pdf-forms 1.0.0 → 1.3.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
-SHA1:
-  metadata.gz: 74a388e3ee6551d601f4d713863ff18d8d71e7bc
-  data.tar.gz: 28c603be2710e14695777208fa1b2219422c67ca
+SHA256:
+  metadata.gz: 40651e7dd3bdc30c69263f76ecc3883c8be80f985e431509ee51e9d51bc0aa8d
+  data.tar.gz: 7e0ab952ddd42be27d4a58de7acdda3d7ef54d1588e889432ec3be922ac1b722
 SHA512:
-  metadata.gz: e7ba4e9bca7cf23c37e431a646c766486a748a9836fb7d3e432ea7ecf814523d1932276f81159c02af298562fdf2653933c2a3c6be16a8d54496e9469d29238b
-  data.tar.gz: b2e6c9002bcc6999c7b33ccf47eb82c22d3699bb709f6085f2b15b23dd0d4e94ad18e34eee14851f4cf0af5ddd031e779445bb6e1c57d207fd055ea12e64df1f
+  metadata.gz: d12c30fbdeabb1443f33d6f52af39e442241ce44158f9095ca909d3de434caef47df83156ba05ae08169d6bed5840220fb8d80dda673c441c3e414fd64ca3a8e
+  data.tar.gz: 9a7c21dc930ffe1249258f730cd628bf77a115fd64f84fef5df994fdef5e88e3be227d09e0cdcb862550a8ede7f25afb7fdb73083297f9c0958e998b748f5b6b

data/README.md

@@ -11,11 +11,35 @@ Fill out PDF forms with [pdftk](http://www.pdflabs.com/tools/pdftk-server/).
 You'll need a working `pdftk` binary. Either get a binary package from
 http://www.pdflabs.com/tools/pdftk-server/ and install it, or run
 `apt-get install pdftk` if you're on Debian or similar.
-[Homebrew Cask](http://caskroom.io) also has a pftk formula.
 
 After that, add `pdf-forms` to your Gemfile or manually install the gem. Nothing
 unusual here.
 
+### Using the Java port of PDFTK
+
+The PDFTK package was dropped from most (all?) current versions of major Linux distributions.
+As contributed in [this issue](https://github.com/jkraemer/pdf-forms/issues/75#issuecomment-698436643), you can use the [Java version of PDFTK](https://gitlab.com/pdftk-java/pdftk)
+with this gem, as well. Just create a small shell script:
+
+~~~shell
+#!/bin/sh
+MYSELF=`which "$0" 2>/dev/null`
+[ $? -gt 0 -a -f "$0" ] && MYSELF="./$0"
+java=java
+if test -n "$JAVA_HOME"; then
+    java="$JAVA_HOME/bin/java"
+fi
+exec "$java" $java_args -jar $MYSELF "$@"
+exit 1
+~~~
+
+Next, concatenate the wrapper script and the Jar file, and you end up with an executable that 
+can be used with pdf-forms:
+
+~~~
+cat stub.sh pdftk-all.jar > pdftk.run && chmod +x pdftk.run
+~~~
+
 
 ## Usage
 
@@ -41,6 +65,9 @@ require 'pdf_forms'
 # add :data_format => 'XFdf' option to generate XFDF instead of FDF when
 # filling a form (XFDF is supposed to have better support for non-western
 # encodings)
+# add :data_format => 'FdfHex' option to generate FDF with values passed in
+# UTF16 hexadecimal format (Hexadecimal format has also proven more reliable
+# for passing latin accented characters to pdftk)
 # add :utf8_fields => true in order to get UTF8 encoded field metadata (this
 # will use dump_data_fields_utf8 instead of dump_data_fields in the call to
 # pdftk)
@@ -52,15 +79,42 @@ pdftk.get_field_names 'path/to/form.pdf'
 # take form.pdf, set the 'foo' field to 'bar' and save the document to myform.pdf
 pdftk.fill_form '/path/to/form.pdf', 'myform.pdf', :foo => 'bar'
 
