pdfium 0.0.1

data/pdfium.gemspec

@@ -0,0 +1,29 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'pdfium/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "pdfium"
+  spec.version       = PDFium::VERSION
+  spec.authors       = ["Nathan Stitt", "Ted Han"]
+  spec.email         = ["nathan@stitt.org"]
+  spec.summary       = %q{Ruby bindings for Google's PDFium project}
+  spec.description   = %q{Ruby bindings for Google's PDFium project.  It supports extracting text and images from PDF's as well as rendering pages to bitmaps}
+  spec.homepage      = "https://github.com/nathanstitt/pdfium-ruby"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files -z`.split("\x0")
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ["lib"]
+
+  spec.add_development_dependency "bundler", "~> 1.7"
+  spec.add_development_dependency "rake", "~> 10.0"
+
+  spec.add_development_dependency "guard-minitest"
+  spec.add_development_dependency "guard-rake"
+  spec.add_development_dependency "fastimage"
+  spec.add_development_dependency "image_science"
+  spec.add_development_dependency "rake-compiler"
+end

data/test/benchmark-docsplit.rb

@@ -0,0 +1,41 @@
+#!/usr/bin/env ruby
+
+require 'rusage'
+require 'docsplit'
+
+
+IMAGE_SIZES = {
+  large:      '1000x',
+  normal:     '700x',
+  small:      '180x',
+  thumbnail:  '60x75!'
+}
+
+output_directory = Dir.mktmpdir
+
+start_time = Time.now
+Dir.glob("test-pdfs/*.pdf").each do |pdf|
+  Docsplit.extract_images(pdf, :format => :gif, :size => IMAGE_SIZES.values, :rolling => true, :output => output_directory)
+end
+elapsed_time = Time.now-start_time
+
+# Process.crusage measures children
+# Process.rusage measures self
+usage = Process.crusage
+
+rss = usage.maxrss.to_f
+# OSX reports in terms of bytes, bsd & linux use kb. "man getrusage"
+rss = rss/1024 if RUBY_PLATFORM =~ /darwin/
+
+du = Dir.glob("#{output_directory}/*/**").inject(0.0){|x,img| x+File.stat(img).size }
+FileUtils.rm_r output_directory
+
+def report(label,value,specifier="")
+  printf("%12s: %8.3f %s\n", label, value, specifier)
+end
+
+report "Elapsed",    elapsed_time, "Seconds"
+report "System CPU", usage.stime
+report "User CPU",   usage.utime
+report "Max Memory", (rss/1024),"MB"
+report "Disk Space", (du/1024/1024), "MB"

data/test/bookmarks_list_spec.rb

@@ -0,0 +1,26 @@
+require_relative 'spec_helper'
+
+describe PDFium::BookmarkList do
+
+    let(:api){ load_document("with_bookmarks") }
+    let(:utf){ load_document("example_utf8")   }
+
+    it "can be created" do
+        assert_kind_of PDFium::BookmarkList, api.bookmarks
+    end
+
+    it "can be empty" do
+        bm = PDFium::Bookmark.new(document: api)
+        assert_kind_of PDFium::BookmarkList, bm.children
+        assert bm.children.empty?, "First bookmark shouldn't have any children"
+    end
+
+    it "can iterate" do
+        count = 0
+        utf.bookmarks.each do | bm |
+            count +=1
+        end
+        assert_equal 16, count
+    end
+
+end

data/test/bookmarks_spec.rb

@@ -0,0 +1,34 @@
+require_relative 'spec_helper'
+
+describe PDFium::Bookmark do
+    let(:api){ load_document("with_bookmarks") }
+
+    it "can be read" do
+        bm = PDFium::Bookmark.new(document: api)
+        assert_equal Encoding::UTF_16LE, bm.title.encoding
+        assert_equal "INDEX", bm.title.encode!("ASCII-8BIT")
+    end
+
+    it "can create siblings" do
+        bm = PDFium::Bookmark.new(document: api)
+        second = bm.next_sibling
+        assert_kind_of PDFium::Bookmark, second
+        assert_equal "Chapter 1", second.title.encode!("ASCII-8BIT")
+        refute second.next_sibling, "PDF shouldn't have 3 top level bookmarks"
+    end
+
+    it "can create children" do
+        bm = PDFium::Bookmark.new(document: api)
+        children = bm.next_sibling.children
+        assert_kind_of PDFium::BookmarkList, children
+        assert_equal 1, children.count
+    end
+
+    it "has destinations" do
+        bm = PDFium::Bookmark.new(document: api).next_sibling
+        assert_kind_of Hash, bm.destination
+        assert_equal :destination, bm.destination[:type]
+        assert_equal 1, bm.destination[:page_number]
+    end
+
+end

