A Ruby gem to transform HTML into PDFs, PNGs or JPEGs using Google Puppeteer and Chromium.
Add this line to your application's Gemfile:
gem 'grover'npm install puppeteer# Grover.new accepts a URL or inline HTML and optional parameters for Puppeteer
grover = Grover.new('https://google.com', format: 'A4')
# Get an inline PDF
pdf = grover.to_pdf
# Get a screenshot
png = grover.to_png
jpeg = grover.to_jpeg
# Get the HTML content (including DOCTYPE)
html = grover.to_html
# Options can be provided through meta tags
Grover.new('<html><head><meta name="grover-page_ranges" content="1-3"')
Grover.new('<html><head><meta name="grover-margin-top" content="10px"')N.B.
- options are underscore case, and sub-options separated with a dash
- all options can be overwritten, including
emulate_mediaanddisplay_url
It's easy to render a normal Rails view template as a PDF, using Rails' render_to_string:
html = MyController.new.render_to_string({
template: 'controller/view',
layout: 'my_layout',
locals: { :@instance_var => ... }
})
pdf = Grover.new(html, **grover_options).to_pdfIf calling Grover directly (not through middleware) you will need to either specify a display_url or modify your
HTML by converting any relative paths to absolute paths before passing to Grover.
This can be achieved using the HTML pre-processor helper (pay attention to the slash at the end of the url):
absolute_html = Grover::HTMLPreprocessor.process relative_html, 'http://my.server/', 'http'This is important because Chromium will try and resolve any relative paths via the display url host. If not provided,
the display URL defaults to http://example.com.
There are many scenarios where specifying a different host of relative paths would be preferred. For example, your server might be behind a NAT gateway and the display URL in front of it. The display URL might be shown in the header/footer, and as such shouldn't expose details of your private network.
If you run into trouble, take a look at the debugging section below which would allow you to inspect the page content and devtools.
Grover can be configured to adjust the layout of the resulting PDF/image.
For available PDF options, see https://github.com/puppeteer/puppeteer/blob/main/docs/api/puppeteer.pdfoptions.md
Also available are the emulate_media, cache, viewport, timeout, requestTimeout, convertTimeout
and launch_args options.
# config/initializers/grover.rb
Grover.configure do |config|
config.options = {
format: 'A4',
margin: {
top: '5px',
bottom: '10cm'
},
user_agent: 'Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:47.0) Gecko/20100101 Firefox/47.0',
viewport: {
width: 640,
height: 480
},
prefer_css_page_size: true,
emulate_media: 'screen',
bypass_csp: true,
media_features: [{ name: 'prefers-color-scheme', value: 'dark' }],
timezone: 'Australia/Sydney',
vision_deficiency: 'deuteranopia',
extra_http_headers: { 'Accept-Language': 'en-US' },
geolocation: { latitude: 59.95, longitude: 30.31667 },
focus: '#some-element',
hover: '#another-element',
cache: false,
timeout: 0, # Timeout in ms. A value of `0` means 'no timeout'
request_timeout: 1000, # Timeout when fetching the content (overloads the `timeout` option)
convert_timeout: 2000, # Timeout when converting the content (overloads the `timeout` option, only applies to PDF conversion)
launch_args: ['--font-render-hinting=medium'],
wait_until: 'domcontentloaded'
}
endFor available PNG/JPEG options, see https://github.com/puppeteer/puppeteer/blob/main/docs/api/puppeteer.page.screenshot.md#remarks
Note that by default the full_page option is set to false and you will get a 800x600 image. You can either specify
the image size using the clip options, or capture the entire page with full_page set to true.
For viewport options, see https://github.com/puppeteer/puppeteer/blob/main/docs/api/puppeteer.page.setviewport.md#remarks
For launch_args options, see http://peter.sh/experiments/chromium-command-line-switches/
Launch parameter args can also be provided using a meta tag:
For timezone IDs see ICUs metaZones.txt.
Passing nil disables timezone emulation.
The vision_deficiency option can be passed one of achromatopsia, deuteranopia, protanopia, tritanopia,
blurredVision or none.