-# optionally, add the :flatten option to prevent editing of a filled out form
+# optionally, add the :flatten option to prevent editing of a filled out form.
+# Other supported options are :drop_xfa and :drop_xmp.
 pdftk.fill_form '/path/to/form.pdf', 'myform.pdf', {:foo => 'bar'}, :flatten => true
+
+# to enable PDF encryption, pass encrypt: true. By default, a random 'owner
+# password' will be used, but you can also set one with the :encrypt_pw option.
+pdftk.fill_form '/path/to/form.pdf', 'myform.pdf', {foo: 'bar'}, encrypt: true, encrypt_options: 'allow printing'
+
+# you can also protect the PDF even from opening by specifying an additional user_pw option:
+pdftk.fill_form '/path/to/form.pdf', 'myform.pdf', {foo: 'bar'}, encrypt: true, encrypt_options: 'user_pw secret'
 ```
 
+Any options shown above can also be set when initializing the PdfForms
+instance. In this case, options given to `fill_form` will override the global
+options.
+
+### Field names with HTML entities
+
+In case your form's field names contain HTML entities (like
+`Straße Hausnummer`), make sure you unescape those before using them, i.e.
+`CGI.unescapeHTML(name)`.  Thanks to @phoet for figuring this out in #65.
+
+### Non-ASCII Characters (UTF8 etc) are not displayed in the filled out PDF
+
+First, check if the field value has been stored properly in the output PDF using `pdftk output.pdf dump_data_fields_utf8`.
+
+If it has been stored but is not rendered, your input PDF lacks the proper font for your kind of characters. Re-create it and embed any necessary fonts.
+If the value has not been stored, there is a problem with filling out the form, either on your side, of with this gem.
+
+Also see [UTF-8 chars are not displayed in the filled PDF](https://github.com/jkraemer/pdf-forms/issues/53)
+
 ### Prior Art
 
 The FDF generation part is a straight port of Steffen Schwigon's PDF::FDF::Simple perl module. Didn't port the FDF parsing, though ;-)
 
 ## License
 
-Created by [Jens Kraemer](http://jkraemer.net/) and licensed under the MIT Liense.
+Created by [Jens Kraemer](http://jkraemer.net/) and licensed under the MIT License.
 

data/lib/pdf_forms.rb

@@ -5,6 +5,7 @@ require 'pdf_forms/normalize_path'
 require 'pdf_forms/data_format'
 require 'pdf_forms/fdf'
 require 'pdf_forms/xfdf'
+require 'pdf_forms/fdf_hex'
 require 'pdf_forms/pdf'
 require 'pdf_forms/pdftk_wrapper'
 

data/lib/pdf_forms/fdf.rb

@@ -28,6 +28,7 @@ module PdfForms
       end
     end
 
+    # pp 559 https://www.adobe.com/content/dam/acom/en/devnet/pdf/pdfs/pdf_reference_archives/PDFReference.pdf
     def header
       header = "%FDF-1.2\n\n1 0 obj\n<<\n/FDF << /Fields 2 0 R"
 
@@ -39,13 +40,16 @@ module PdfForms
       header << "/ID[" << options[:id].join << "]" if options[:id]
 
       header << ">>\n>>\nendobj\n2 0 obj\n["
-      return header
+      header
     end
 
+    # pp 561 https://www.adobe.com/content/dam/acom/en/devnet/pdf/pdfs/pdf_reference_archives/PDFReference.pdf
     def field(key, value)
-      "<</T(#{key})/V" +
-        (Array === value ? "[#{value.map{ |v|"(#{quote(v)})" }.join}]" : "(#{quote(value)})") +
-        ">>\n"
+      field = "<<"
+      field << "/T" + "(#{key})"
+      field << "/V" + (Array === value ? "[#{value.map{ |v|"(#{quote(v)})" }.join}]" : "(#{quote(value)})")
+      field << ">>\n"
+      field
     end
 
     def quote(value)

data/lib/pdf_forms/fdf_hex.rb

@@ -0,0 +1,39 @@
+# coding: UTF-8
+
+module PdfForms
+  # Map keys and values to Adobe's FDF format.
+  #
+  # This is a variation of the original Fdf data format, values are encoded in UTF16 hexadesimal
+  # notation to improve compatibility with non ascii charsets.
+  #
+  # Information about hexadesimal FDF values was found here:
+  #
+  # http://stackoverflow.com/questions/6047970/weird-characters-when-filling-pdf-with-pdftk
+  #
+  class FdfHex < Fdf
+    private
+
+    def field(key, value)
+      "<</T(#{key})/V" +
+        (Array === value ? encode_many(value) : encode_value_as_hex(value)) +
+        ">>\n"
+    end
+
+    def encode_many(values)
+      "[#{values.map { |v| encode_value_as_hex(v) }.join}]"
+    end
+
+    def encode_value_as_hex(value)
+      value = value.to_s
