Wicked PDF uses the shell utility wkhtmltopdf to serve a PDF file to a user from HTML. In other words, rather than dealing with a PDF generation DSL of some sort, you simply write an HTML view as you would normally, then let Wicked PDF take care of the hard stuff.
Wicked PDF has been verified to work on Ruby versions 2.2 through 2.6; Rails 4 through 6.1
Add this to your Gemfile and run bundle install
:
gem 'wicked_pdf'
Then create the initializer with
rails generate wicked_pdf
You may also need to add
Mime::Type.register "application/pdf", :pdf
to config/initializers/mime_types.rb
in older versions of Rails.
Because wicked_pdf
is a wrapper for wkhtmltopdf, you'll need to install that, too.
The simplest way to install all of the binaries on most Linux or OSX systems is through the gem wkhtmltopdf-binary. Builds for other systems are available here To install that gem, add this:
gem 'wkhtmltopdf-binary'
To your Gemfile and run bundle install
.
This gem currently installs version 0.12.x of wkhtmltopdf
. Some of the options listed below are specific 0.9 or below, and others are for 0.12 and up.
You can see what flags are supported for the current version in wkhtmltopdf's auto-generated manual
If your wkhtmltopdf executable is not on your webserver's path, you can configure it in an initializer:
WickedPdf.config = {
exe_path: '/usr/local/bin/wkhtmltopdf'
}
For more information about wkhtmltopdf
, see the project's homepage.
class ThingsController < ApplicationController
def show
respond_to do |format|
format.html
format.pdf do
render pdf: "file_name" # Excluding ".pdf" extension.
end
end
end
end
The wkhtmltopdf binary is run outside of your Rails application; therefore, your normal layouts will not work. If you plan to use any CSS, JavaScript, or image files, you must modify your layout so that you provide an absolute reference to these files. The best option for Rails without the asset pipeline is to use the wicked_pdf_stylesheet_link_tag
, wicked_pdf_image_tag
, and wicked_pdf_javascript_include_tag
helpers or to go straight to a CDN (Content Delivery Network) for popular libraries such as jQuery.
<!doctype html>
<html>
<head>
<meta charset='utf-8' />
<%= wicked_pdf_stylesheet_link_tag "pdf" -%>
<%= wicked_pdf_javascript_include_tag "number_pages" %>
</head>
<body onload='number_pages'>
<div id="header">
<%= wicked_pdf_image_tag 'mysite.jpg' %>
</div>
<div id="content">
<%= yield %>
</div>
</body>
</html>
Using wicked_pdf_helpers with asset pipeline raises Asset names passed to helpers should not include the "/assets/" prefix.
error. To work around this, you can use wicked_pdf_asset_base64
with the normal Rails helpers, but be aware that this will base64 encode your content and inline it in the page. This is very quick for small assets, but large ones can take a long time.
<!doctype html>
<html>
<head>
<meta charset='utf-8' />
<%= stylesheet_link_tag wicked_pdf_asset_base64("pdf") %>
<%= javascript_include_tag wicked_pdf_asset_base64("number_pages") %>
</head>
<body onload='number_pages'>
<div id="header">
<%= image_tag wicked_pdf_asset_base64('mysite.jpg') %>
</div>
<div id="content">
<%= yield %>
</div>
</body>
</html>
wicked_pdf supports webpack stylesheets and javascript assets.
Use wicked_pdf_stylesheet_pack_tag
for stylesheets
Use wicked_pdf_javascript_pack_tag
for javascripts
It is best to precompile assets used in PDF views. This will help avoid issues when it comes to deploying, as Rails serves asset files differently between development and production (config.assets.compile = false
), which can make it look like your PDFs work in development, but fail to load assets in production.
config.assets.precompile += ['blueprint/screen.css', 'pdf.css', 'jquery.ui.datepicker.js', 'pdf.js', ...etc...]
In this case, you can use that standard Rails helpers and point to the current CDN for whichever framework you are using. For jQuery, it would look somethng like this, given the current versions at the time of this writing.
