Extracting Ruby and HTML

Extraction pulls one language out of a template and blanks out the other. It is how a tool that only understands Ruby, such as RuboCop or a Ruby parser, gets something it can read, and how an HTML tool gets markup without ERB in the middle of it.

Extract the Ruby

Ruby

source = %(<p>Hello <%= user.name %></p>)

Herb.extract_ruby(source)
# => "             user.name  ;    "

JavaScript

const source = "<p>Hello <%= user.name %></p>"

Herb.extractRuby(source)
// => "             user.name  ;    "

Java

import org.herb.Herb;

String ruby = Herb.extractRuby("<p>Hello <%= user.name %></p>");
// "             user.name  ;    "

Rust

use herb::extract_ruby;

let ruby = extract_ruby("<p>Hello <%= user.name %></p>").unwrap();
// "             user.name  ;    "

By default every character keeps the position it had in the template, with the HTML replaced by spaces. A Ruby error on line 12 column 30 of the extracted output is on line 12 column 30 of the template, so positions need no translation.

Options

Option	Type	Default	Description
semicolons	Boolean	true	Add ; at the end of each ERB tag to separate statements
comments	Boolean	false	Include ERB comments (<%# %>) in the output
preserve_positions	Boolean	true	Maintain character positions by padding with whitespace

Ruby

Herb.extract_ruby(source, semicolons: false)

JavaScript

Herb.extractRuby(source, { semicolons: false })

Java

import org.herb.ExtractRubyOptions;

ExtractRubyOptions options = ExtractRubyOptions.create().semicolons(false);

Herb.extractRuby(source, options);

Rust

use herb::{extract_ruby_with_options, ExtractRubyOptions};

let options = ExtractRubyOptions { semicolons: false, ..Default::default() };

extract_ruby_with_options(source, &options).unwrap();

What each option changes, shown in Ruby and the same everywhere:

source = "<% x = 1 %> <% y = 2 %>"

Herb.extract_ruby(source)
# => "   x = 1  ;    y = 2  ;"

Herb.extract_ruby(source, semicolons: false)
# => "   x = 1       y = 2   "

Herb.extract_ruby("<%# comment %>\n<% code %>", comments: true)
# => "  # comment   \n   code  ;"

Herb.extract_ruby("<%# comment %><%= something %>", preserve_positions: false, comments: true)
# => "# comment \n something "

TIP

Use preserve_positions: false when you want readable Ruby. Keep the default when you need positions in the output to match positions in the template.

Extract the HTML

Ruby

Herb.extract_html(%(<p>Hello <%= user.name %></p>))
# => "<p>Hello                 </p>"

JavaScript

Herb.extractHTML("<p>Hello <%= user.name %></p>")
// => "<p>Hello                 </p>"

Java

String html = Herb.extractHTML("<p>Hello <%= user.name %></p>");
// "<p>Hello                 </p>"

Rust

use herb::extract_html;

let html = extract_html("<p>Hello <%= user.name %></p>").unwrap();
// "<p>Hello                 </p>"

The ERB is replaced by spaces, so the HTML keeps its positions the same way the Ruby does.

Next

Working with the tree is the other way to get at one language, by walking the nodes instead of blanking out text.