pdf_margins 0.1.1 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 6065ffaa732ecbbbe1d9addb6275715ea9613596
-  data.tar.gz: 8a3e756c74ba210add152540fee7948255bd40ff
+  metadata.gz: d41e2247fb4dead96b77524b5c4524bfc3802692
+  data.tar.gz: 73d2bd7d01f22f3232716376e64ee52a63c81a20
 SHA512:
-  metadata.gz: 5f89e8ca71ee8d320e7afc9cae7181a839429c23e306b6ce916a4294a6991bc9a4d99c407405d6f102648ab4bf489a0b5d73e97fff1fa68f4cd8eb8a4d569195
-  data.tar.gz: bc12bde2c429fe3eb0ed5b79daa775b837717a773b9ad391209f11c7747b544c650c9c984aecbe6c3c793a6f1d1b6fad57ecf504ffad4b9d5ddc58c3f29747bf
+  metadata.gz: 2231c4e3028feddc1b17ce1413d938518b27b11e14e84913683e90a1187c21b2ec47c91c1927116f063d11096d04822301b0c86fda17d3559b83cc3d2430e3b9
+  data.tar.gz: 1fe63baac8954e2b26774d7aa644ffaab0754c936a64fffca2c1f0e7edd3158995cf98a0d77a118c61695eee1d1b770ff799d752332dcf99470ed4e7564c9627

data/README.md

@@ -5,18 +5,31 @@ pdf/margins
 Status](https://travis-ci.org/newspaperclub/pdf_margins.png)](https://travis-ci.org/newspaperclub/pdf_margins)
 
 Simple Ruby library to check a PDF for clear margin areas, by rendering to
-a CMYK bitmap, ensuring that the pixel area defined by each margin is empty.
-It's pretty slow, but this is a lot easier than attempting to work out which
-object in a PDF is rendering into the margin area.
+a bitmap, ensuring that the pixel area defined by each margin is empty. 
 
-Depends on RMagick, and thus requires a Ruby interpreter that supports
-C extensions, such as MRI.
+This is a bit of a clumsy technique, but it's faster than parsing and
+understanding all the permutations of the PDF format that may result in an
+object being visible in the margins.
 
-Written by [Tom Taylor](http://scraplab.net), [Newspaper Club](http://www.newspaperclub.com).
+It shells out to `mudraw`, part of [mupdf] [1], for rastering the PDF to PNGs,
+and then uses [Chunky PNG] [2] to examine each page's margins.
+
+On Ubuntu, you may wish to install `mupdf` from the [PPA] [3].
+
+Written by [Tom Taylor] [3], [Newspaper Club] [4].
+
+[1]: http://www.mupdf.com
+[2]: https://github.com/wvanbergen/chunky_png
+[3]: https://launchpad.net/~mupdf/+archive/stable
+[4]: http://scraplab.net
+[5]: http://www.newspaperclub.com
 
 Example
 --
 
+    # accepts a `spreads` boolean which assumes margin dimensions refer to
+    # spreads, and so alternates which pages we check left and right margins on
+    spreads = true
     # measurements in mm, ordered top, right, bottom, left (same as CSS)
-    checker = PDF::Margins::Checker.new('example.pdf', 10, 10, 10, 10)
+    checker = PDF::Margins::Checker.new('example.pdf', 10, 10, 10, 10, spreads)
     puts checker.issues.inspect

data/lib/pdf/margins/checker.rb

@@ -1,5 +1,7 @@
 require 'pdf/margins/issue'
-require 'RMagick'
+require 'chunky_png'
+require 'oily_png'
+require 'tmpdir'
 
 module PDF
   module Margins
@@ -12,12 +14,12 @@ module PDF
       SCALE_MULTIPLIER = 1 
       RESOLUTION = DEFAULT_RESOLUTION * SCALE_MULTIPLIER
 
-      attr_reader :file, :top_margin, :right_margin, :bottom_margin, :left_margin, :spreads
+      attr_reader :file_path, :top_margin, :right_margin, :bottom_margin, :left_margin, :spreads
 
       # Dimensions are in mm, to be converted to PDF points later. Pass spreads
       # as true to check left and right margins of spreads, not pages.
-      def initialize(file, top_margin, right_margin, bottom_margin, left_margin, spreads=false)
-        @file          = file
+      def initialize(file_path, top_margin, right_margin, bottom_margin, left_margin, spreads=false)
+        @file_path     = file_path
         @top_margin    = top_margin
         @right_margin  = right_margin
         @bottom_margin = bottom_margin
@@ -26,36 +28,41 @@ module PDF
       end
 
       def issues
-        image_list = Magick::Image.read(file) do
-          # Force CMYK colorspace, because ImageMagick autodetects PDF colorspace
-          # depending on the contents of the file. By forcing the colorspace we
-          # ensure that the subsequent checks in the `dirty_pixels?` method work.
-          self.colorspace = Magick::CMYKColorspace
-          self.density    = RESOLUTION
-          self.antialias  = false
-        end
+        temp_dir_path = Dir.mktmpdir("pdf_margins")
+
+        begin
+          # This produces greyscale PNGs - we throw the colour away because we
+          # don't need it to check for margins.
+          system("mudraw -g -b 0 -r #{RESOLUTION} -o #{temp_dir_path}/%d.png #{file_path}") || raise
+
+          issues = []
 