<!doctype html>
<html>
<head>
<%= javascript_include_tag "http://code.jquery.com/jquery-1.10.0.min.js" %>
<%= javascript_include_tag "http://code.jquery.com/ui/1.10.3/jquery-ui.min.js" %>
NOTE: Certain options are only supported in specific versions of wkhtmltopdf.
class ThingsController < ApplicationController
def show
respond_to do |format|
format.html
format.pdf do
render pdf: 'file_name',
disposition: 'attachment', # default 'inline'
template: 'things/show',
file: "#{Rails.root}/files/foo.erb",
inline: '<!doctype html><html><head></head><body>INLINE HTML</body></html>',
layout: 'pdf', # for a pdf.pdf.erb file
wkhtmltopdf: '/usr/local/bin/wkhtmltopdf', # path to binary
show_as_html: params.key?('debug'), # allow debugging based on url param
orientation: 'Landscape', # default Portrait
page_size: 'A4, Letter, ...', # default A4
page_height: NUMBER,
page_width: NUMBER,
save_to_file: Rails.root.join('pdfs', "#{filename}.pdf"),
save_only: false, # depends on :save_to_file being set first
default_protocol: 'http',
proxy: 'TEXT',
basic_auth: false # when true username & password are automatically sent from session
username: 'TEXT',
password: 'TEXT',
title: 'Alternate Title', # otherwise first page title is used
cover: 'URL, Pathname, or raw HTML string',
dpi: 'dpi',
encoding: 'TEXT',
user_style_sheet: 'URL',
cookie: ['_session_id SESSION_ID'], # could be an array or a single string in a 'name value' format
post: ['query QUERY_PARAM'], # could be an array or a single string in a 'name value' format
redirect_delay: NUMBER,
javascript_delay: NUMBER,
window_status: 'TEXT', # wait to render until some JS sets window.status to the given string
image_quality: NUMBER,
no_pdf_compression: true,
zoom: FLOAT,
page_offset: NUMBER,
book: true,
default_header: true,
disable_javascript: false,
grayscale: true,
lowquality: true,
enable_plugins: true,
disable_internal_links: true,
disable_external_links: true,
print_media_type: true,
disable_smart_shrinking: true,
use_xserver: true,
background: false, # background needs to be true to enable background colors to render
no_background: true,
viewport_size: 'TEXT', # available only with use_xserver or patched QT
extra: '', # directly inserted into the command to wkhtmltopdf
raise_on_all_errors: nil, # raise error for any stderr output. Such as missing media, image assets
log_level: 'info', # Available values: none, error, warn, or info - only available with wkhtmltopdf 0.12.5+
quiet: false, # `false` is same as `log_level: 'info'`, `true` is same as `log_level: 'none'`
outline: { outline: true,
outline_depth: LEVEL },
margin: { top: SIZE, # default 10 (mm)
bottom: SIZE,
left: SIZE,
right: SIZE },
header: { html: { template: 'users/header', # use :template OR :url
layout: 'pdf_plain', # optional, use 'pdf_plain' for a pdf_plain.html.pdf.erb file, defaults to main layout
url: 'www.example.com',
locals: { foo: @bar }},
center: 'TEXT',
font_name: 'NAME',
font_size: SIZE,
left: 'TEXT',
right: 'TEXT',
spacing: REAL,
line: true,
content: 'HTML CONTENT ALREADY RENDERED'}, # optionally you can pass plain html already rendered (useful if using pdf_from_string)
footer: { html: { template:'shared/footer', # use :template OR :url
layout: 'pdf_plain.html', # optional, use 'pdf_plain' for a pdf_plain.html.pdf.erb file, defaults to main layout
url: 'www.example.com',
locals: { foo: @bar }},
center: 'TEXT',
font_name: 'NAME',
font_size: SIZE,
left: 'TEXT',
right: 'TEXT',
spacing: REAL,
line: true,
content: 'HTML CONTENT ALREADY RENDERED'}, # optionally you can pass plain html already rendered (useful if using pdf_from_string)
toc: { font_name: "NAME",
depth: LEVEL,
header_text: "TEXT",
header_fs: SIZE,
text_size_shrink: 0.8,
l1_font_size: SIZE,
l2_font_size: SIZE,
l3_font_size: SIZE,
l4_font_size: SIZE,
l5_font_size: SIZE,
l6_font_size: SIZE,
l7_font_size: SIZE,
level_indentation: NUM,
l1_indentation: NUM,
l2_indentation: NUM,
l3_indentation: NUM,
l4_indentation: NUM,
l5_indentation: NUM,
l6_indentation: NUM,
l7_indentation: NUM,
no_dots: true,
disable_dotted_lines: true,
disable_links: true,
disable_toc_links: true,
disable_back_links:true,
xsl_style_sheet: 'file.xsl'}, # optional XSLT stylesheet to use for styling table of contents
progress: proc { |output| puts output } # proc called when console output changes
end
end
end
end
By default, it will render without a layout (layout: false) and the template for the current controller and action.
