pdf-reader 2.4.0 → 2.6.0

data/lib/pdf/reader/filter/flate.rb

@@ -8,6 +8,9 @@ class PDF::Reader
   module Filter # :nodoc:
     # implementation of the Flate (zlib) stream filter
     class Flate
+      ZLIB_AUTO_DETECT_ZLIB_OR_GZIP = 47  # Zlib::MAX_WBITS + 32
+      ZLIB_RAW_DEFLATE              = -15 # Zlib::MAX_WBITS * -1
+
       def initialize(options = {})
         @options = options
       end
@@ -15,25 +18,34 @@ class PDF::Reader
       ################################################################################
       # Decode the specified data with the Zlib compression algorithm
       def filter(data)
-        deflated = nil
+        deflated = zlib_inflate(data) || zlib_inflate(data[0, data.bytesize-1])
+
+        if deflated.nil?
+          raise MalformedPDFError,
+            "Error while inflating a compressed stream (no suitable inflation algorithm found)"
+        end
+        Depredict.new(@options).filter(deflated)
+      end
+
+      private
+
+      def zlib_inflate(data)
         begin
-          deflated = Zlib::Inflate.new.inflate(data)
-        rescue Zlib::DataError => e
+          return Zlib::Inflate.new(ZLIB_AUTO_DETECT_ZLIB_OR_GZIP).inflate(data)
+        rescue Zlib::DataError
           # by default, Ruby's Zlib assumes the data it's inflating
-          # is RFC1951 deflated data, wrapped in a RFC1951 zlib container.
-          # If that fails, then use an undocumented 'feature' to attempt to inflate
-          # the data as a raw RFC1951 stream.
-          #
-          # See
-          # - http://blade.nagaokaut.ac.jp/cgi-bin/scat.rb/ruby/ruby-talk/243545
-          # - http://www.gzip.org/zlib/zlib_faq.html#faq38
-          deflated = Zlib::Inflate.new(-Zlib::MAX_WBITS).inflate(data)
+          # is RFC1951 deflated data, wrapped in a RFC1950 zlib container. If that
+          # fails, swallow the exception and attempt to inflate the data as a raw
+          # RFC1951 stream.
         end
-        Depredict.new(@options).filter(deflated)
-      rescue Exception => e
-        # Oops, there was a problem inflating the stream
-        raise MalformedPDFError,
-          "Error occured while inflating a compressed stream (#{e.class.to_s}: #{e.to_s})"
+
+        begin
+          return Zlib::Inflate.new(ZLIB_RAW_DEFLATE).inflate(data)
+        rescue StandardError
+          # swallow this one too, so we can try some other fallback options
+        end
+
+        nil
       end
     end
   end

data/lib/pdf/reader/font.rb

@@ -131,7 +131,9 @@ class PDF::Reader
       if obj[:ToUnicode]
         # ToUnicode is optional for Type1 and Type3
         stream = @ohash.object(obj[:ToUnicode])
-        @tounicode = PDF::Reader::CMap.new(stream.unfiltered_data)
+        if stream.is_a?(PDF::Reader::Stream)
+          @tounicode = PDF::Reader::CMap.new(stream.unfiltered_data)
+        end
       end
     end
 

data/lib/pdf/reader/glyph_hash.rb

@@ -103,19 +103,25 @@ class PDF::Reader
 
     # returns a hash that maps glyph names to unicode codepoints. The mapping is based on
     # a text file supplied by Adobe at:
-    # http://www.adobe.com/devnet/opentype/archives/glyphlist.txt
+    # https://github.com/adobe-type-tools/agl-aglfn
     def load_adobe_glyph_mapping
       keyed_by_name      = {}
       keyed_by_codepoint = {}
 
-      File.open(File.dirname(__FILE__) + "/glyphlist.txt", "r:BINARY") do |f|
-        f.each do |l|
-          _m, name, code = *l.match(/([0-9A-Za-z]+);([0-9A-F]{4})/)
-          if name && code
-            cp = "0x#{code}".hex
-            keyed_by_name[name.to_sym]   = cp
-            keyed_by_codepoint[cp]     ||= []
-            keyed_by_codepoint[cp]     << name.to_sym
+      paths = [
+        File.dirname(__FILE__) + "/glyphlist.txt",
+        File.dirname(__FILE__) + "/glyphlist-zapfdingbats.txt",
+      ]
+      paths.each do |path|
+        File.open(path, "r:BINARY") do |f|
+          f.each do |l|
+            _m, name, code = *l.match(/([0-9A-Za-z]+);([0-9A-F]{4})/)
+            if name && code
+              cp = "0x#{code}".hex
+              keyed_by_name[name.to_sym]   = cp
+              keyed_by_codepoint[cp]     ||= []
+              keyed_by_codepoint[cp]     << name.to_sym
+            end
           end
         end
       end