The focus option takes a CSS selector and will focus on the first matching element after rendering is complete
(including waiting for the specified wait_for_selector).
The hover option takes a CSS selector and will hover on the first matching element after rendering is complete
(including waiting for the specified wait_for_selector).
<meta name="grover-launch_args" content="['--disable-speech-api']" />For wait_until option, default for URLs is networkidle2 and for HTML content networkidle0.
For available options see https://github.com/puppeteer/puppeteer/blob/main/docs/api/puppeteer.page.goto.md#remarks
The wait_for_selector option can also be used to wait until an element appears on the page. Additional waiting parameters can be set with the wait_for_selector_options options hash. For available options, see: https://github.com/puppeteer/puppeteer/blob/main/docs/api/puppeteer.page.waitforselector.md#remarks.
The wait_for_function option can be used to wait until a specific function returns a truthy value. Additional parameters can be set with the wait_for_function_options options hash. For available options, see: https://github.com/puppeteer/puppeteer/blob/main/docs/api/puppeteer.page.waitforfunction.md#remarks
The wait_for_timeout option can also be used to wait the specified number of milliseconds have elapsed.
The raise_on_request_failure option, when enabled, will raise a Grover::JavaScript::RequestFailedError
if the initial content request or any subsequent asset request returns a bad response or times out.
The Chrome/Chromium executable path can be overridden with the executable_path option.
Supplementary JavaScript can be executed on the page (after render and before conversion to PDF/image)
by passing it to the execute_script option.
Grover.new(<some url>, { execute_script: 'document.getElementsByTagName("footer")[0].innerText = "Hey"' }).to_pdfFor requesting a page with basic authentication, username and password options can be provided. Note that this
only really makes sense if you're calling Grover directly (and not via middleware).
Grover.new('<some URI with basic authentication', username: 'the username', password: 'super secret').to_pdfBy default, Grover launches a local Chromium instance. You can connect to a remote/external
Chromium with the browser_ws_endpoint options.
For example, to connect to a chrome instance started with docker using docker run -p 3000:3000 ghcr.io/browserless/chrome:latest:
options = {"browser_ws_endpoint": "ws://localhost:3000/chrome"}
grover = Grover.new("https://mysite.com/path/to/thing", options)
File.open("grover.png", "wb") { |f| f << grover.to_png }You can also pass launch flags like this: ws://localhost:3000/chrome?--disable-speech-api
If you are only using remote chromium, you can install the puppeteer-core node package instead of puppeteer to avoid downloading chrome.
Grover will use puppeteer or fallback to puppeteer-core if it is available.
npm install puppeteer-coreTo set request cookies when requesting a URL, pass an array of hashes as such
N.B. Only the name and value properties are required.
See page.setCookie documentation for more details (old documentation with more detailed description is available here).
myCookies = [
{ name: 'sign_username', value: '[email protected]', domain: 'mydomain' },
{ name: '_session_id', value: '9c014df0b699d8dc08d1c472f8cc594c', domain: 'mydomain' }
]
Grover.new('<some URI with cookies', cookies: myCookies).to_pdfIf you need to forward the cookies from the original request, you could extract them as such:
def header_cookies
request.headers['Cookie'].split('; ').map do |cookie|
key, value = cookie.split '='
{ name: key, value: value, domain: request.headers['Host'] }
end
endAnd give that array to Grover:
Grover.new('<some URI with cookies', cookies: header_cookies).to_pdfTo add style tags, pass an array of style tag options as such See page.addStyleTag documentation for more details (old documentation with more detailed description is available here).
style_tag_options = [
{ url: 'http://example.com/style.css' },
{ path: 'style.css' },
{ content: '.body{background: red}' }
]
Grover.new('<html><body><h1>Heading</h1></body></html>', style_tag_options: style_tag_options).to_pdfTo add script tags, pass an array of script tag options as such See documentation for more details page.addScriptTag (old documentation is available here).
script_tag_options = [
{ url: 'http://example.com/script.js' },
{ path: 'script.js' },
{ content: 'document.querySelector("h1").style.display = "none"' }
]
Grover.new('<html><body><h1>Heading</h1></body></html>', script_tag_options: script_tag_options).to_pdfIf you want to have the header or footer display the page URL, Grover requires that this is passed through via the
display_url option. This is because the page URL is not available in the raw HTML!
