
         nTree << ">>\n"
@@ -342,7 +357,7 @@ class PDFBeads::PDFBuilder
       fin.each do |fl|
         next if /^\#/.match( fl )
 
-        if /^\/?([A-Za-z]+)[ \t]*:[ \t]+\"(.*)\"/.match( fl )
+        if /^\/?([A-Za-z]+)[         ]*:[         ]+\"(.*)\"/.match( fl )
           key = $1
           if keys.include? key
             begin

data/lib/pdfbeads/pdffont.rb

@@ -396,9 +396,6 @@ class PDFBeads::PDFBuilder::FontDataProvider
       0xFB01 => ["/fi", 556],
       0xFB02 => ["/fl", 556],
     ]
-    @chardata.default = proc do |fd, uni|
-      fd[uni] = [ sprintf( "/uni%04X",uni ), 500 ]
-    end
 
     @encodings = Array.new()
     @wlists = Array.new()
@@ -410,7 +407,7 @@ class PDFBeads::PDFBuilder::FontDataProvider
     w = 0.0
     line.each_char do |uc|
       begin
-        w += @chardata[uc.ord][1] * size / 1000.0
+        w += chardata( uc.ord )[1] * size / 1000.0
       rescue
         rawbytes = uc.unpack( 'C*' )
         bs = ''
@@ -427,7 +424,7 @@ class PDFBeads::PDFBuilder::FontDataProvider
   def getEncoding( enc )
     ret = Array.new()
     enc.each do |char|
-      ret << @chardata[char.ord][0]
+      ret << chardata( char.ord )[0]
     end
     ret
   end
@@ -437,7 +434,7 @@ class PDFBeads::PDFBuilder::FontDataProvider
   def getWidths( enc )
     ret = Array.new()
     enc.each do |char|
-      ret << @chardata[char.ord][1]
+      ret << chardata( char.ord )[1]
     end
     ret
   end
@@ -530,4 +527,10 @@ class PDFBeads::PDFBuilder::FontDataProvider
     ], Zlib::Deflate.deflate( cmap,9 ) )
     toUnicode
   end
+
+  def chardata( uni )
+    @chardata.fetch( uni ) do |u|
+      [ sprintf( "/uni%04X",uni ), 500 ]
+    end
+  end
 end

data/lib/pdfbeads/pdflabels.rb

@@ -107,7 +107,7 @@ class PDFBeads::PDFBuilder::PDFLabels < Array
     ]
     res = ''
 
-    numerals.keys.sort{ |a,b| b <=> a }.reverse.each do |val|
+    numerals.keys.sort{ |a,b| b <=> a }.each do |val|
       while num >= val
         res << numerals[val]
         num -= val

data/lib/pdfbeads/pdfpage.rb

@@ -259,6 +259,7 @@ class PDFBeads::PageDataProvider < Array
           $stderr.puts( "This version of ImageMagick doesn't support JPEG2000 compression." )
           $stderr.puts( "\tI'll use JPEG compression instead." )
           bgf = 'JPG'
+          bgpath = "#{@basename}.bg." << bgf.downcase
         end
 
         writeImage( img,bgpath,bgf )

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: pdfbeads
 version: !ruby/object:Gem::Version 
-  hash: 17
-  prerelease: false
+  hash: 29
+  prerelease: 
   segments: 
   - 1
   - 0
-  - 3
-  version: 1.0.3
+  - 5
+  version: 1.0.5
 platform: ruby
 authors: 
 - Alexey Kryukov
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-11-16 00:00:00 +03:00
+date: 2012-02-04 00:00:00 +04:00
 default_executable: pdfbeads
 dependencies: []
 
@@ -74,7 +74,7 @@ requirements:
 - RMagick, v2.13.0 or greater
 - Hpricot, v0.8.3 or greater
 rubyforge_project: PDFBeads
-rubygems_version: 1.3.7
+rubygems_version: 1.5.0
 signing_key: 
 specification_version: 3
 summary: PDFBeads -- convert scanned images to a single PDF file.