-        image_list.each_with_index.map do |image, index|
+          Dir.glob("#{temp_dir_path}/*.png").each_with_index do |png_path, index|
+            image = ChunkyPNG::Image.from_file(png_path)
+            page_number = index + 1
 
-          page_number = index + 1
-          [].tap do |page_issues|
-            if dirty_pixels?(top_pixels(image, top_margin))
-              page_issues << Issue.new(page_number, :top)
+            if dirty_top_margin?(image, top_margin)
+              issues << Issue.new(page_number, :top)
             end
 
-            if dirty_pixels?(bottom_pixels(image, bottom_margin))
-              page_issues << Issue.new(page_number, :bottom)
+            if dirty_bottom_margin?(image, bottom_margin)
+              issues << Issue.new(page_number, :bottom)
             end
 
-            if (!spreads || page_number % 2 == 0) && dirty_pixels?(left_pixels(image, left_margin))
-              page_issues << Issue.new(page_number, :left)
+            if (!spreads || page_number % 2 == 0) && dirty_left_margin?(image, left_margin)
+              issues << Issue.new(page_number, :left)
             end
             
-            if (!spreads || page_number % 2 != 0) && dirty_pixels?(right_pixels(image, right_margin))
-              page_issues << Issue.new(page_number, :right)
+            if (!spreads || page_number % 2 != 0) && dirty_right_margin?(image, right_margin)
+              issues << Issue.new(page_number, :right)
             end
           end
-        end.flatten
+
+        ensure
+          FileUtils.remove_entry(temp_dir_path)
+        end
+
+        return issues
       end
 
       private
@@ -64,31 +71,38 @@ module PDF
         (mm * MM_TO_PTS * SCALE_MULTIPLIER).floor
       end
 
-      def top_pixels(image, mm)
-        width_px = mm_to_pixels(mm)
-        image.get_pixels(0, 0, image.columns, width_px)
+      def dirty_top_margin?(image, mm)
+        px = mm_to_pixels(mm)
+        dirty_pixels?(image, 0, px-1, 0, image.width-1)
       end
 
-      def left_pixels(image, mm)
-        width_px = mm_to_pixels(mm)
-        image.get_pixels(0, 0, width_px, image.rows)
+      def dirty_left_margin?(image, mm)
+        px = mm_to_pixels(mm)
+        dirty_pixels?(image, 0, image.height-1, 0, px-1)
       end
 
-      def right_pixels(image, mm)
-        width_px = mm_to_pixels(mm)
-        x = image.columns - width_px
-        image.get_pixels(x, 0, width_px, image.rows)
+      def dirty_right_margin?(image, mm)
+        px = mm_to_pixels(mm)
+        offset = image.width - px - 1
+        dirty_pixels?(image, 0, image.height-1, offset, image.width-1)
       end
 
-      def bottom_pixels(image, mm)
-        width_px = mm_to_pixels(mm)
-        y = image.rows - width_px
-        image.get_pixels(0, y, image.columns, width_px)
+      def dirty_bottom_margin?(image, mm)
+        px = mm_to_pixels(mm)
+        offset = image.height - px - 1
+        dirty_pixels?(image, offset, image.height-1, 0, image.width-1)
       end
 
-      def dirty_pixels?(pixels)
-        pixels.any? do |p|
-          (p.cyan | p.magenta | p.yellow | p.black) > 0
+      def dirty_pixels?(image, row_start, row_end, column_start, column_end)
+        rows = (row_start..row_end).to_a
+        columns = (column_start..column_end).to_a
+
+        white = ChunkyPNG::Color::WHITE
+
+        rows.any? do |row|
+          columns.any? do |column|
+            image[column, row] != white
+          end
         end
       end
 

data/lib/pdf/margins/version.rb

@@ -1,5 +1,5 @@
 module PDF
   module Margins
-    VERSION = "0.1.1"
+    VERSION = "0.2.0"
   end
 end

metadata

@@ -1,29 +1,43 @@
 --- !ruby/object:Gem::Specification
 name: pdf_margins
 version: !ruby/object:Gem::Version
-  version: 0.1.1
+  version: 0.2.0
 platform: ruby
 authors:
 - Tom Taylor
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2013-07-11 00:00:00.000000000 Z
+date: 2013-07-15 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
-  name: rmagick
+  name: chunky_png
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: 2.13.2
+        version: 1.2.8
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - ~>
       - !ruby/object:Gem::Version
-        version: 2.13.2
+        version: 1.2.8
+- !ruby/object:Gem::Dependency
+  name: oily_png
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.1.0
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.1.0
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement
@@ -94,7 +108,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.0.2
+rubygems_version: 2.0.3
 signing_key: 
 specification_version: 4
 summary: Simple library to checks whether the margins are clear in a PDF