One usecase is adding metadata to your code that can then be used to apply syntax highlighting.
How?
copy the config.js and edit it in order to specify how
certain tokens are to be surrounded/replaced
replace the undefined of each token you want to configure with one of the following
{String} config
'before:after'
wraps the token inside before/after
{Object} config
{ _before: 'before', _after: 'after' }
wraps token inside before/after
Missing before and after resolution for {String} and {Object} config
For the {String} and {Object} configurations, 'before' or 'after' may be omitted:
{String}:
'before:' (omitting 'after')
':after' (omitting 'before')
{Object}:
{ _before: 'before' } (omitting '_after')
{ _after: 'after' } (omitting '_before')
In these cases the missing half is resolved as follows:
from the parent._default (i.e., Keyword._default) if found
otherwise from the config._default if found
otherwise '' (empty string)
{Function} config
function (tokenString, info) { return {String}|{Object}; }
Inputs
tokenString: the content of the token that is currently being processed
info: an object with the following structure
{ // {Int} // the index of the token being processed inside tokens tokenIndex // {Array} // all tokens that are being processed including comments // (i.e. the result of merging esprima tokens and comments) , tokens // {Object} // the abstract syntax tree of the parsed code , ast // {String} // the code that was parsed (same string as the one passed to redeyed(code ..) , code}
In most cases the tokenString is all you need. The extra info object is passed in case you need to gather more
information about the token's surroundings in order to decide how to transform it.
See: replace-log-example
Output
You can return a {String} or an {Object} from a {Function} config.
when returning a {String}, the token value will be replaced with it
when returning an {Object}, it should be of the following form:
{ // {String} // the string that should be substituted for the value of the current and all skipped tokens replacement // {Object} (Token) // the token after which processing should continue // all tokens in between the current one and this one inclusive will be ignored , skipPastToken}
Transforming JavaScript code
redeyed(code, config[, opts])
Invoke redeyed with your configuration, a code snippet and maybe opts as in the below example:
var redeyed = require('redeyed') , config = require('./path/to/config') , code = 'var a = 3;' , result;// redeyed will throw an error (caused by the esprima parser) if the code has invalid javascripttry { result = redeyed(code, config); console.log(result.code);} catch(err) { console.error(err);}
opts:
{ // {Boolean} // if true `result.ast` property contains the abstract syntax tree of the code // if false (default) `result.ast` is not assigned and therefore `undefined` buildAst: true|false // {Boolean} // if `true`, jsx syntax is supported, default `false` // due to how esprima works, the AST is built when this option is `true`, even if // `buildAST` is `false`, jsx: true|false // {Boolean} // if true `result.code` is not assigned and therefore `undefined` // if false (default) `result.code` property contains the result of `split.join` nojoin: true|false // {Object} // overrides default parser `esprima-fb` and needs to be compatible with it parser: require('esprima') }