data/test/debug.rb

@@ -0,0 +1,24 @@
+#!/usr/bin/env ruby
+
+require_relative '../lib/pdfium'
+require 'pathname'
+
+
+path = Pathname.new(__FILE__).dirname.join('pdfs','example_images.pdf')
+
+# guide = PDFium::Document.new( path )
+# `rm /tmp/images/*`
+# page = guide.page_at(0)
+
+
+# path.write page.as_image(height: 120).data('jpg')
+
+data = path.read
+puts data.length
+pdf = PDFium::Document.from_memory(data)
+
+puts pdf.page_count
+
+# page.each_image do |img|
+#     img.save("/tmp/images/#{img.index}.png")
+# end

data/test/document_spec.rb

@@ -0,0 +1,49 @@
+require_relative 'spec_helper'
+
+describe PDFium::Document do
+    let(:guide){ load_document("example_images") }
+
+    it "can create a new empty pdf" do
+        pdf = PDFium::Document.new
+        assert pdf
+    end
+
+    it "can be initialized from string" do
+        data = pdf_path("with_bookmarks").read
+        pdf = PDFium::Document.from_memory(data)
+        assert pdf
+        assert_equal 3, pdf.page_count
+    end
+
+    it "counts pdf pages" do
+        assert_equal 3, guide.page_count
+    end
+
+    it "can save to a file" do
+        pdf = PDFium::Document.new
+        PDFium::Page.create(pdf,0)
+        Tempfile.open(['test','.pdf']) do |f|
+            pdf.save(f.path)
+            reloaded = PDFium::Document.new(f.path)
+            assert_equal 1, reloaded.page_count
+        end
+    end
+
+    it "returns pages" do
+        assert guide
+        assert_kind_of PDFium::PageList, guide.pages
+    end
+
+    it "can read metadata" do
+        assert_equal "mPDF 5.1", guide.metadata[:producer].encode!("ASCII-8BIT")
+    end
+
+    it "can write metadata" do
+        guide.metadata do | md |
+            md[:author] = "My Little Writer"
+        end
+        after_saving(guide) do | saved |
+            assert_equal "My Little Writer", saved.metadata[:author].encode!("ASCII-8BIT")
+        end
+    end
+end

data/test/image_list_spec.rb

@@ -0,0 +1,18 @@
+require_relative 'spec_helper'
+
+describe PDFium::ImageList do
+
+    let(:image_doc){  load_document("example_images").page_at(0) }
+    let(:blank_page){ load_document("example_utf8").page_at(0)   }
+
+    it "can be empty" do
+        assert blank_page.images.none?, "images found where there should not be"
+    end
+
+    it "can iterate" do
+        count = 0
+        image_doc.images.each{|i| count+=1 }
+        assert_equal 26, count
+    end
+
+end

data/test/image_spec.rb

@@ -0,0 +1,53 @@
+require_relative 'spec_helper'
+require 'image_science'
+
+describe PDFium::Image do
+    let(:guide){ load_document("example_images") }
+    let(:page) { guide.page_at(0) }
+
+    it "saves as various formats" do
+        %w{png jpeg tiff bmp gif}.each do | ext |
+            file = Tempfile.new(['test', ".#{ext}"])
+            width  = rand(200) + 100
+            height = rand(300) + 100
+            page.as_image(width: width, height: height).save(file.path)
+            assert_size "#{width}x#{height}", file.path
+        end
+    end
+
+
+    it "dumps to string" do
+        file = Tempfile.new(['test',".jpeg"])
+        file.write page.as_image(height: 120).data('jpg')
+        file.flush
+        assert_size "84x120", file.path
+    end
+
+    it "iterates over page images" do
+        valid_sizes = [567, 284, 386, 227, 500, 939, 950, 959]
+        count = 0
+        page.each_image do |img|
+            count += 1
+            next if count % 4 == 0 # to speed up spec runs only sample 1/4 of time
+            assert_kind_of PDFium::Image, img
+            file = Tempfile.new(['test',".png"])
+            img.save(file.path)
+            assert_includes valid_sizes, FastImage.size(file).first
+            assert_includes valid_sizes, FastImage.size(file).last
+        end
+        assert_equal 26, count, "Incorrect # of images counted"
+    end
+
+    it "can return an ImageScience instance" do
+        image = page.images.first
+        assert image
+        ims = image.as_science
+        file = Tempfile.new(['test',"jpg"])
+        ims.cropped_thumbnail(100) do |thumb|
+           thumb.save file.path
+           assert_size "100x100", file.path
+        end
+
+    end
+
+end