Some of the options above are being passed to wkhtmltopdf
binary. They can be used to control the options used in Webkit rendering before generating the PDF.
Examples of those options are:
print_media_type: true # Passes `--print-media-type`
no_background: true # Passes `--no-background`
You can see the complete list of options under "Global Options" in wkhtmltopdf usage docs.
If you need to just create a pdf and not display it:
# create a pdf from a string
pdf = WickedPdf.new.pdf_from_string('<h1>Hello There!</h1>')
# create a pdf file from a html file without converting it to string
# Path must be absolute path
pdf = WickedPdf.new.pdf_from_html_file('/your/absolute/path/here')
# create a pdf from a URL
pdf = WickedPdf.new.pdf_from_url('https://github.com/mileszs/wicked_pdf')
# create a pdf from string using templates, layouts and content option for header or footer
pdf = WickedPdf.new.pdf_from_string(
render_to_string('templates/pdf', layout: 'pdfs/layout_pdf.html'),
footer: {
content: render_to_string(
'templates/footer',
layout: 'pdfs/layout_pdf.html'
)
}
)
# It is possible to use footer/header templates without a layout, in that case you need to provide a valid HTML document
pdf = WickedPdf.new.pdf_from_string(
render_to_string('templates/full_pdf_template'),
header: {
content: render_to_string('templates/full_header_template')
}
)
# or from your controller, using views & templates and all wicked_pdf options as normal
pdf = render_to_string pdf: "some_file_name", template: "templates/pdf", encoding: "UTF-8"
# then save to a file
save_path = Rails.root.join('pdfs','filename.pdf')
File.open(save_path, 'wb') do |file|
file << pdf
end
# you can also track progress on your PDF generation, such as when using it from within a Resque job
class PdfJob
def perform
blk = proc { |output|
match = output.match(/\[.+\] Page (?<current_page>\d+) of (?<total_pages>\d+)/)
if match
current_page = match[:current_page].to_i
total_pages = match[:total_pages].to_i
message = "Generated #{current_page} of #{total_pages} pages"
at current_page, total_pages, message
end
}
WickedPdf.new.pdf_from_string(html, progress: blk)
end
end
If you need to display utf encoded characters, add this to your pdf views or layouts:
<meta charset="utf-8" />
If you need to return a PDF in a controller with Rails in API mode:
pdf_html = ActionController::Base.new.render_to_string(template: 'controller_name/action_name', layout: 'pdf')
pdf = WickedPdf.new.pdf_from_string(pdf_html)
send_data pdf, filename: 'file.pdf'
You can control page breaks with CSS.
