* Create Mustache Syntax.md * Update Mustache Syntax.md * Create Transforms.md * Update Transforms.md * Update Mustache Syntax.md * Update Mustache Syntax.md * Create Lambdas.md * Update Lambdas.md
HummingbirdMustache
Package for rendering Mustache templates. Mustache is a logicless templating language commonly used in web and mobile platforms. You can find out more about Mustache here.
While Hummingbird Mustache has been designed to be used with the Hummingbird server framework it has no dependencies and can be used as a standalone library.
Usage
Load your templates from the filesystem
let library = HBMustacheLibrary("folder/my/templates/are/in")
This will look for all the files with the extension ".mustache" in the specified folder and subfolders and attempt to load them. Each file is registed with the name of the file (with subfolder, if inside a subfolder) minus the "mustache" extension.
Render an object with a template
let output = library.render(object, withTemplate: "myTemplate")
HummingbirdMustache will render both dictionaries and objects via Mirror reflection. The following two examples will both produce the same output
let object = ["name": "John Smith", "age": 68]
let output = library.render(object, withTemplate: "myTemplate")
and
struct Person {
let name: String
let age: Int
}
let object = Person(name: "John Smith", age: 68)
let output = library.render(object, withTemplate: "myTemplate")
Support
Hummingbird Mustache supports all standard Mustache tags and is fully compliant with the Mustache spec with the exception of the Lambda support.
Additional features
Hummingbird Mustache includes some features that are specific to its implementation.
Lambda Implementation
The library doesn't provide a lambda implementation but it does provide something akin to the lambda feature.
Add a HBMustacheLambda to the object you want to be rendered and it can be used in a similar way to lambdas are used in Mustache. When you create a section referencing the lambda the contents of the section are passed as a template along with the current object to the lamdba function. This is slightly different from the standard implementation where the unprocessed text is passed to the lambda.
Given the following mustache template
let mustache = "{{#wrapped}}{{name}} is awesome.{{/wrapped}}"
let template = try HBMustacheTemplate(string: mustache)
The following object john
struct Object {
let name: String
let wrapped: HBMustacheLambda
}
let john = Object(
name: "John",
wrapped: HBMustacheLambda({ object, template in
return "<b>\(template.render(object))</b>"
})
)
let output = template.render(john)
Will render as
<b>John is awesome.</b>
Transforms
Transforms are similar to lambdas in that they are functions run on an object but with the difference they return a new object instead of rendered text. Transforms are formatted as a function call inside a tag eg
{{uppercase(string)}}
They can be applied to variable, section and inverted section tags. If you apply them to a section or inverted section tag the handler name should be included in the end section tag as well eg
{{#sorted(array)}}{{.}}{{/sorted(array)}}
The library comes with a series of transforms for the Swift standard objects.
- String/Substring
- capitalized: Return string with first letter capitalized
- lowercase: Return lowercased version of string
- uppercase: Return uppercased version of string
- reversed: Reverse string
- Int/UInt/Int8/Int16...
- plusone: Add one to integer
- minusone: Subtract one from integer
- odd: return if integer is odd
- even: return if integer is even
- Array
- first: Return first element of array
- last: Return last element of array
- count: Return number of elements in array
- reversed: Reverse array
- sorted: If the elements of the array are comparable sort them
- Dictionary
- count: Return number of elements in dictionary
- enumerated: Return dictionary as array of key, value pairs
- sorted: If the keys are comparable return as array of key, value pairs sorted by key
If a transform is applied to an object that doesn't recognise it then nil is returned.
Sequence context transforms
Sequence context transforms are transforms applied to the current position in the sequence. They are formatted as a function that takes no parameter eg
{{#first()}}First{{/first()}}
The following sequence context transforms are available
- first: Is this the first element of the sequence
- last: Is this the last element of the sequence
- index: Returns the index of the element within the sequence
- odd: Returns if the index of the element is odd
- even: Returns if the index of the element is even
Template inheritance
Template inheritance is not part of the Mustache spec yet but it is a commonly implemented feature. Template inheritance allows you to override elements of an included partial. It allows you to create a base page template and override elements of it with your page content. A partial that includes overriding elements is indicated with a {{<partial}}. Note this is different from the normal partial reference which uses >. This is a section tag so needs a ending tag as well. Inside the section the tagged sections to override are added using the syntax {{$tag}}contents{{/tag}}. If your template and partial were as follows
{{! mypage.mustache }}
{{<base}}
{{$head}}<title>My page title</title>{{/head}}
{{$body}}Hello world{{/body}}
{{/base}}
{{! base.mustache }}
<html>
<head>
{{$head}}{{/head}}
</head>
<body>
{{$body}}Default text{{/body}}
</body>
</html>
You would get the following output when rendering mypage.mustache.
<html>
<head>
<title>My page title</title>
</head>
<body>
Hello world
</body>
Note the {{$head}} section in base.mustache is replaced with the {{$head}} section included inside the {{<base}} partial reference from mypage.mustache. The same occurs with the {{$body}} section. In that case though a default value is supplied for the situation where a {{$body}} section is not supplied.
Pragma
The syntax {{% var: value}} can be used to set template rendering configuration variables specific to Hummingbird Mustache. The only variable you can set at the moment is CONTENT_TYPE. This can be set to either to HTML or TEXT and defines how variables are escaped. A content type of TEXT means no variables are escaped and a content type of HTML will do HTML escaping of the rendered text. The content type defaults to HTML.
Given input object "<>", template {{%CONTENT_TYPE: HTML}}{{.}} will render as <> and {{%CONTENT_TYPE: TEXT}}{{.}} will render as <>.