For Rack middleware conversions, the original request URL (without the .pdf extension) will be passed through and
assigned to display_url for you. You can of course override this by using a meta tag in the downstream HTML response.
For raw HTML conversions, if the display_url is not provided http://example.com will be used as the default.
Should be valid HTML markup with following classes used to inject printing values into them:
dateformatted print datetitledocument titleurldocument locationpageNumbercurrent page numbertotalPagestotal pages in the document
In respective controller's action use:
respond_to do |format|
format.html do
response.headers['Content-Disposition'] = %(attachment; filename="lorem_ipsum.pdf")
render layout: 'pdf'
end
endThe node_env_vars configuration option enables you to set custom environment variables for the spawned node process. For example you might need to disable jemalloc in some environments (#80).
# config/initializers/grover.rb
Grover.configure do |config|
config.node_env_vars = { "LD_PRELOAD" => "" }
endGrover comes with a middleware that allows users to get a PDF, PNG or JPEG view of any page on your site by appending .pdf, .png or .jpeg/.jpg to the URL.
Non-Rails Rack apps
# in config.ru
require 'grover'
use Grover::MiddlewareRails apps
# in application.rb
require 'grover'
config.middleware.use Grover::MiddlewareN.B. by default PNG and JPEG are not modified in the middleware to prevent breaking standard behaviours. To enable them, there are configuration options for each image type as well as an option to disable the PDF middleware (on by default).
If either of the image handling middleware options are enabled, the ignore_path and/or ignore_request should also be configured, otherwise assets are likely to be handled which would likely result in 404 responses.
# config/initializers/grover.rb
Grover.configure do |config|
config.use_png_middleware = true
config.use_jpeg_middleware = true
config.use_pdf_middleware = false
endThe root_url option can be specified either when configuring the middleware or as a global option. This is needed
when running the Grover middleware behind a URL rewriting proxy or within a containerised system.
As a middleware option:
# in application.rb
require 'grover'
config.middleware.use Grover::Middleware, root_url: 'https://my.external.domain'or as a global option:
# config/initializers/grover.rb
Grover.configure do |config|
config.root_url = 'https://my.external.domain'
endThe ignore_path configuration option can be used to tell Grover's middleware whether it should handle/modify
the response. There are three ways to set up the ignore_path:
- a
Stringwhich matches the start of the request path. - a
Regexpwhich could match any part of the request path. - a
Procwhich accepts the request path as a parameter.
# config/initializers/grover.rb
Grover.configure do |config|
# assigning a String
config.ignore_path = '/assets/'
# matches `www.example.com/assets/foo.png` and not `www.example.com/bar/assets/foo.png`
# assigning a Regexp
config.ignore_path = /my\/path/
# matches `www.example.com/foo/my/path/bar.png`
# assigning a Proc
config.ignore_path = ->(path) do
/\A\/foo\/.+\/[0-9]+\.png\z/.match path
end
# matches `www.example.com/foo/bar/123.png`
endThe ignore_request configuration option can be used to tell Grover's middleware whether it should handle/modify
the response. It should be set with a Proc which accepts the request (Rack::Request) as a parameter.
# config/initializers/grover.rb
Grover.configure do |config|
# assigning a Proc
config.ignore_request = ->(req) do
req.host == 'www.example.com'
end
# matches `www.example.com/foo/bar/123.png`
config.ignore_request = ->(req) do
req.has_header?('X-BLOCK')
end
# matches `HTTP Header X-BLOCK`
endThe allow_file_uris option can be used to render an html document from the file system.
This should be used with EXTREME CAUTION. If used improperly it could potentially be manipulated to reveal
sensitive files on the system. Do not enable if rendering content from outside entities
(user uploads, external URLs, etc).
It defaults to false preventing local system files from being read.