data/test/page_list_spec.rb

@@ -0,0 +1,24 @@
+require_relative 'spec_helper'
+
+describe PDFium::PageList do
+
+    let(:guide){ load_document("with_bookmarks") }
+
+    it "can be empty" do
+        pdf = PDFium::Document.new
+        pages = pdf.pages
+        assert_kind_of PDFium::PageList, pages
+        assert pages.none?, "A freshly created Document shouldn't have any pages"
+    end
+
+    it "can iterate" do
+        count = 0
+        guide.pages.each{ count += 1 }
+        assert_equal 3, count
+    end
+
+    it "supports access by index" do
+        assert_equal 1, guide.pages[0].number
+        assert_equal 2, guide.pages[1].number
+    end
+end

data/test/page_spec.rb

@@ -0,0 +1,91 @@
+# -*- coding: utf-8 -*-
+require_relative 'spec_helper'
+
+describe PDFium::Page do
+    let(:utfdoc){ PDFium::Document.new( pdf_path("example_utf8") ) }
+    let(:textdoc){ PDFium::Document.new( pdf_path("example_divs") ) }
+
+    let(:page) { PDFium::Page.open(utfdoc,0) }
+
+
+    it "opens existing page" do
+        assert_kind_of PDFium::Page, page
+    end
+
+    it "creates a page" do
+        page = PDFium::Page.create(utfdoc,0)
+        assert_kind_of PDFium::Page, page
+    end
+
+    it "adds pages to an existing document" do
+        pdf = PDFium::Document.new
+        PDFium::Page.create(pdf)
+        PDFium::Page.create(pdf)
+        assert_equal 2, pdf.page_count
+    end
+
+    it "has dimensions" do
+        assert_in_delta 595.28, page.width
+        assert_in_delta 841.89, page.height
+    end
+
+    it "refuses to open invalid page ranges" do
+        assert_raises(RangeError) do
+            PDFium::Page.open(utfdoc,-1)
+        end
+        assert_raises(RangeError) do
+            PDFium::Page.open(utfdoc,90)
+        end
+    end
+
+    it "creates new pages" do
+        pdf = PDFium::Document.new
+        page = PDFium::Page.create(pdf,0, width:100, height:180)
+        assert_equal 100, page.width
+        after_saving(pdf) do | saved |
+            assert_equal 1, saved.page_count
+            assert_equal 100.0, saved.page_at(0).width
+        end
+    end
+
+    it "can't be created using new" do
+        assert_raises(RuntimeError){ PDFium::Page.new }
+    end
+
+    it "can load/unload page" do
+        pdf = PDFium::Document.new
+        10.times do
+          page = PDFium::Page.create(pdf)
+          page.unload
+        end
+        assert_equal 10, pdf.page_count
+        page = pdf.page_at(1)
+        assert_equal 612.0, page.width
+        page.unload
+        assert_equal 612.0, page.width
+    end
+
+    it "can read text" do
+        ascii_text = PDFium::Page.open(textdoc,1).text.encode!("ASCII-8BIT")
+        assert_match /Cras tellus. Fusce aliquet/, ascii_text
+    end
+
+    it "can read utf text" do
+        strings = [
+            "Жълтата дюля беше щастлива",
+            "Jove xef, porti whisky amb quinze glaçons d'hidrogen",
+            "Příliš žluťoučký kůň úpěl ďábelské ódy",
+            "Høj bly gom vandt fræk sexquiz på wc",
+            "Doch Bep, flink sexy qua vorm, zwijgt",
+            "Törkylempijä vongahdus",
+            "Falsches Üben von Xylophonmusik quält jeden größeren Zwerg"
+        ]
+        text = PDFium::Page.open(utfdoc,0).text
+        utf8 = text.encode("UTF-8")
+        strings.each do | sentence |
+            assert_match sentence, utf8
+            assert_match sentence.encode("UTF-16LE"), text
+        end
+    end
+
+end

data/test/pdfium_spec.rb

@@ -0,0 +1,15 @@
+require_relative 'spec_helper'
+require 'tempfile'
+
+
+describe PDFium do
+
+    it "creates classes" do
+        assert PDFium::Document
+        assert PDFium::Page
+        assert PDFium::Bookmark
+        assert PDFium::BookmarkList
+    end
+
+
+end

data/test/profile.rb

@@ -0,0 +1,29 @@
+#!/usr/bin/env ruby
+
+require_relative '../lib/pdfium'
+require 'pathname'
+
+puts "Waiting for profiler attachment (PID: #{Process.pid})\nPress enter to continue"
+gets
+
+path = Pathname.new(__FILE__).dirname.join('pdfs','example_images.pdf').to_s
+
+guide = PDFium::Document.new( path )
+`rm /tmp/images/*`
+page = guide.page_at(0)
+page.each_image do |img|
+    img.save("/tmp/images/#{img.index}.png")
+end
+
+# count = 0
+# pdf.bookmarks.each do | bm |
+#     count +=1
+#     print count.to_s + "   "
+#     puts bm.title
+# end
+
+# pdf.each_page do | page |
+#     page.width
+# end
+
+# GC.start(full_mark: true, immediate_sweep: true)