Skip to content
This repository has been archived by the owner on Jul 10, 2023. It is now read-only.

Latest commit

 

History

History
61 lines (47 loc) · 2.25 KB

README.md

File metadata and controls

61 lines (47 loc) · 2.25 KB
Documentation

xom

Transform XML trees into performant JavaScript DOM calls at compile-time using Nim code.

hello.nim

The above will compile to the following JavaScript code:

var p0 = document.createElement("p");
p0.appendChild(document.createTextNode("Hello!"));
document.body.appendChild(p0);

Installation

xom supports Nim 1.4.0+ and can be installed using Nimble:

$ nimble install xom

Customizing code generation

xom can be customized to generate code that is optimized for a particular use case. Nodes can be modified in-place, new child nodes can be created, the code generation can be suppressed all together for particular nodes, and variables are only created for nodes that you specify.

All this customization is done through the use of the two simple callbacks of the Xom context object:

  • onEnter*: XmlNode -> Command
  • onEmitNamed*: (XmlNode, NimNode) -> void

onEnter is called for every node that is found, and onEmitNamed is called for every node for which a variable has been requested. onEnter returns a Command object, which is an enum type that can be one of the following:

  • Emit: the node will be emitted but no variable will be created (default for all nodes).
  • EmitNamed: the node will be emitted and a variable for it will be created. This also triggers a call to onEmitNamed on the node.
  • Skip: the node will be skipped and no variable will be created.

Inside both callbacks, you can modify nodes in-place, and changes will be reflected in the generated code at compile-time.

By default, no variables are created for nodes that are not requested if not necessary (and necessary but not requested variables are always scoped by default). If you want to reference a node inside onEmitNamed, you have to return EmitNamed from onEnter for that node. In particular, text nodes are always merged together unless a variable is being emitted for them. Having a variable for a node is useful for dinamically modifying the node in separately generated code (see examples/ and tests/ for some simple use cases).