icu-messageformat-parser v2.0.0

messageformat-parser

A PEG.js parser for ICU MessageFormat strings. Outputs an AST defined by parser.pegjs.

The generated parser function takes two parameters, first the string to be parsed, and a second optional parameter options, an object.

The options object contains arrays of keywords for cardinal and ordinal rules for the current locale – these are used to validate plural and selectordinal keys. If options or its fields are missing or set to false, the full set of valid Unicode CLDR keys is used: 'zero', 'one', 'two', 'few', 'many', 'other'. To disable this check, pass in an empty array.

The options object also supports two settings that make the parser follow the ICU MessageFormat spec more closely: strictNumberSign and strictFunctionParams.

Inside a plural or selectordinal statement, a pound symbol (#) is replaced with the input number. By default, # is parsed as a special character in nested statements too, and can be escaped using apostrophes ('#').

Setting strictNumberSign to true will only parse # as a special character directly inside a plural or selectordinal statement. Outside those, # and '#' are parsed as literal text.

By default, function parameters are split on commas and trimmed, so the parameters in {x,fn, a, b } are parsed as ['a','b']. Setting strictFunctionParams to true will result in a params array with a single element: [' a, b '].

The parser only supports the DOUBLE_OPTIONAL apostrophe mode. A single apostrophe only starts quoted literal text if preceded by a curly brace ({}) or a pound symbol (#) inside a plural or selectordinal statement, depending on the value of strictNumberSign. Otherwise, it is a literal apostrophe. A double apostrophe is always a literal apostrophe.

Installation

npm install icu-messageformat-parser

Usage

> var parse = require('messageformat-parser').parse;

> parse('So {wow}.')
[ 'So ', { type: 'argument', arg: 'wow' }, '.' ]

> parse('Such { thing }. { count, selectordinal, one {First} two {Second}' +
        '                  few {Third} other {#th} } word.')
[ 'Such ',
  { type: 'argument', arg: 'thing' },
  '. ',
  { type: 'selectordinal',
    arg: 'count',
    offset: 0,
    cases:
     [ { key: 'one', tokens: [ 'First' ] },
       { key: 'two', tokens: [ 'Second' ] },
       { key: 'few', tokens: [ 'Third' ] },
       { key: 'other', tokens: [ { type: 'octothorpe' }, 'th' ] } ] },
  ' word.' ]

> parse('Many{type,select,plural{ numbers}selectordinal{ counting}' +
                         'select{ choices}other{ some {type}}}.')
[ 'Many',
  { type: 'select',
    arg: 'type',
    cases:
     [ { key: 'plural', tokens: [ ' numbers' ] },
       { key: 'selectordinal', tokens: [ ' counting' ] },
       { key: 'select', tokens: [ ' choices' ] },
       { key: 'other', tokens: [ ' some',
                                 { type: 'argument', arg: 'type' } ] } ] },
  '.' ]

> parse('{Such compliance')
// SyntaxError: Expected ",", "}" or [ \t\n\r] but "c" found.

> var msg = '{words, plural, zero{No words} one{One word} other{# words}}';
> var englishKeys = { cardinal: [ 'one', 'other' ],
                      ordinal: [ 'one', 'two', 'few', 'other' ] };
> parse(msg)
[ { type: 'plural',
    arg: 'words',
    offset: 0,
    cases:
     [ { key: 'zero', tokens: [ 'No words' ] },
       { key: 'one', tokens: [ 'One word' ] },
       { key: 'other', tokens: [ { type: 'octothorpe' }, ' words' ] } ] } ]

> parse(msg, englishKeys)
// Error: Invalid key `zero` for argument `words`. Valid plural keys for this
//        locale are `one`, `other`, and explicit keys like `=0`.