raw vs. html_safe vs. h to unescape html

Considering Rails 3:

html_safe actually "sets the string" as HTML Safe (it's a little more complicated than that, but it's basically it). This way, you can return HTML Safe strings from helpers or models at will.

h can only be used from within a controller or view, since it's from a helper. It will force the output to be escaped. It's not really deprecated, but you most likely won't use it anymore: the only usage is to "revert" an html_safe declaration, pretty unusual.

Prepending your expression with raw is actually equivalent to calling to_s chained with html_safe on it, but is declared on a helper, just like h, so it can only be used on controllers and views.

"SafeBuffers and Rails 3.0" is a nice explanation on how the SafeBuffers (the class that does the html_safe magic) work.

I think it bears repeating: html_safe does not HTML-escape your string. In fact, it will prevent your string from being escaped.

<%= "<script>alert('Hello!')</script>" %>

will put:

<script>alert(&#x27;Hello!&#x27;)</script>

into your HTML source (yay, so safe!), while:

<%= "<script>alert('Hello!')</script>".html_safe %>

will pop up the alert dialog (are you sure that's what you want?). So you probably don't want to call html_safe on any user-entered strings.

The difference is between Rails’ html_safe() and raw(). There is an excellent post by Yehuda Katz on this, and it really boils down to this:

def raw(stringish)  stringish.to_s.html_safeend

Yes, raw() is a wrapper around html_safe() that forces the input to String and then calls html_safe() on it. It’s also the case that raw() is a helper in a module whereas html_safe() is a method on the String class which makes a new ActiveSupport::SafeBuffer instance — that has a @dirty flag in it.

Refer to "Rails’ html_safe vs. raw".