data/lib/pdf/reader/glyphlist-zapfdingbats.txt

@@ -0,0 +1,245 @@
+# -----------------------------------------------------------
+# Copyright 2002-2019 Adobe (http://www.adobe.com/).
+#
+# Redistribution and use in source and binary forms, with or
+# without modification, are permitted provided that the
+# following conditions are met:
+#
+# Redistributions of source code must retain the above
+# copyright notice, this list of conditions and the following
+# disclaimer.
+#
+# Redistributions in binary form must reproduce the above
+# copyright notice, this list of conditions and the following
+# disclaimer in the documentation and/or other materials
+# provided with the distribution.
+#
+# Neither the name of Adobe nor the names of its contributors
+# may be used to endorse or promote products derived from this
+# software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND
+# CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES,
+# INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
+# MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR
+# CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
+# NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
+# LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
+# HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR
+# OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
+# SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+# -----------------------------------------------------------
+# Name:          ITC Zapf Dingbats Glyph List
+# Table version: 2.0
+# Date:          September 20, 2002
+# URL:           https://github.com/adobe-type-tools/agl-aglfn
+#
+# Format: two semicolon-delimited fields:
+#   (1) glyph name--upper/lowercase letters and digits
+#   (2) Unicode scalar value--four uppercase hexadecimal digits
+#
+a100;275E
+a101;2761
+a102;2762
+a103;2763
+a104;2764
+a105;2710
+a106;2765
+a107;2766
+a108;2767
+a109;2660
+a10;2721
+a110;2665
+a111;2666
+a112;2663
+a117;2709
+a118;2708
+a119;2707
+a11;261B
+a120;2460
+a121;2461
+a122;2462
+a123;2463
+a124;2464
+a125;2465
+a126;2466
+a127;2467
+a128;2468
+a129;2469
+a12;261E
+a130;2776
+a131;2777
+a132;2778
+a133;2779
+a134;277A
+a135;277B
+a136;277C
+a137;277D
+a138;277E
+a139;277F
+a13;270C
+a140;2780
+a141;2781
+a142;2782
+a143;2783
+a144;2784
+a145;2785
+a146;2786
+a147;2787
+a148;2788
+a149;2789
+a14;270D
+a150;278A
+a151;278B
+a152;278C
+a153;278D
+a154;278E
+a155;278F
+a156;2790
+a157;2791
+a158;2792
+a159;2793
+a15;270E
+a160;2794
+a161;2192
+a162;27A3
+a163;2194
+a164;2195
+a165;2799
+a166;279B
+a167;279C
+a168;279D
+a169;279E
+a16;270F
+a170;279F
+a171;27A0
+a172;27A1
+a173;27A2
+a174;27A4
+a175;27A5
+a176;27A6
+a177;27A7
+a178;27A8
+a179;27A9
+a17;2711
+a180;27AB
+a181;27AD
+a182;27AF
+a183;27B2
+a184;27B3
+a185;27B5
+a186;27B8
+a187;27BA
+a188;27BB
+a189;27BC
+a18;2712
+a190;27BD
+a191;27BE
+a192;279A
+a193;27AA
+a194;27B6
+a195;27B9
+a196;2798
+a197;27B4
+a198;27B7
+a199;27AC
+a19;2713
+a1;2701
+a200;27AE
+a201;27B1
+a202;2703
+a203;2750
+a204;2752
+a205;276E
+a206;2770
+a20;2714
+a21;2715
+a22;2716
+a23;2717
+a24;2718
+a25;2719
+a26;271A
+a27;271B
+a28;271C
+a29;2722
+a2;2702
+a30;2723
+a31;2724
+a32;2725
+a33;2726
+a34;2727
+a35;2605
+a36;2729
+a37;272A
+a38;272B
+a39;272C
+a3;2704
+a40;272D
+a41;272E
+a42;272F
+a43;2730
+a44;2731
+a45;2732
+a46;2733
+a47;2734
+a48;2735
+a49;2736
+a4;260E
+a50;2737
+a51;2738
+a52;2739
+a53;273A
+a54;273B
+a55;273C
+a56;273D
+a57;273E
+a58;273F
+a59;2740
+a5;2706
+a60;2741
+a61;2742
+a62;2743
+a63;2744
+a64;2745
+a65;2746
+a66;2747
+a67;2748
+a68;2749
+a69;274A
+a6;271D
+a70;274B
+a71;25CF
+a72;274D
+a73;25A0
+a74;274F
+a75;2751
+a76;25B2
+a77;25BC
+a78;25C6
+a79;2756
+a7;271E
+a81;25D7
+a82;2758
+a83;2759
+a84;275A
+a85;276F
+a86;2771
+a87;2772
+a88;2773
+a89;2768
+a8;271F
+a90;2769
+a91;276C
+a92;276D
+a93;276A
+a94;276B
+a95;2774
+a96;2775
+a97;275B
+a98;275C
+a99;275D
+a9;2720
+# END