# config/initializers/grover.rb
Grover.configure do |config|
config.allow_file_uris = true
endAnd used as such:
# Grover.new accepts a file URI and optional parameters for Puppeteer
grover = Grover.new('file:///some/local/file.html', format: 'A4')
# Get an inline PDF of the local file
pdf = grover.to_pdfSince the header/footer for Puppeteer is configured globally, displaying of front/back cover pages (with potentially different headers/footers etc) is not possible.
To get around this, Grover's middleware allows you to specify relative paths for the cover page contents. For direct execution, you can make multiple calls and combine the resulting PDFs together.
You can specify relative paths to the cover page contents using the front_cover_path and back_cover_path
options either via the global configuration, or via meta tags. These paths (with query parameters) are then
requested from the downstream app.
The cover pages are converted to PDF in isolation, and then combined together with the original PDF response, before being returned back up through the Rack stack.
N.B To simplify things, the same request method and body are used for the cover page requests.
# config/initializers/grover.rb
Grover.configure do |config|
config.options = {
front_cover_path: '/some/global/cover/page?foo=bar'
}
endOr via the meta tags in the original response:
<html>
<head>
<meta name="grover-back_cover_path" content="/back/cover/page?bar=baz" />
</head>
...
</html>To add a cover page using direct execution, you can make multiple calls and combine the results using the
combine_pdf gem.
require 'combine_pdf'
# ...
def invoke(file_path)
pdf = CombinePDF.parse(Grover.new(pdf_report_url).to_pdf)
pdf >> CombinePDF.parse(Grover.new(pdf_front_cover_url).to_pdf)
pdf << CombinePDF.parse(Grover.new(pdf_back_cover_url).to_pdf)
pdf.save file_path
endTo run Grover (Puppeteer) on Heroku follow these steps:
-
Add the
nodebuildpack. Puppeteer requires a node environment to run.heroku buildpacks:add heroku/nodejs --index=1 [--remote yourappname] -
Add the
puppeteerbuildpack. Make sure thepuppeteerbuildpack runs after thenodebuildpack and before the mainrubybuildpack.heroku buildpacks:add jontewks/puppeteer --index=2 [--remote yourappname] -
Next, tell Grover to run Puppeteer in the "no-sandbox" mode by setting an ENV variable
GROVER_NO_SANDBOX=trueon your app dyno. Make sure that you trust all the HTML/JS you provide to Grover.heroku config:set GROVER_NO_SANDBOX=true [--remote yourappname] -
Finally, if using puppeteer 19+ (the default) add the following to a
.puppeteerrc.cjsfile in the root of your project:const {join} = require('path'); /** * @type {import("puppeteer").Configuration} * */ module.exports = { cacheDirectory: join(__dirname, '.cache', 'puppeteer'), };
If you're having trouble with converting the HTML content, you can enable some debugging options to help. These can be
enabled as global options via Grover.configure, by passing through to the Grover initializer, or using meta tag
options.
debug: {
headless: false, # Default true. When set to false, the Chromium browser will be displayed
devtools: true # Default false. When set to true, the browser devtools will be displayed.
}N.B.
- The headless option disabled is not compatible with exporting of the PDF.
- If showing the devtools, the browser will halt resulting in a navigation timeout
Bug reports and pull requests are welcome on GitHub at https://github.com/Studiosity/grover.
Note that spec tests are appreciated to minimise regressions. Before submitting a PR, please ensure that:
$ rspecand
$ rubocopboth succeed
To run tests tagged with remote_browser, you need to start a browser in a container:
docker run -p 3000:3000 browserless/chrome:latest
and run:
rspec --tag remote_browser
Thanks are given to the great work done in the PDFKit project. The middleware and HTML preprocessing components were used heavily in the implementation of Grover.
Thanks are also given to the excellent Schmooze project. The Ruby to NodeJS interface in Grover is heavily based off that work. Grover previously used that gem, however migrated away due to differing requirements over persistence/cleanup of the NodeJS worker process.
The gem is available as open source under the terms of the MIT License.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent