Working with PDF Forms in Rails

Using PDFTK and the pdf-forms gem

We recently worked on a client project that involved using a Rails backend to fill out PDF Forms. PDF Forms are those PDFs with editable fields, so you can insert text fields, checkboxes, signatures, etc in the file. There weren’t many resources out there for working with PDF forms so here is some of what we learned.

The most popular gem (in terms of downloads on RubyGems) for working with these files is pdf-forms. So that is the one we settled on trying out first. It offers some basic functionality to read the fields and then fill the fields. Here is a brief overview of how to work with PDF Forms in a Ruby On Rails application.


First, we need a few tools, the aforementioned gem pdf-forms, and the command line tool PDFtk.


pdf-forms gem

The first step is to get the pdf-forms gem in your project. If you have a gemfile add it with gem 'pdf-forms'

Then run bundle install. Otherwise, install it with gem install pdf-forms and pull the gem into your ruby file with require pdf-forms


Now since the gem is a wrapper for the PDFtk CLI we will need to install that as well. PDFtk is a tool from PDFLabs that allows you to interact with PDFs from the command line. The link on the page takes you to an older version of the tool which didn’t work for me on macOS Catalina 10.15. The answer on this StackOverflow thread provides a link to the updated Mac package, which you can download here.

After installing PDFtk you can explore it in the command line. I mostly tried out the dump_data_fields command which you run with the prompt pdftk your_document.pdf dump_data_fields replacing your_document.pdf with the file path for the PDF you want to explore. Dump_data_fields will print out the metadata for the fields on the PDF. For example:

FieldType: Text
FieldName: Buyer
FieldNameAlt: Buyer 
FieldFlags: 5768935
FieldJustification: Left
FieldType: Button
FieldName: Joint Buyer
FieldNameAlt: Property
FieldFlags: 0
FieldValue: Off
FieldJustification: Left
FieldStateOption: Off
FieldStateOption: On

With that set up you and working are ready to go!


When instantiating an instance of the PdfForms class, it takes in one argument which is the file path to PDFtk in your environment. I did not do anything special when installing so my path was the default one: '/usr/local/bin/pdftk'

This means I create a new PdfForms instance like so:

form_reader ='/usr/local/bin/pdftk’)


With our new PDFForms instance created, the next step is to read the fields on the pdf form.

There are two helpful methods given to us from the gem: 

Returns an array of Field objects that each hold the information about an individual field on the PDF.

1. get_fields

get_fields takes a single argument, the whole file path to the pdf form you want to read.



   #<PdfForms::Field:0x00007f7bfde14c10 @flags=”5768935″, @justification=”Left”, @name=”Buyer”, @name_alt=”Buyer”, @type=”Text”>,
   #<PdfForms::Field:0x00007f7bbfc3cdb8 @flags=”0″, @justification=”Left”, @name=”Joint Buyer”, @name_alt=”Property”, @options=[“Off”, “On”], @type=”Button”, @value=”Off”>,

You can see this same info matches what pdftk dump_data_fields returned, but now conveniently turned into Ruby objects for us to interact with.

2.  get_field_names

get_field_names does what it sounds like, it returns an array of just the name attribute from all those field objects.



“Joint Buyer”


Now the fun part is writing your custom data onto the pdf. Use the method fill_form to do this.

It takes several arguments, including an optional one. First is the full file path to the pdf form you are using. Then it takes a name to be used as for the filled-out PDF form (which will be a new file in your project’s root directory, the original pdf form will be unchanged). The third is a hash of the data you want to be filled in. Lastly is an optional arguments hash. Here is an example without the optional arguments

file_path = “your_document.pdf”
new_file_name = “completed_form.pdf”
form_data = {
              "Buyer" =>"Orson Welles",
              "Joint Tenants" =>"On",
form_reader.fill_form(file_path, new_file_name, form_data)

To build the hash with the field values, the key will be the name of the field (exactly as it appears in the pdf metadata)  and the value will be a string of whatever data you want to be added to the form.

The tricky part is working with other field types. Checkboxes have two options defined on the PDF form. One of them will make the box checked and the other will make it unchecked. A common one is Yes’and Off. Why? I don’t know. But this is where reading the field data will be necessary either with form_reader.get_fields or from the command line with PDFtk dump_data_fields. Each checkbox should have two FieldTypeOptions defined on it, and that is your source of truth for that specific field. When using the gem you can see these combined into an array and stored as Field.options.

More info about how to check checkboxes

On my test PDF form, I had a field with unchecked defined as Off and checked defined as Checkbox:plus:option:plus:2 whatever that means… So you will have to check your own PDF fields to confirm what values you need.

Lastly, when working with Signature field types I have not found a way to write a value with the gem or PDFtk. It seems that there are some additional layers that Adobe has attached to signature fields for security purposes. If you find a way to write to Signature fields, let us know!

There is a quick overview of how to work with PDF Forms in Rails using the pdf-forms gem. Hopefully this is helpful!