html_validation helps you validate, track, and fix or accept HTML validation issues without depending on an external web service. The idea is to take an HTML markup string associated with a particular resource (file or URL) and validate it locally. No web connection or dependency on a 3rd party web service is involved–just an install of HTML Tidy, which is supported by the W3C. An RSpec matcher is included.
HTML Validation is intended to be used in acceptance tests, test suites or rake tasks to alert you to changes in your html’s validity so you can fix them, or barring that, review and accept any errors and warnings.
The current (as of May 2013) default version of Tidy doesn’t support HTML 5. To get HTML 5 support, install this fork of tidy: github.com/w3c/tidy-html5
Download the repo as a zip file and extract somewhere, then follow the instructions given on the page (using sudo for make commands if on Linux). Add it to the PATH. Note that OS X and Debian like to install the old version of tidy in /usr/bin so if you don’t remove it or place the newer one earlier on the PATH, sadness may ensue.
On linux/OS X machines, confirm the right installation path using: which tidy
Add to spec/spec_helper.rb: require 'html_validation' include PageValidations Now, within your Integration Tests, you can do this: page.should have_valid_html or with the new syntax: expect(page).to have_valid_html It can be convenient to see the html printed in the RSpec failure messages. To set this feature (globally for all): HaveValidHTML.show_html_in_failures = true All Tidy flags, including using a config file can be handled. A shortcut to disable Tidy Warning messages (they are on by default) has been created for your convenience. However, some things that might be called "errors" show up as warnings, so the recommended approach is run, then accept the errors / warnings that are noted. At your peril, the shortcut is: HTMLValidation.show_warnings = false If you want to ignore specific errors, you can do the following : HTMLValidation.ignored_attribute_errors = ['tabindex'] HTMLValidation.ignored_tag_errors = ['inline'] HTMLValidation.ignored_errors = ['missing li']
require 'html_validation'
h = PageValidations::HTMLValidation.new
v = h.validation("<html>foo</html>", 'http://somesite.com/index.html')
# Note: The second argument (resource) just serves as an identifier used to
# name the results files, and no data is actually retrieved for you from on the URL!
v.valid?
-> false
v.exceptions
-> line 1 column 1 - Warning: missing <!DOCTYPE> declaration
-> line 1 column 9 - Warning: plain text isn't allowed in <head> elements
-> line 1 column 9 - Warning: inserting missing 'title' element
v.accept!
v.valid?
-> true
After the exceptions string has been accepted, the exceptions
are still available in v.exceptions even though valid? will return true.
This is so you can print them out, nag yourself, etc.
The exception string is treated and matched as a whole. The current
implementation does not allow partial acceptance of individual
results in a result string. However, line numbers are ignored, so there is
some flexibility in terms of being able to make changes without re-reviewing
what passes. Of course, if your HTML is totally clean, no worries.
The HTML Validation gem stores failure data per-resource. The results are stored in the data_folder. Typically this is the default folder: .validation in the project root. After a validation fails, run the thor task to accept or reject validation exceptions. From your project's root folder, run: html_validation review The result string is checked as a whole, so changes require re-review. Note, if you configure an alternate data_folder, you may pass it as an option to the Thor task.
This is untested on Windows as I don't have access to a Windows machine at the moment. It is written so it could (probably) work on Windows. Pull requests in this area would be welcomed.
HTML Tidy needs to be on the PATH. Make sure the right Tidy is being picked up with: which tidy
-
Check out the latest master to make sure the feature hasn’t been implemented or the bug hasn’t been fixed yet
-
Check out the issue tracker to make sure someone already hasn’t requested it and/or contributed it
-
Fork the project
-
Start a feature/bugfix branch
-
Commit and push until you are happy with your contribution
-
Make sure to add tests for it. This is important so I don’t break it in a future version unintentionally.
After tidy has been installed, in the gem folder, run: rspec spec
Copyright © 2012 Eric Beland. See LICENSE.txt for further details.