+      utf_16 = value.encode('UTF-16BE', :invalid => :replace, :undef => :replace)
+      hex = utf_16.unpack('H*').first
+      hex.force_encoding 'ASCII-8BIT' # jruby
+      '<FEFF' + hex.upcase + '>'
+    end
+
+    # Fdf implementation encodes to ISO-8859-15 which we do not want here.
+    def encode_data(fdf)
+      fdf
+    end
+  end
+end

data/lib/pdf_forms/field.rb

@@ -9,35 +9,45 @@ module PdfForms
     # FieldStateOption: Ja
     # FieldStateOption: Off
     #
-    # Represenation of a PDF Form Field
+    # Representation of a PDF Form Field
     def initialize(field_description)
+      last_value = nil
       field_description.each_line do |line|
-        case line
-        when /FieldStateOption:\s*(.*?)\s*$/
-          (@options ||= []) << $1
-        else
-          line.strip!
-          key, value = line.split(": ")
-          key.gsub!(/Field/, "")
-          key = key.split(/(?=[A-Z])/).map(&:downcase).join('_').split(":")[0]
-          
-          instance_variable_set("@#{key}", value)
-          
-          # dynamically add in fields that we didn't anticipate in ATTRS
-          unless self.respond_to?(key.to_sym)
-            proc = Proc.new { instance_variable_get("@#{key}".to_sym) }
-            self.class.send(:define_method, key.to_sym, proc)
+        line.chomp!
+
+        if line =~ /^Field([A-Za-z]+):\s+(.*)/
+          _, key, value = *$~
+
+          if key == 'StateOption'
+            (@options ||= []) << value
+
+          else
+            value.chomp!
+            last_value = value
+            key = key.split(/(?=[A-Z])/).map(&:downcase).join('_')
+            instance_variable_set("@#{key}", value)
+
+            # dynamically add in fields that we didn't anticipate in ATTRS
+            unless self.respond_to?(key.to_sym)
+              proc = Proc.new { instance_variable_get("@#{key}".to_sym) }
+              self.class.send(:define_method, key.to_sym, proc)
+            end
           end
+
+        else
+          # pdftk returns a line that doesn't start with "Field"
+          # It happens when a text field has multiple lines
+          last_value << "\n" << line
         end
       end
     end
-    
+
     def to_hash
       hash = {}
       ATTRS.each do |attribute|
         hash[attribute] = self.send(attribute)
       end
-      
+
       hash
     end
 

data/lib/pdf_forms/normalize_path.rb

@@ -1,12 +1,13 @@
 # coding: UTF-8
 
+require 'pathname'
+
 module PdfForms
 
   module NormalizePath
 
     def normalize_path(path)
-      path = path.to_path if path.respond_to? :to_path
-      path.to_str
+      Pathname(path).to_path
     end
 
   end

data/lib/pdf_forms/pdftk_wrapper.rb

@@ -3,6 +3,7 @@
 require 'tempfile'
 require 'cliver'
 require 'safe_shell'
+require 'securerandom'
 
 module PdfForms
   class PdftkError < StandardError
@@ -22,6 +23,9 @@ module PdfForms
     #
     # The pdftk binary may also be explecitly specified:
     # PdftkWrapper.new('/usr/bin/pdftk', :flatten => true, :encrypt => true, :encrypt_options => 'allow Printing')
+    #
+    # Besides the options shown above, the drop_xfa or drop_xmp options are
+    # also supported.
     def initialize(*args)
       pdftk, options = normalize_args *args
       @pdftk = Cliver.detect! pdftk
@@ -40,8 +44,7 @@ module PdfForms
       fill_options = {:tmp_path => tmp.path}.merge(fill_options)
 
       args = [ q_template, 'fill_form', normalize_path(tmp.path), 'output', q_destination ]
-      append_options args, fill_options
-      result = call_pdftk *args
+      result = call_pdftk(*(append_options(args, fill_options)))
 
       unless File.readable?(destination) && File.size(destination) > 0
         fdf_path = nil
@@ -85,13 +88,28 @@ module PdfForms
       SafeShell.execute pdftk, *(args.flatten)
     end
 
-    # concatenate documents
+    # concatenate documents, can optionally specify page ranges
     #
-    # args: in_file1, in_file2, ... , in_file_n, output
+    # args: in_file1, {in_file2 => ["1-2", "4-10"]}, ... , in_file_n, output
     def cat(*args)
-      arguments = args.flatten.compact.map{|path| normalize_path(path)}
-      output = arguments.pop
-      call_pdftk arguments, 'output', output
+      in_files = []
+      page_ranges = []
+      file_handle = "A"
+      output = normalize_path args.pop
+
+      args.flatten.compact.each do |in_file|
+        if in_file.is_a? Hash
+          path = in_file.keys.first
+          page_ranges.push *in_file.values.first.map {|range| "#{file_handle}#{range}"}
+        else
+          path = in_file
+          page_ranges.push "#{file_handle}"
+        end
+        in_files.push "#{file_handle}=#{normalize_path(path)}"
+        file_handle.next!
+      end
+
+      call_pdftk in_files, 'cat', page_ranges, 'output', output
     end
 
     # stamp one pdf with another
@@ -118,17 +136,30 @@ module PdfForms
       local[attrib] || options[attrib]
     end
 
+    ALLOWED_OPTIONS = %i(
+      drop_xmp
+      drop_xfa
+      flatten
+      need_appearances
+    ).freeze
+
     def append_options(args, local_options = {})
-      return if options.empty? && local_options.empty?
-      if option_or_global(:flatten, local_options)
-        args << 'flatten'
+      return args if options.empty? && local_options.empty?
+      args = args.dup
+      ALLOWED_OPTIONS.each do |option|
+        if option_or_global(option, local_options)
+          args << option.to_s
+        end
       end
       if option_or_global(:encrypt, local_options)
         encrypt_pass = option_or_global(:encrypt_password, local_options)
-        encrypt_pass ||= option_or_global(:tmp_path, local_options)
+        encrypt_pass ||= SecureRandom.urlsafe_base64
         encrypt_options = option_or_global(:encrypt_options, local_options)
-        args += ['encrypt_128bit', 'owner_pw', encrypt_pass, encrypt_options].flatten.compact
+        encrypt_options = encrypt_options.split if String === encrypt_options
+        args << ['encrypt_128bit', 'owner_pw', encrypt_pass, encrypt_options]
       end
+      args.flatten!
+      args.compact!
       args
     end
 
@@ -141,11 +172,11 @@ module PdfForms
         when String
           pdftk = first_arg
           options = args.shift || {}
-          raise InvalidArgumentError.new("expected hash, got #{options.class.name}") unless Hash === options
+          raise ArgumentError.new("expected hash, got #{options.class.name}") unless Hash === options
         when Hash
           options = first_arg
         else
-          raise InvalidArgumentError.new("expected string or hash, got #{first_arg.class.name}")
+          raise ArgumentError.new("expected string or hash, got #{first_arg.class.name}")
         end
       end
       [pdftk, options]

data/lib/pdf_forms/version.rb

@@ -1,5 +1,5 @@
 # coding: UTF-8
 
 module PdfForms
-  VERSION = '1.0.0'
+  VERSION = '1.3.0'
 end

data/lib/pdf_forms/xfdf.rb

@@ -1,5 +1,7 @@
 # coding: UTF-8
 
+require 'rexml/document'
+
 module PdfForms
   # Map keys and values to Adobe's XFDF format.
   class XFdf < DataFormat
@@ -14,7 +16,6 @@ module PdfForms
     end
 
     def quote(value)
-      require 'rexml/document'
       REXML::Text.new(value.to_s).to_s
     end
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: pdf-forms
 version: !ruby/object:Gem::Version
-  version: 1.0.0
+  version: 1.3.0
 platform: ruby
 authors:
 - Jens Krämer
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-10-30 00:00:00.000000000 Z
+date: 2020-10-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: cliver
@@ -48,14 +48,14 @@ dependencies:
   name: bundler
   requirement: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
         version: '1.7'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
-    - - ">="
+    - - "~>"
       - !ruby/object:Gem::Version
         version: '1.7'
 - !ruby/object:Gem::Dependency
@@ -86,6 +86,7 @@ files:
 - lib/pdf_forms.rb
 - lib/pdf_forms/data_format.rb
 - lib/pdf_forms/fdf.rb
+- lib/pdf_forms/fdf_hex.rb
 - lib/pdf_forms/field.rb
 - lib/pdf_forms/normalize_path.rb
 - lib/pdf_forms/pdf.rb
@@ -112,9 +113,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: 1.3.6
 requirements: []
 rubyforge_project: pdf-forms
-rubygems_version: 2.4.5.1
+rubygems_version: 2.7.6.2
 signing_key: 
 specification_version: 4
 summary: Fill out PDF forms with pdftk (http://www.accesspdf.com/pdftk/).
 test_files: []
-has_rdoc: