combine_pdf 1.0.9 → 1.0.10

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: f416a0cbf333fe7e6d49c844a9ff6bbc2c735077bb325d5702019b86c01e0cc7
-  data.tar.gz: d5af83b8d3bbb89456edb98fd8961d6de10405d06f15b80297576bc16692a7a5
+  metadata.gz: 7fced5e9fb501777c32a0c483d00c9ae0ff3875406cd6367eb8a1ac5193a5ba3
+  data.tar.gz: '08bdd4f49d69a25a2d83b0a28dc13d6e07c2e100f986404a516f1995f11db2b9'
 SHA512:
-  metadata.gz: ed94ad7367e9521a5753656148fe654a225c33427b66c89724ea9ea175799b8f20467bb3adb201770ce61848ca1c22e767ac626cfeaa708bde18609645a4ed9a
-  data.tar.gz: eccf69bbd4de266e6b9e7de1f7a0a6ecbf615eb14921940808725c4407181f1e62e89215033a9af15b29659041d7b1b0b97f45a9086adedff38e0312d7206b73
+  metadata.gz: b55bfcb34ebc64b3fc9749dfb119bd719d873d814fa340443bf346aeab233de151c6d57f2c90cf25d50053c3880b0d3f946fa028ef353b4356e02537ef729467
+  data.tar.gz: e1307d1b99aa5c174b2bde19311213fcd69248ba2fcb622e1497b1981e06fd026f7493c33fc30e38861460849c25f5072644223a241e7cf9e4b1489170c93cd4

data/CHANGELOG.md

@@ -2,9 +2,13 @@
 
 ***
 
+#### Change log v.1.0.10
+
+**Fix**: Fixed an issue related to issue #131 where parsing would fail if the `xref` section appears to be misplaced within the PDF. Credit to @bharat303 (Bharat Godhani) for exposing this issue.
+
 #### Change log v.1.0.9
 
-**Fix**: Fixed issue #136 where the `#fix_rotation` function would rotate the page to the wrong direction. Credit to @dmkash for exposing this issue
+**Fix**: Fixed issue #136 where the `#fix_rotation` function would rotate the page to the wrong direction. Credit to @dmkash for exposing this issue.
 
 #### Change log v.1.0.8
 

data/lib/combine_pdf/parser.rb

@@ -80,6 +80,7 @@ module CombinePDF
       # puts @parsed
 
       unless (@parsed.select { |i| !i.is_a?(Hash) }).empty?
+        # p @parsed.select
         raise ParsingError, 'Unknown PDF parsing error - malformed PDF file?'
       end
 