Add a few styles like this to your stylesheet or page:
div.alwaysbreak { page-break-before: always; }
div.nobreak:before { clear:both; }
div.nobreak { page-break-inside: avoid; }
A bit of javascript can help you number your pages. Create a template or header/footer file with this:
<html>
<head>
<script>
function number_pages() {
var vars={};
var x=document.location.search.substring(1).split('&');
for(var i in x) {var z=x[i].split('=',2);vars[z[0]] = decodeURIComponent(z[1]);}
var x=['frompage','topage','page','webpage','section','subsection','subsubsection'];
for(var i in x) {
var y = document.getElementsByClassName(x[i]);
for(var j=0; j<y.length; ++j) y[j].textContent = vars[x[i]];
}
}
</script>
</head>
<body onload="number_pages()">
Page <span class="page"></span> of <span class="topage"></span>
</body>
</html>
Anything with a class listed in "var x" above will be auto-filled at render time.
If you do not have explicit page breaks (and therefore do not have any "page" class), you can also use wkhtmltopdf's built in page number generation by setting one of the headers to "[page]":
render pdf: 'filename', header: { right: '[page] of [topage]' }
You can put your default configuration, applied to all pdf's at "wicked_pdf.rb" initializer.
If you would like to have WickedPdf automatically generate PDF views for all (or nearly all) pages by appending .pdf to the URL, add the following to your Rails app:
# in application.rb (Rails3) or environment.rb (Rails2)
require 'wicked_pdf'
config.middleware.use WickedPdf::Middleware
If you want to turn on or off the middleware for certain URLs, use the :only
or :except
conditions like so:
# conditions can be plain strings or regular expressions, and you can supply only one or an array
config.middleware.use WickedPdf::Middleware, {}, only: '/invoice'
config.middleware.use WickedPdf::Middleware, {}, except: [ %r[^/admin], '/secret', %r[^/people/\d] ]
If you use the standard render pdf: 'some_pdf'
in your app, you will want to exclude those actions from the middleware.
To include a rendered pdf file in an email you can do the following:
attachments['attachment.pdf'] = WickedPdf.new.pdf_from_string(
render_to_string('link_to_view.pdf.erb', layout: 'pdf')
)
This will render the pdf to a string and include it in the email. This is very slow so make sure you schedule your email delivery in a job.
Mike Ackerman's post How To Create PDFs in Rails
Andreas Happe's post Generating PDFs from Ruby on Rails
JESii's post WickedPDF, wkhtmltopdf, and Heroku...a tricky combination
Berislav Babic's post Send PDF attachments from Rails with WickedPdf and ActionMailer
StackOverflow questions with the tag "wicked-pdf"
Now you can use a debug param on the URL that shows you the content of the pdf in plain html to design it faster.
First of all you must configure the render parameter show_as_html: params.key?('debug')
and then just use it like you normally would but add "debug" as a GET param in the URL:
http://localhost:3001/CONTROLLER/X.pdf?debug
However, the wicked_pdf_* helpers will use file:/// paths for assets when using :show_as_html, and your browser's cross-domain safety feature will kick in, and not render them. To get around this, you can load your assets like so in your templates:
<%= params.key?('debug') ? image_tag('foo') : wicked_pdf_image_tag('foo') %>
If one image from your HTML cannot be found (relative or wrong path for example), others images with right paths may not be displayed in the output PDF as well (it seems to be an issue with wkhtmltopdf).
wkhtmltopdf may render at different resolutions on different platforms. For example, Linux prints at 75 dpi (native for WebKit) while on Windows it's at the desktop's DPI (which is normally 96 dpi). Use :zoom => 0.78125
(75/96) to match Linux rendering to Windows.
You may have noticed: this plugin is heavily inspired by the PrinceXML plugin princely. PrinceXML's cost was prohibitive for me. So, with a little help from some friends (thanks jqr), I tracked down wkhtmltopdf, and here we are.
- Fork it
- Create your feature branch (
git checkout -b my-new-feature
) - Run the test suite and check the output (
rake
) - Add tests for your feature or fix (please)
- Commit your changes (
git commit -am 'Add some feature'
) - Push to the branch (
git push origin my-new-feature
) - Create new Pull Request
Also, thanks to unixmonkey, galdomedia, jcrisp, lleirborras, tiennou, and everyone else for all their hard work and patience with my delays in merging in their enhancements.