data/lib/pdf/reader/object_hash.rb

@@ -331,7 +331,9 @@ class PDF::Reader
     def decrypt(ref, obj)
       case obj
       when PDF::Reader::Stream then
-        obj.data = sec_handler.decrypt(obj.data, ref)
+        # PDF 32000-1:2008 7.5.8.2: "The cross-reference stream shall not be encrypted [...]."
+        # Therefore we shouldn't try to decrypt it.
+        obj.data = sec_handler.decrypt(obj.data, ref) unless obj.hash[:Type] == :XRef
         obj
       when Hash                then
         arr = obj.map { |key,val| [key, decrypt(ref, val)] }.flatten(1)

data/lib/pdf/reader/orientation_detector.rb

@@ -22,8 +22,8 @@ class PDF::Reader
     def detect_orientation
       llx,lly,urx,ury = @attributes[:MediaBox]
       rotation        = @attributes[:Rotate].to_i
-      width           = urx.to_i - llx.to_i
-      height          = ury.to_i - lly.to_i
+      width           = (urx.to_i - llx.to_i).abs
+      height          = (ury.to_i - lly.to_i).abs
       if width > height
         (rotation % 180).zero? ? 'landscape' : 'portrait'
       else

data/lib/pdf/reader/page.rb

@@ -124,6 +124,34 @@ module PDF
         }.join(" ")
       end
 
+      # returns the angle to rotate the page clockwise. Always 0, 90, 180 or 270
+      #
+      def rotate
+        value = attributes[:Rotate].to_i
+        case value
+        when 0, 90, 180, 270
+          value
+        else
+          0
+        end
+      end
+
+      # returns the "boxes" that define the page object.
+      # values are defaulted according to section 7.7.3.3 of the PDF Spec 1.7
+      #
+      def boxes
+        mediabox = attributes[:MediaBox]
+        cropbox = attributes[:Cropbox] || mediabox
+
+        {
+          MediaBox: objects.deref!(mediabox),
+          CropBox: objects.deref!(cropbox),
+          BleedBox: objects.deref!(attributes[:BleedBox] || cropbox),
+          TrimBox: objects.deref!(attributes[:TrimBox] || cropbox),
+          ArtBox: objects.deref!(attributes[:ArtBox] || cropbox)
+        }
+      end
+
       private
 
       def root

data/lib/pdf/reader/page_layout.rb

@@ -2,6 +2,7 @@
 # frozen_string_literal: true
 
 require 'pdf/reader/overlapping_runs_filter'
+require 'pdf/reader/zero_width_runs_filter'
 
 class PDF::Reader
 
@@ -17,22 +18,27 @@ class PDF::Reader
     def initialize(runs, mediabox)
       raise ArgumentError, "a mediabox must be provided" if mediabox.nil?
 
