A tiny, secure JavaScript micro-templating script.
Tim lets you write simple templates that uses JavaScript's familiar dot notation. You pass in a JavaScript object that contains all the relevant strings, and they are then substituted into the template. For example:
tim("Hello {{place}}", {place: "world"});
// "Hello world"
- by Premasagar Rose (Dharmafly)
- source: github.com/premasagar/tim (MIT license)
- ~200 bytes minified & gzipped
Don't you just hate having to write HTML with a mess of string concatenation that clutters up your JavaScript?:
var myHTML = "<ul class='" + myClass + "'>" +
"<li id='" + theId + "'>" + liContents + "</li>" +
// etc, etc, etc
Yuck. There's no need to do this. Simply prepare a JavaScript object with the required properties, and inject it into a simple template string. The templates can all be tidily kept together with the rest of the markup in an HTML document (see below).
It is safe and secure: it doesn't use eval or (new Function), so it cannot execute malicious code. As such, it can be used in secure widgets and apps that disallow eval - e.g. Adobe Air sandboxes, AdSafe ads, etc.
It doesn't include a whole bloat load of features that are unlikely to get used when you just want to get some simple templating up and running.
It comes in two versions: "lite" (which is tiny), and "standard", which has advanced functionality and allows extensibility with plugins.
It's easy to debug.
For these reasons, it is now in use in Sqwidget, the JavaScript widget library: github.com/premasagar/sqwidget
There are two versions of Tim: the "standard" (full) version, and a stripped down "lite" version. The core functionality of both versions is identical, and is described below.
Tim can be used to replace tokens within a text string with specified data.
For example:
var template = "Hello {{place}}. My name is {{person.name}}.",
data = {
place: "Brighton",
person: {
name: "Prem"
}
};
tim(template, data);
// "Hello Brighton. My name is Prem."
In addition to plain and simple text, you can use Tim to populate HTML or other types of template.
For example:
var template = "<p><a href='{{url}}'>{{title}}</a></p>",
data = {
title: "Dharmafly",
url: "http://dharmafly.com"
};
tim(myTemplate, data);
// "<p><a href='http://dharmafly.com'>Dharmafly</a></p>"
...and so on, all the way up to a full-blown HTML document.
Sometimes, you will want to reuse the same template multiple times in a loop, and then wrapped within a bigger template - e.g. when creating an HTML <ul> list tag.
This is easily achieved:
var ul = "<ul>{{list}}</ul>",
li = "<li>{{contents}}</li>",
myList = "",
i;
for (i=100; i<103; i++){
myList += tim(li, {contents: i});
}
tim(ul, {list: myList});
// "<ul><li>100</li><li>101</li><li>102</li></ul>"
If your template references a path in the data object that could not actually be found, then Tim will throw an error, to help with debugging:
tim("Hello {{config.foo.bar}}", {config: {moo: "blah"}});
// tim: 'foo' not found in {{config.foo.bar}}
The data can be, or can include, an array. Use dot notation to access the array elements.
e.g:
tim("Hello {{0}}", ["world"]);
// "Hello world"
or:
tim("Hello {{places.0}}", {places: ["world"]});
// "Hello world"
Further examples are discussed in the section "iterating through arrays" below.
By default, template tags are delimited by "{{" and "}}" tokens.
To change this, edit the start and end vars in the code.
A little known aspect of browser parsing of HTML documents is that if a document contains a <script> tag that has a non-standard type attribute, then the browser will not attempt to parse that script element. However, JavaScript can still access the contents of the element. This leads us to a very useful coding pattern:
in the HTML, add a script tag with a non-standard type, e.g. "text/tim:
<script type="text/tim" id="foo">
<p><a href='{{url}}'>{{title}}</a></p>
</script>
and in the JavaScript, grab the script element by its id and extract its contents:
var template = document.getElementById("foo").innerHTML,
data = {
title: "Dharmafly",
url: "http://dharmafly.com"
};
tim(template, data);
// "<p><a href='http://dharmafly.com'>Dharmafly</a></p>"
then do something with your newly populated HTML, such as inject it into an element in the DOM:
document.getElementById("bar").innerHTML = myHTML;
Working in this way brings some sanity back to application development, where you want to specify the markup structure, but swap in specific text and attributes, or whole blocks of markup.
