Scaml is a very DRY way of writing XHTML templates. It is Scala version of Haml. Scaml functions as a replacement for inline page templating systems such as PHP, ERB, and ASP. However, Scaml avoids the need for explicitly coding XHTML into the template, because it uses a very concise white space active XHMLT notation.
Scaml supports both the original Haml notation and a newer Jade notation. The Jade dialect of Haml recognizes that Haml based notations are best used for rendering structured XHTML markup and that content is best rendered using something like markdown. It therefore, simplifies the element notation in exchange for complicating the plain content notation.
%html
%body
The quick brown fox jumps
over the lazy dog
html
body
| The quick brown fox jumps
| over the lazy dog
Both examples above will render to the following:
<html>
<body>
The quick brown fox jumps
over the lazy dog
</body>
</html>
The only difference between the Haml and Jade notation styles are that in the Jade notation style:
%
|
The rest of this document will assume you are using the Haml notation style. All the examples will work with the Jade notation provided you apply the differences listed above.
A substantial portion of any HTML document is its content, which is plain old text. Any Scaml line that's not interpreted as something else is taken to be plain text, and passed through unmodified.
For example:
%gee
%whiz
Wow this is cool!
<gee>
<whiz>
Wow this is cool!
</whiz>
</gee>
Note that HTML tags are passed through unmodified as well. If you have some HTML you don't want to convert to Scaml, or you're converting a file line-by-line, you can just include it as-is.
%p
<div id="blah">Blah!</div>
<p>
<div id="blah">Blah!</div>
</p>
\
The backslash character escapes the first character of a line, allowing use of otherwise interpreted characters as plain text.
%title
= title
\= title
<title>
MyPage
= title
</title>
%
The percent character is placed at the beginning of a line.
It's followed immediately by the name of an element,
then optionally by modifiers (see below), a space,
and text to be rendered inside the element.
It creates an element in the form of <element></element>
.
%one
%two
%three Hey there
<one>
<two>
<three>Hey there</three>
</two>
</one>
Any string is a valid element name; Scaml will automatically generate opening and closing tags for any element.
{
}
or (
)
Brackets represent a Scala Map that is used for specifying the attributes of an element. Ruby hash syntax is used instead of Scala syntax to preserve a higher level of compatibility with the original Haml implementation. It is translated and evaluated as a Scala Map, so logic will work in it and local variables may be used. Quote characters within the attribute will be replaced by appropriate escape sequences. The hash is placed after the tag is defined.
%html{:xmlns => "http://www.w3.org/1999/xhtml", "xml:lang" => "en", :lang => "en"}
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en"></html>
Attribute hashes can also be stretched out over multiple lines to accommodate many attributes. However, newlines may only be placed immediately after commas.
%script{:type => "text/javascript",
:src => "javascripts/script"}
<script type="text/javascript" src="javascripts/script"/>
Complex expression are supported if you wrap them between the {
and }
characters.
%li{:counter={3+4}} Stuff
<li counter="7">Stuff</li>
()
Scaml also supports a terser, less Scala-specific attribute syntax based on HTML's attributes. These are used with parentheses instead of brackets, like so:
%html(xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en")
Scala variables can be used by omitting the quotes. For example:
%a(title=title href=href) Stuff
This is the same as:
%a{:title =>title, :href => href} Stuff
Complex expression are supported if you wrap them between the {
and }
characters.
%li(counter={3+4}) Stuff
<li counter="7">Stuff</li>
You can use both syntaxes together:
%a(title="Hello"){:href => "http://scalate.github.io/scalate"} Stuff
You can also use \#{}
interpolation to insert complicated expressions
in a HTML-style attribute:
%span(class="widget_#{widget.number}")
HTML-style attributes can be stretched across multiple lines just like hash-style attributes:
%script(type="text/javascript"
src="javascripts/script")
Some attributes, such as “checked” for input
tags or “selected” for option
tags,
are “boolean” in the sense that their values don't matter -
it only matters whether or not they're present.
In HTML (but not XHTML), these attributes can be written as
<input selected>
To do this in Scaml using hash-style attributes, just assign a Scala
true
value to the attribute:
%input{:selected => true}
In XHTML, the only valid value for these attributes is the name of the attribute. Thus this will render in XHTML as
<input selected="selected"/>
To set these attributes to false, simply assign them to a Scala false value. In both XHTML and HTML
%input{:selected => false}
will just render as
<input/>
HTML-style boolean attributes can be written just like HTML:
%input(selected)
or using true
and false
:
%input(selected=true)
.
and #
The period and pound sign are borrowed from CSS.
They are used as shortcuts to specify the class
and id
attributes of an element, respectively.
Multiple class names can be specified in a similar way to CSS,
by chaining the class names together with periods.
They are placed immediately after the tag and before an attributes hash.
%div#things
%span#rice Chicken Fried
%p.beans{ :food => "true" } The magical fruit
%h1.class.otherclass#id La La La
<div id="things">
<span id="rice">Chicken Fried</span>
<p class="beans" food="true">The magical fruit</p>
<h1 id="id" class="class otherclass">La La La</h1>
</div>
And,
#content
.articles
.article.title Doogie Howser Comes Out
.article.date 2006-11-05
.article.entry
Neil Patrick Harris would like to dispel any rumors that he is straight
<div id="content">
<div class="articles">
<div class="article title">Doogie Howser Comes Out</div>
<div class="article date">2006-11-05</div>
<div class="article entry">
Neil Patrick Harris would like to dispel any rumors that he is straight
</div>
</div>
</div>
'
element'
Sometimes you have to generate markup with weird element names. Element names
like <funny.element/>
. Since Scaml interprets the period as a class name for the
element, the following example:
%funny.element
<funny class="element"/>
does not give you the desired result of <funny.element/>
. In these cases you must single
quote the element name.
Example:
%'funny.element'
<funny.element/>
Because divs are used so often, they're the default elements.
If you only define a class and/or id using .
or #
,
a div is automatically used.
#collection
.item
.description What a cool item!
%div#collection
%div.item
%div.description What a cool item!
and is rendered to:
<div id="collection">
<div class="item">
<div class="description">What a cool item!</div>
</div>
</div>
/
The forward slash character, when placed at the end of a tag definition, causes the tag to be self-closed. For example:
%br/
%meta{"http-equiv" => "Content-Type", :content => "text/html"}/
<br/>
<meta http-equiv="Content-Type" content="text/html"/>
Some tags are automatically closed, as long as they have no content.
meta
, img
, link
, script
, br
, and hr
tags are closed by default.
This list can be customized by setting the ScamlOptions.autoclose
option.
%br
%meta{"http-equiv" => "Content-Type", :content => "text/html"}
<br/>
<meta http-equiv="Content-Type" content="text/html"/>
>
and <
>
and <
give you more control over the whitespace near a tag.
>
will remove all whitespace surrounding a tag,
while <
will remove all whitespace immediately within a tag.
You can think of them as alligators eating the whitespace:
>
faces out of the tag and eats the whitespace on the outside,
and <
faces into the tag and eats the whitespace on the inside.
They're placed at the end of a tag definition,
after class, id, and attribute declarations
but before /
or =
.
%blockquote<
%div
Foo!
<blockquote><div>
Foo!
</div></blockquote>
And:
%img
%img>
%img
<img/><img/><img/>
And:
%p<= "Foo\nBar"
<p>Foo
Bar</p>
And finally:
%img
%pre><
foo
bar
%img
<img /><pre>foo
bar</pre><img />
!!! format
When describing HTML documents with Scaml,
you can have a document type or XML prolog generated automatically
by including the characters !!!
.
!!! XML
!!!
%html
%head
%title Myspace
%body
%h1 I am the international space station
%p Sign my guestbook
<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html>
<head>
<title>Myspace</title>
</head>
<body>
<h1>I am the international space station</h1>
<p>Sign my guestbook</p>
</body>
</html>
You can also specify the specific doctype after the !!!
When the format
is set to :xhtml
(the default),
the following doctypes are supported:
!!!
XHTML 1.0 Transitional
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
!!! Strict
XHTML 1.0 Strict
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
!!! Frameset
XHTML 1.0 Frameset
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Frameset//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-frameset.dtd">
!!! 1.1
XHTML 1.1
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
!!! Basic
XHTML Basic 1.1
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML Basic 1.1//EN" "http://www.w3.org/TR/xhtml-basic/xhtml-basic11.dtd">
!!! Mobile
XHTML Mobile 1.2
<!DOCTYPE html PUBLIC "-//WAPFORUM//DTD XHTML Mobile 1.2//EN" "http://www.openmobilealliance.org/tech/DTD/xhtml-mobile12.dtd">
When the format
option is set to :html4
,
the following doctypes are supported:
!!!
HTML 4.01 Transitional
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">
!!! Strict
HTML 4.01 Strict
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
!!! Frameset
HTML 4.01 Frameset
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Frameset//EN" "http://www.w3.org/TR/html4/frameset.dtd">
!!! 5
HTML 5
<!DOCTYPE html>
When the format
option is set to :html5
,
!!!
is always <!DOCTYPE html>
.
If you're not using the UTF-8 character set for your document, you can specify which encoding should appear in the XML prolog in a similar way.
!!! XML iso-8859-1
<?xml version="1.0" encoding="iso-8859-1" ?>
Scaml supports two sorts of comments: those that show up in the HTML output and those that don't.
/
The forward slash character, when placed at the beginning of a line, wraps all text after it in an HTML comment.
%peanutbutterjelly
/ This is the comment
I like sandwiches!
<peanutbutterjelly>
<!-- This is the comment -->
I like sandwiches!
</peanutbutterjelly>
The forward slash can also wrap indented sections of code. For example:
/
%p This doesn't render...
%div
%h1 Because it's commented out!
<!--
<p>This doesn't render...</p>
<div>
<h1>Because it's commented out!</h1>
</div>
-->
/[]
You can also use Internet Explorer conditional comments
by enclosing the condition in square brackets after the /
.
/[if IE]
%a{ :href => "http://www.mozilla.com/en-US/firefox/" }
%h1 Get Firefox
<!--[if IE]>
<a href="http://www.mozilla.com/en-US/firefox/">
<h1>Get Firefox</h1>
</a>
<![endif]-->
-#
The hyphen followed immediately by the pound sign signifies a silent comment. Any text following this isn't rendered in the resulting document at all.
For example:
%p foo
-# This is a comment
%p bar
<p>foo</p>
<p>bar</p>
You can also nest text beneath a silent comment. None of this text will be rendered.
%p foo
-#
This won't be displayed
Nor will this
%p bar
<p>foo</p>
<p>bar</p>
-@
When a Scalate template is rendered, the caller can pass an attribute map which the template in charge of rendering. To bind the attribute to a Scala variable, a Scaml template uses the hyphen character followed by an at sign and then a Scala variable declaration statement.
For example To define an attribute use the following declaration
-@ val foo: MyType
If the attribute map does not contain a “foo” entry, then a NoValueSetException is thrown when the the template is rendered.
To avoid this exception, a default value can be configured. For example:
-@ val bar: String = "this is the default value"
The attribute is now available for use as an expression.
Its very common to have a template based on a single object who's members are f frequently accessed. In this cases, it's convenient to import all the object's members. This can be done by adding the import keyword to the attribute declaration.
For example:
-@ import val model: Person
%p Hello #{name}, what is the weather like in #{city}
is the same as:
-@ val model: Person
- import model._
%p Hello #{name}, what is the weather like in #{city}
Which is the same as:
-@ val model: Person
%p Hello #{model.name}, what is the weather like in #{model.city}
=
The equals character is followed by Scala code. This code is evaluated and the output is inserted into the document.
%p
= List("hi", "there", "reader!").mkString(" ")
= "yo"
<p>
hi there reader!
yo
</p>
The default setting for the TemplateEngine.escapeMarkup
option is
true. When TemplateEngine.escapeMarkup
is enabled, =
will sanitize any
HTML-sensitive characters generated by the script.
= """<script>alert("I'm evil!");</script>"""
<script>alert("I'm evil!");</script>
=
can also be used at the end of a tag to insert Scala code within that tag.