-      @runs    = merge_runs(OverlappingRunsFilter.exclude_redundant_runs(runs))
+      runs = ZeroWidthRunsFilter.exclude_zero_width_runs(runs)
+      runs = OverlappingRunsFilter.exclude_redundant_runs(runs)
+      @runs = merge_runs(runs)
       @mean_font_size   = mean(@runs.map(&:font_size)) || DEFAULT_FONT_SIZE
       @mean_font_size = DEFAULT_FONT_SIZE if @mean_font_size == 0
-      @mean_glyph_width = mean(@runs.map(&:mean_character_width)) || 0
-      @page_width  = mediabox[2] - mediabox[0]
-      @page_height = mediabox[3] - mediabox[1]
-      @x_offset = @runs.map(&:x).sort.first
+      @median_glyph_width = median(@runs.map(&:mean_character_width)) || 0
+      @page_width  = (mediabox[2] - mediabox[0]).abs
+      @page_height = (mediabox[3] - mediabox[1]).abs
+      @x_offset = @runs.map(&:x).sort.first || 0
+      lowest_y = @runs.map(&:y).sort.first || 0
+      @y_offset = lowest_y > 0 ? 0 : lowest_y
     end
 
     def to_s
       return "" if @runs.empty?
+      return "" if row_count == 0
 
       page = row_count.times.map { |i| " " * col_count }
       @runs.each do |run|
         x_pos = ((run.x - @x_offset) / col_multiplier).round
-        y_pos = row_count - (run.y / row_multiplier).round
+        y_pos = row_count - ((run.y - @y_offset) / row_multiplier).round
         if y_pos <= row_count && y_pos >= 0 && x_pos <= col_count && x_pos >= 0
           local_string_insert(page[y_pos-1], run.text, x_pos)
         end
@@ -64,7 +70,7 @@ class PDF::Reader
     end
 
     def col_count
-      @col_count ||= ((@page_width  / @mean_glyph_width) * 1.05).floor
+      @col_count ||= ((@page_width  / @median_glyph_width) * 1.05).floor
     end
 
     def row_multiplier
@@ -83,12 +89,12 @@ class PDF::Reader
       end
     end
 
-    def each_line(&block)
-      @runs.sort.group_by { |run|
-        run.y.to_i
-      }.map { |y, collection|
-        yield y, collection
-      }
+    def median(collection)
+      if collection.size == 0
+        0
+      else
+        collection.sort[(collection.size * 0.5).floor]
+      end
     end
 
     # take a collection of TextRun objects and merge any that are in close

data/lib/pdf/reader/page_state.rb

@@ -30,7 +30,7 @@ class PDF::Reader
         @xobject_stack = [page.xobjects]
         @cs_stack      = [page.color_spaces]
         @stack         = [DEFAULT_GRAPHICS_STATE.dup]
-        state[:ctm]    = identity_matrix
+        state[:ctm]  = identity_matrix
       end
 
       #####################################################
@@ -322,11 +322,13 @@ class PDF::Reader
         th = state[:h_scaling]
         # optimise the common path to reduce Float allocations
         if th == 1 && tj == 0 && tc == 0 && tw == 0
-          glyph_width = w0 * fs
-          tx = glyph_width
+          tx = w0 * fs
+        elsif tj != 0
+          # don't apply spacing to TJ displacement
+          tx = (w0 - (tj/1000.0)) * fs * th
         else
-          glyph_width = ((w0 - (tj/1000.0)) * fs) * th
-          tx = glyph_width + ((tc + tw) * th)
+          # apply horizontal scaling to spacing values but not font size
+          tx = ((w0 * fs) + tc + tw) * th
         end
 
         # TODO: I'm pretty sure that tx shouldn't need to be divided by

data/lib/pdf/reader/page_text_receiver.rb

@@ -41,13 +41,17 @@ module PDF
       # starting a new page
       def page=(page)
         @state = PageState.new(page)
+        @page = page
         @content = []
         @characters = []
         @mediabox = page.objects.deref(page.attributes[:MediaBox])
+        device_bl = apply_rotation(*@state.ctm_transform(@mediabox[0], @mediabox[1]))
+        device_tr = apply_rotation(*@state.ctm_transform(@mediabox[2], @mediabox[3]))
+        @device_mediabox = [ device_bl.first, device_bl.last, device_tr.first, device_tr.last]
       end
 
       def content
-        PageLayout.new(@characters, @mediabox).to_s
+        PageLayout.new(@characters, @device_mediabox).to_s
       end
 
       #####################################################
@@ -101,6 +105,8 @@ module PDF
         glyphs.each_with_index do |glyph_code, index|
           # paint the current glyph
           newx, newy = @state.trm_transform(0,0)
+          newx, newy = apply_rotation(newx, newy)
+
           utf8_chars = @state.current_font.to_utf8(glyph_code)
 