It also makes your code more maintainable - the markup templates can live with the rest of the static markup on a page, where the markup coder can access it, and the JavaScript logic lives in an entirely different place.
This time, using class names on the script element, and a little jQuery:
<!-- HTML -->
<script type="text/tim" class="foo">
<p><a href='{{url}}'>{{title}}</a></p>
</script>
/* JavaScript */
var template = jQuery("script[type=text/tim].foo").html(),
data = {
title: "Dharmafly",
url: "http://dharmafly.com"
};
jQuery("#bar").html(
tim(template, data)
);
E.g. by looping through all the template elements:
<!-- HTML -->
<script type="text/tim" class="foo">
<p><a href='{{url}}'>{{title}}</a></p>
</script>
/* JavaScript */
var templates = {}; // template cache object
jQuery("script[type=text/tim]").each(function(){
templates[this.className] = this.innerHTML;
});
tim(templates.foo, data);
The following methods are also available within the full version of Tim.
Templates can be cached and retrieved by Tim:
tim.templates("foo"); // get template named "foo"
tim.templates("foo", "bar"); // set template named "foo" to "bar"
tim.templates("foo", false); // delete template named "foo"
tim.templates({foo:"bar", blah:false}); // set/delete multiple templates
tim.templates(false); // delete all templates
An example use of this:
tim.templates("foo", "Cached templates are {{word}}"); // set template named "foo"
var foo = tim.templates("foo"); // get template named "foo"
tim(foo, {word:"useful"});
// Cached templates are useful
Caching templates is more efficient than parsing a template every time it is used.
When using Tim on the client-side, it automatically caches all templates it finds in the DOM that match the pattern <script type="text/tim" class="TEMPLATE_ID">. In this way, calling tim("TEMPLATE_ID", data) is a shortcut for templates embedded in the DOM.
It is possible to update Tim's default settings such as the template type, and start/end string delineators - {{ and }}
This allows you to parse micro-templates containing different types of data, such as:
<script type="text/tim+erb" id="myErbTest">
ERB is also <% adj %>, btw.
</script>
To parse this:
// Change the template's start/end delimiters and the target script element type
var erb = tim.parser({start:"<%", end:"%>", type:"text/tim+erb"});
erb("myErbTest", {adj:"lovely"});
// ERB is also lovely, btw.
Rather than targeting micro-templates via their class/id, or by their type attribute, it is also possible to specify the DOM attribute with which to target templates, such as data-*.
The following template:
<script type="text/tim" data-tim="myAttrTest">
data-* attributes are {{words.modifier}} {{words.adj}}.
</script>
Can be targeted by changing the target attribute:
tim.dom({attr:"data-tim"});
tim("myAttrTest", {words:{modifier:"really", adj:"useful"}});
// data-* attributes are really useful
Tim supports the use of boolean conditionals, to allow you to only output part of a template based on the value of a passed parameter.
<script type="text/tim" class="bool">
{{isWeekend}}
{{day}}
{{/isWeekend}}
</script>
The parameter {{day}} will only be output if isWeekend is true:
tim("bool", {isWeekend:true, day:"Sunday"}); // outputs: Sunday
tim("bool", {isWeekend:false, day:"Thursday"}); // no output
tim("bool", {isWeekend:true, day:"Saturday"}); // outputs: Saturday
Tim can be passed an array, which it will iterate through.
<script type="text/tim" class="loops">
Some fruit:
{{fruit}}
{{name}} are {{colour}},
{{/fruit}}
</script>
In this example, Tim will iterate through the array fruit, and add to its output each time:
tim("loops", {fruit:[
{ name:"apples", colour:"green" },
{ name:"oranges", colour:"orange" },
{ name:"bananas", colour:"yellow" }
]});
// apples are green, oranges are orange, bananas are yellow,
Tim can also be passed an object, through which it can iterate.
<script type="text/tim" class="objects">
Object iteration:
{{things}}
{{_key}} are {{_content}},
{{/things}}
</script>
In this example, Tim will iterate through the object things, using the keywords _key and _content to access object properties
tim("objects", {things:{
roses:"red",
violets:"blue",
dogs:"green",
oranges:"round"
}});
// roses are red, violets are blue, dogs are green, oranges are round,
Do you have any issues, questions or suggestions, or are you finding Tim useful in one of your projects? See github.com/premasagar/tim/issues, or get in touch (@premasagar).