@@ -385,17 +386,6 @@ module CombinePDF
           out.last[:Dest] = unify_string(out.last[:Dest].to_s) if out.last[:Dest] && out.last[:Dest].is_a?(Symbol)
         # puts "!!!!!!!!! Error with :indirect_reference_id\n\nObject #{out.last}  :indirect_reference_id = #{out.last[:indirect_reference_id]}" unless out.last[:indirect_reference_id].is_a?(Numeric)
         ##########################################
-        ## Parse a comment
-        ##########################################
-        elsif str = @scanner.scan(/\%/)
-          # is a comment, skip until new line
-          loop do
-            # break unless @scanner.scan(/[^\d\r\n]+/)
-            break if @scanner.check(/([\d]+[\s]+[\d]+[\s]+obj[\s]+\<\<)|([\n\r]+)/) || @scanner.eos? # || @scanner.scan(/[^\d]+[\r\n]+/) ||
-            @scanner.scan(/[^\d\r\n]+/) || @scanner.pos += 1
-          end
-        # puts "AFTER COMMENT: #{@scanner.peek 8}"
-        ##########################################
         ## Parse an Object Reference
         ##########################################
         elsif @scanner.scan(/R/)
@@ -414,32 +404,57 @@ module CombinePDF
         elsif @scanner.scan(/null/)
           out << nil
         ##########################################
+        ## Parse file trailer
+        ##########################################
+        elsif @scanner.scan(/trailer/)
+          if @scanner.skip_until(/<</)
+            data = _parse_
+            (@root_object ||= {}).clear
+            @root_object[data.shift] = data.shift while data[0]
+          end
+        ##########################################
         ## XREF - check for encryption... anything else?
         ##########################################
-        elsif @scanner.scan(/(startxref)|(xref)/)
-          ##########
-          ## get root object to check for encryption
-          @scanner.scan_until(/(trailer)|(\%EOF)/)
-          fresh = true
-          if @scanner.matched[-1] == 'r'
-            if @scanner.skip_until(/<</)
-              data = _parse_
-              (@root_object ||= {}).clear
-              @root_object[data.shift] = data.shift while data[0]
-            end
-            ##########
-            ## skip untill end of segment, maked by %%EOF
-            @scanner.skip_until(/\%\%EOF/)
-            ##########
-            ## If this was the last valid segment, ignore any trailing garbage
-            ## (issue #49 resolution)
-            break unless @scanner.exist?(/\%\%EOF/)
-
+        elsif @scanner.scan(/xref/)
+          # skip first xref line
+          @scanner.scan(/[\s]+[\d]+[\s]+[\d]+[\s]+/)
+          while @scanner.scan(/[\d]+[\s][\d]+[\s]+[nf][\s]+/)
+            # skip all xref lines
+            nil
           end
-
+        ##########################################
+        ## XREF location can be ignored
+        ##########################################
+        elsif @scanner.scan(/startxref/)
+          @scanner.scan(/[\s]+[\d]+[\s]+/)
+        ##########################################
+        ## Skip Whitespace
+        ##########################################
         elsif @scanner.scan(/[\s]+/)
           # Generally, do nothing
           nil
+        ##########################################
+        ## EOF?
+        ##########################################
+        elsif @scanner.scan(/\%\%EOF/)
+          ##########
+          ## If this was the last valid segment, ignore any trailing garbage
+          ## (issue #49 resolution)
+          break unless @scanner.exist?(/\%\%EOF/)
+        ##########################################
+        ## Parse a comment
+        ##########################################
+        elsif str = @scanner.scan(/\%/)
+          # is a comment, skip until new line
+          loop do
+            # break unless @scanner.scan(/[^\d\r\n]+/)
+            break if @scanner.check(/([\d]+[\s]+[\d]+[\s]+obj[\s]+\<\<)|([\n\r]+)/) || @scanner.eos? # || @scanner.scan(/[^\d]+[\r\n]+/) ||
+            @scanner.scan(/[^\d\r\n]+/) || @scanner.pos += 1
+          end
+        # puts "AFTER COMMENT: #{@scanner.peek 8}"
+        ##########################################
+        ## Fix wkhtmltopdf - missing 'endobj' keywords
+        ##########################################
         elsif @scanner.scan(/obj[\s]*/)
           # Fix wkhtmltopdf PDF authoring issue - missing 'endobj' keywords
           unless fresh || (out[-4].nil? || out[-4].is_a?(Hash))
@@ -460,6 +475,9 @@ module CombinePDF
             out << keep.pop
           end
           fresh = false
+        ##########################################
+        ## Unknown, warn and advance
+        ##########################################
         else
           # always advance
           # warn "Advancing for unknown reason... #{@scanner.string[@scanner.pos - 4, 8]} ... #{@scanner.peek(4)}" unless @scanner.peek(1) =~ /[\s\n]/

data/lib/combine_pdf/version.rb

@@ -1,3 +1,3 @@
 module CombinePDF
-  VERSION = '1.0.9'.freeze
+  VERSION = '1.0.10'.freeze
 end

data/test/automated

@@ -95,6 +95,7 @@ pdf.save('07_named destinations_numbered.pdf')
 CombinePDF.load("./Ruby/test\ pdfs/Scribus-unknown_err.pdf").save '08_1-unknown-err-empty-str.pdf'
 CombinePDF.load("./Ruby/test\ pdfs/Scribus-unknown_err2.pdf").save '08_2-unknown-err-empty-str.pdf'
 CombinePDF.load("./Ruby/test\ pdfs/Scribus-unknown_err3.pdf").save '08_3-unknown-err-empty-str.pdf'
+CombinePDF.load("./Ruby/test\ pdfs/xref_in_middle.pdf").save '08_4-xref-in-middle.pdf'
 
 CombinePDF.load("/Users/2Be/Ruby/test\ pdfs/nil_object.pdf").save('09_nil_in_parsed_array.pdf')
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: combine_pdf
 version: !ruby/object:Gem::Version
-  version: 1.0.9
+  version: 1.0.10
 platform: ruby
 authors:
 - Boaz Segev
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-02-07 00:00:00.000000000 Z
+date: 2018-05-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ruby-rc4