           # apply to glyph displacment for the current glyph so the next
@@ -115,6 +121,21 @@ module PDF
         end
       end
 
+      def apply_rotation(x, y)
+        if @page.rotate == 90
+          tmp = x
+          x = y
+          y = tmp * -1
+        elsif @page.rotate == 180
+          y *= -1
+        elsif @page.rotate == 270
+          tmp = x
+          x = y * -1
+          y = tmp * -1
+        end
+        return x, y
+      end
+
     end
   end
 end

data/lib/pdf/reader/parser.rb

@@ -175,15 +175,18 @@ class PDF::Reader
       return "".dup.force_encoding("binary") if str == ")"
       Error.assert_equal(parse_token, ")")
 
-      str.gsub!(/\\([nrtbf()\\\n]|\d{1,3})?|\r\n?|\n\r/m) do |match|
-        MAPPING[match] || "".dup
+      str.gsub!(/\\(\r\n|[nrtbf()\\\n\r]|([0-7]{1,3}))?|\r\n?/m) do |match|
+        if $2.nil? # not octal digits
+          MAPPING[match] || "".dup
+        else # must be octal digits
+          ($2.oct & 0xff).chr # ignore high level overflow
+        end
       end
       str.force_encoding("binary")
     end
 
     MAPPING = {
       "\r"   => "\n",
-      "\n\r" => "\n",
       "\r\n" => "\n",
       "\\n"  => "\n",
       "\\r"  => "\r",
@@ -194,10 +197,9 @@ class PDF::Reader
       "\\)"  => ")",
       "\\\\" => "\\",
       "\\\n" => "",
+      "\\\r" => "",
+      "\\\r\n" => "",
     }
-    0.upto(9)   { |n| MAPPING["\\00"+n.to_s] = ("00"+n.to_s).oct.chr }
-    0.upto(99)  { |n| MAPPING["\\0"+n.to_s]  = ("0"+n.to_s).oct.chr }
-    0.upto(377) { |n| MAPPING["\\"+n.to_s]   = n.to_s.oct.chr }
 
     ################################################################################
     # Decodes the contents of a PDF Stream and returns it as a Ruby String.

data/lib/pdf/reader/width_calculator/built_in.rb

@@ -37,23 +37,15 @@ class PDF::Reader
       def glyph_width(code_point)
         return 0 if code_point.nil? || code_point < 0
 
-        m = @metrics.char_metrics_by_code[code_point]
-        if m.nil?
-          names = @font.encoding.int_to_name(code_point)
+        names = @font.encoding.int_to_name(code_point)
+        metrics = names.map { |name|
+          @metrics.char_metrics[name.to_s]
+        }.compact.first
 
-          m = names.map { |name|
-            @metrics.char_metrics[name.to_s]
-          }.compact.first
-        end
-
-        if m
-          m[:wx]
-        elsif @font.widths[code_point - 1]
-          @font.widths[code_point - 1]
-        elsif control_character?(code_point)
-          0
+        if metrics
+          metrics[:wx]
         else
-          0
+          @font.widths[code_point - 1] || 0
         end
       end
 

data/lib/pdf/reader/xref.rb

@@ -131,6 +131,9 @@ class PDF::Reader
             generation = buf.token.to_i
             state = buf.token
 
+            # Some PDF writers start numbering at 1 instead of 0. Fix up the number.
+            # TODO should this fix be logged?
+            objid = 0 if objid == 1 and offset == 0 and generation == 65535 and state == 'f'
             store(objid, generation, offset + @junk_offset) if state == "n" && offset > 0
             objid += 1
             params.clear
@@ -146,7 +149,9 @@ class PDF::Reader
       end
 
       load_offsets(trailer[:XRefStm])   if trailer.has_key?(:XRefStm)
-      load_offsets(trailer[:Prev].to_i) if trailer.has_key?(:Prev)
+      # Some PDF creators seem to use '/Prev 0' in trailer if there is no previous xref
+      # It's not possible for an xref to appear at offset 0, so can safely skip the ref
+      load_offsets(trailer[:Prev].to_i) if trailer.has_key?(:Prev) and trailer[:Prev].to_i != 0
 
       trailer
     end

data/lib/pdf/reader/zero_width_runs_filter.rb

@@ -0,0 +1,11 @@
+# coding: utf-8
+
+class PDF::Reader
+  # There's no point rendering zero-width characters
+  class ZeroWidthRunsFilter
+
+    def self.exclude_zero_width_runs(runs)
+      runs.reject { |run| run.width == 0 }
+    end
+  end
+end