%p= "hello"
<p>hello</p>
-
The hyphen character is also followed by Scala code. This code is evaluated but not inserted into the document.
**It is not recommended that you use this widely; almost all processing code and logic should be restricted to the Controller, the Helper, or partials.**
- var foo = "hello"
- foo += " there"
- foo += " you!"
%p= foo
<p>hello there you!</p>
Or alternatively, if you have a large block of Scala code, you can nest it under the hyphen character as demonstrated by the following example:
-
var foo = "hello"
foo += " there"
foo += " you!"
%p= foo
<p>hello there you!</p>
Scala blocks, like XHTML tags, don't need to be explicitly closed in Scaml. Rather, they're automatically closed, based on indentation. A block begins whenever the indentation is increased after a Scala insertion or evaluation command. It ends when the indentation decreases.
- for(i <- 42 to 46)
%p= i
%p See, I can count!
<p>42</p>
<p>43</p>
<p>44</p>
<p>45</p>
<p>46</p>
<p>See, I can count!</p>
And,
%p
- 2 match
- case 1 =>
= "one"
- case 2 =>
= "two"
- case 3 =>
= "three"
<p>
two
</p>
When inserting evaluated statements, it can also take advantage of Scala blocks. It can be handy for passing partial functions.
For example:
%p
= List(1,2,3).foldLeft("result: ")
- (a,x)=>
- a+x
is the same as:
%p
= List(1,2,3).foldLeft("result: ") { (a,x)=> { a+x } }
would be rendered to:
<p>
result: 123
</p>
~
~
works just like =
, except that it preserves the white space
formating on its input.
Scaml always produces HTML source which is easy to read since it properly indented. Even dynamically generated output is properly indented.
%html
%p
= "line1\nline2\nline3"
<html>
<p>
line1
line2
line3
</p>
</html>
Sometimes you don't want Scaml to indent the dynamically generated content.
For example, tags like pre
and textarea
are whitespace-sensitive;
indenting the text makes them render wrong.
When you use ~
instead of =
,
Scaml will convert newlines to the XHTML newline escape code,
and avoid
adding spaces for indentation.
%html
%pre
~ "line1\nline2\nline3"
<html>
<pre>
line1
line2
line3
</pre>
</html>
~~
Sometimes, you don't want Scaml to indent or apply the whitespace transformation on
the evaluated expression. When this is the case, use ~~
to use ugly whitespace
preservation. We call it ugly because the produce HTML will not properly indented.
%html
%p
~~ "line1\nline2\nline3"
<html>
<p>
line1
line2
line3
</p>
</html>
#{}
Scala code can be interpolated within plain text using #{}
.
%p This is #{quality} cake!
%p= "This is the "+(quality)+" cake!"
and renders to
<p>This is scrumptious cake!</p>
Backslashes can be used to escape #{
strings,
but they don't act as escapes anywhere else in the string.
%p
A slash make a difference here: \#{name} is set to: \\#{name}
But is ignored for: \# or \\
<p>
A slash make a difference here: #{name} is set to: \Hiram
But is ignored for: \# or \\
</p>
&=
An ampersand followed by one or two equals characters evaluates Scala code just like the equals without the ampersand, but sanitizes any HTML-sensitive characters in the result of the code.
&= "I like cheese & crackers"
I like cheese & crackers
When the TemplateEngine.escapeMarkup
option is set
to true, =
behaves identically to &=
.
&
can also be used on its own so that #{}
interpolation is escaped.
& I like #{"cheese & crackers"}
I like cheese & crackers
!=
An exclamation mark followed by one or two equals characters evaluates Scala code just like the equals would, but never sanitizes the HTML.
When the TemplateEngine.escapeMarkup
option is set to false, =
behaves identically to !=
.
However, if the TemplateEngine.escapeMarkup
option is set to true, =
will sanitize the HTML, but !=
still won't.
For example, if TemplateEngine.escapeMarkup
is true:
= "I feel <strong>!"
!= "I feel <strong>!"
I feel <strong>!
I feel <strong>!
!
can also be used on its own so that #{}
interpolation is unescaped.
! I feel #{"<strong>"}!
I feel <strong>!
:
The colon character designates a filter. This allows you to pass an indented block of text as input to another filtering program and add the result to the output of Haml.
The syntax is a colon followed by an optional list of filter flags and then a colon separated list of filter names.
%p
:markdown
Markdown
========
Hello, *World*
<p>
<h1>Markdown</h1>
<p>Hello, <em>World</em></p>
</p>
If you use the !
or &
filter flags, you can have Scala code
interpolated with #{}
expressions. It is invalid to use both
the !
and &
flags at the same time.
The &
flag enables sanitized interpolations.
- var flavor = "<raspberry/>"
#content
:&markdown
I *really* prefer #{flavor} jam.
<div id="content">
<p>I <em>really</em> prefer <raspberry/> jam.</p>
</div>
The !
flag enables non-sanitized interpolations.
- var flavor = "<raspberry/>"
#content
:!markdown
I *really* prefer #{flavor} jam.
<div id="content">
<p>I <em>really</em> prefer <raspberry/>; jam.</p>
</div>
The ~
filter flag enables preserves the white space of the content.
The indent level is left unchanged and newlines are converted to
%html
%p<
:~plain
Indentation levels are not enforced in filters.
#{Interpolation} is disabled by default
Last line
<html>
<p> Indentation levels are not enforced in filters.
 #{Interpolation} is disabled by default
Last line</p>
</html>
More than one filter can be be used by separating each filter name with a colon. When this is done, the filters are chained together so that the output of filter on right, is passed as input to the filter on the left.
%pre
:escaped :javascript
alert("Hello");
<pre>
<script type='text/javascript'>
//<![CDATA[
alert("Hello");
//]]>
</script>
</pre>
Scaml has the following filters defined:
:plain
Does not parse the filtered text. This is useful for large blocks of text or HTML. Really handy when when you don't want lines starting with
.
or-
to be parsed.
:javascript
Surrounds the filtered text with
<script>
and CDATA tags. Useful for including inline Javascript.
:css
Surrounds the filtered text with
<style>
and CDATA tags. Useful for including inline CSS.
{#cdata-filter}
:cdata
Surrounds the filtered text with CDATA tags.
:escaped
Works the same as plain, but HTML-escapes the text before placing it in the document.
:sass
Parses the filtered text with Sass to produce CSS output.
Only works if you have thescalate-jruby
module on the class path. You normally want to combine with the:css
filter. For example:css:sass
:scss
Parses the filtered text with Scss to produce CSS output. Only works if you have the
scalate-jruby
module on the class path. You normally want to combine with the:css
filter. For example:css:scss
:textile
Parses the filtered text with Textile. Only works the scalate-wikitext module is found on the class path.
:markdown
Parses the filtered text with Markdown. Only works if scalamd or MarkdownJ is found on the class path.
There are several global options you can configure to customize how Scaml renders the output. You will need to configure these before any of your scaml templates are compiled as they affect the generated scala template classes.
ScamlOptions.indent
The ScamlOptions.indent
option is used to control what kind of indenting characters to
use in the rendered markup. It defaults to two spaces but can be set to the tab character
or set to the empty string to disable indenting alltogether.
%gee
%whiz
Wow this is cool!
<gee>
<whiz>
Wow this is cool!
</whiz>
</gee>
ScamlOptions.nl
The ScamlOptions.nl
option is used to control what kind of new line seperator to
use in the rendered markup. It defaults to \n
. Some folks may want to set it and the indent
to the empty string to reduce the generated document sizes.
For example, if ScamlOptions.indent = ""
and ScamlOptions.nl = ""
then:
%gee
%whiz
Wow this is cool!
<gee><whiz>Wow this is cool!</whiz></gee>
ScamlOptions.ugly
Enabling the ScamlOptions.ugly
option makes =
statements work like ~~
statements. The dynamic expressions
will not be indented and they the new line preservation transformation method will not be applied. Enabling the
ugly option will significantly reduce the CPU overhead of processing dynamic expressions.
%html
%p
= "line1\nline2\nline3"
<html>
<p>
line1
line2
line3
</p>
</html>