-- Copyright 2006-2013 JMS. See LICENSE. -- Scala LPeg Lexer. local l = require('lexer') local token, word_match = l.token, l.word_match local P, R, S = lpeg.P, lpeg.R, lpeg.S local M = {_NAME = 'scala'} -- Whitespace. local ws = token(l.WHITESPACE, l.space^1) -- Comments. local line_comment = '//' * l.nonnewline_esc^0 local block_comment = '/*' * (l.any - '*/')^0 * P('*/')^-1 local comment = token(l.COMMENT, line_comment + block_comment) -- Strings. local symbol = "'" * l.word local dq_str = l.delimited_range('"', true) local tq_str = '"""' * (l.any - '"""')^0 * P('"""')^-1 local string = token(l.STRING, tq_str + symbol + dq_str) -- Numbers. local number = token(l.NUMBER, (l.float + l.integer) * S('LlFfDd')^-1) -- Keywords. local keyword = token(l.KEYWORD, word_match{ 'abstract', 'case', 'catch', 'class', 'def', 'do', 'else', 'extends', 'false', 'final', 'finally', 'for', 'forSome', 'if', 'implicit', 'import', 'lazy', 'match', 'new', 'null', 'object', 'override', 'package', 'private', 'protected', 'return', 'sealed', 'super', 'this', 'throw', 'trait', 'try', 'true', 'type', 'val', 'var', 'while', 'with', 'yield' }) -- Types. local type = token(l.TYPE, word_match{ 'Array', 'Boolean', 'Buffer', 'Byte', 'Char', 'Collection', 'Double', 'Float', 'Int', 'Iterator', 'LinkedList', 'List', 'Long', 'Map', 'None', 'Option', 'Set', 'Short', 'SortedMap', 'SortedSet', 'String', 'TreeMap', 'TreeSet' }) -- Identifiers. local identifier = token(l.IDENTIFIER, l.word) -- Operators. local operator = token(l.OPERATOR, S('+-/*%<>!=^&|?~:;.()[]{}')) -- Functions. local func = token(l.FUNCTION, l.word) * #P('(') -- Classes. local class_sequence = token(l.KEYWORD, P('class')) * ws^1 * token(l.CLASS, l.word) M._rules = { {'whitespace', ws}, {'class', class_sequence}, {'keyword', keyword}, {'type', type}, {'function', func}, {'identifier', identifier}, {'string', string}, {'comment', comment}, {'number', number}, {'operator', operator}, } M._foldsymbols = { _patterns = {'[{}]', '/%*', '%*/', '//'}, [l.OPERATOR] = {['{'] = 1, ['}'] = -1}, [l.COMMENT] = {['/*'] = 1, ['*/'] = -1, ['//'] = l.fold_line_comments('//')} } return M