Jump To …

lexer.coffee

#

The CoffeeScript Lexer. Uses a series of token-matching regexes to attempt matches against the beginning of the source code. When a match is found, a token is produced, we consume the match, and start again. Tokens are in the form:

[tag, value, lineNumber]

Which is a format that can be fed directly into Jison.

{Rewriter} = require './rewriter'
#

Import the helpers we need.

{include, count, starts, compact} = require('./helpers').helpers
#

The Lexer Class

#

The Lexer class reads a stream of CoffeeScript and divvys it up into tagged tokens. Some potential ambiguity in the grammar has been avoided by pushing some extra smarts into the Lexer.

exports.Lexer = class Lexer
#

tokenize is the Lexer's main method. Scan by attempting to match tokens one at a time, using a regular expression anchored at the start of the remaining code, or a custom recursive token-matching method (for interpolations). When the next token has been recorded, we move forward within the code past the token, and begin again.

Each tokenizing method is responsible for incrementing @i by the number of characters it has consumed. @i can be thought of as our finger on the page of source.

Before returning the token stream, run it through the Rewriter unless explicitly asked not to.

  tokenize: (code, options) ->
    code     = code.replace /(\r|\s+$)/g, ''
    o        = options or {}
    @code    = code         # The remainder of the source code.
    @i       = 0            # Current character position we're parsing.
    @line    = o.line or 0  # The current line.
    @indent  = 0            # The current indentation level.
    @indebt  = 0            # The over-indentation at the current level.
    @outdebt = 0            # The under-outdentation at the current level.
    @indents = []           # The stack of all current indentation levels.
    @tokens  = []           # Stream of parsed tokens in the form ['TYPE', value, line]
    while @i < @code.length
      @chunk = @code[@i..]
      @extractNextToken()
    @closeIndentation()
    return @tokens if o.rewrite is off
    (new Rewriter).rewrite @tokens
#

At every position, run through this list of attempted matches, short-circuiting if any of them succeed. Their order determines precedence: @literalToken is the fallback catch-all.

  extractNextToken: ->
    return if @identifierToken()
    return if @commentToken()
    return if @whitespaceToken()
    return if @lineToken()
    return if @heredocToken()
    return if @stringToken()
    return if @numberToken()
    return if @regexToken()
    return if @jsToken()
    return    @literalToken()
#

Tokenizers

#

Matches identifying literals: variables, keywords, method names, etc. Check to ensure that JavaScript reserved words aren't being used as identifiers. Because CoffeeScript reserves a handful of keywords that are allowed in JavaScript, we're careful not to tag them as keywords when referenced as property names here, so you can still do jQuery.is() even though is means === otherwise.

  identifierToken: ->
    return false unless id = @match IDENTIFIER, 1
    @i += id.length
    forcedIdentifier = @tagAccessor() or @match ASSIGNED, 1
    tag = 'IDENTIFIER'
    tag = id.toUpperCase() if include(JS_KEYWORDS, id) or (not forcedIdentifier and include(COFFEE_KEYWORDS, id))
    tag = 'LEADING_WHEN'   if tag is 'WHEN' and include LINE_BREAK, @tag()
    tag = 'ALL'            if id is 'all' and @tag() is 'FOR'
    tag = 'UNARY'          if include UNARY, tag
    if include(JS_FORBIDDEN, id)
      if forcedIdentifier
        tag = 'STRING'
        id  = "\"#{id}\""
        if forcedIdentifier is 'accessor'
          close_index = true
          @tokens.pop() if @tag() isnt '@'
          @token 'INDEX_START', '['
      else if include(RESERVED, id)
        @identifierError id
    unless forcedIdentifier
      tag = id = CONVERSIONS[id] if include COFFEE_ALIASES, id
      tag = 'LOGIC' if include LOGIC, id
      tag = 'UNARY' if id is '!'
    @token tag, id
    @token ']', ']' if close_index
    true
#

Matches numbers, including decimals, hex, and exponential notation. Be careful not to interfere with ranges-in-progress.

  numberToken: ->
    return false unless number = @match NUMBER, 1
    return false if @tag() is '.' and starts number, '.'
    @i += number.length
    @token 'NUMBER', number
    true
#

Matches strings, including multi-line strings. Ensures that quotation marks are balanced within the string's contents, and within nested interpolations.

  stringToken: ->
    return false unless starts(@chunk, '"') or starts(@chunk, "'")
    return false unless string =
      @balancedToken(['"', '"'], ['#{', '}']) or
      @balancedToken ["'", "'"]
    @interpolateString string.replace /\n/g, '\\\n'
    @line += count string, "\n"
    @i += string.length
    true
#

Matches heredocs, adjusting indentation to the correct level, as heredocs preserve whitespace, but ignore indentation to the left.

  heredocToken: ->
    return false unless match = @chunk.match HEREDOC
    quote = match[1].substr 0, 1
    doc = @sanitizeHeredoc match[2] or match[4] or '', {quote}
    @interpolateString quote + doc + quote, heredoc: yes
    @line += count match[1], "\n"
    @i += match[1].length
    true
#

Matches and consumes comments.

  commentToken: ->
    return false unless match = @chunk.match(COMMENT)
    @line += count match[1], "\n"
    @i += match[1].length
    if match[2]
      @token 'HERECOMMENT', @sanitizeHeredoc match[2],
        herecomment: true, indent: Array(@indent + 1).join(' ')
      @token 'TERMINATOR', '\n'
    true
#

Matches JavaScript interpolated directly into the source via backticks.

  jsToken: ->
    return false unless starts @chunk, '`'
    return false unless script = @balancedToken ['`', '`']
    @token 'JS', script.replace JS_CLEANER, ''
    @i += script.length
    true
#

Matches regular expression literals. Lexing regular expressions is difficult to distinguish from division, so we borrow some basic heuristics from JavaScript and Ruby, borrow slash balancing from @balancedToken, and borrow interpolation from @interpolateString.

  regexToken: ->
    return false unless first = @chunk.match REGEX_START
    return false if first[1] is ' ' and @tag() not in ['CALL_START', '=']
    return false if     include NOT_REGEX, @tag()
    return false unless regex = @balancedToken ['/', '/']
    return false unless end = @chunk.substr(regex.length).match REGEX_END
    regex += flags = end[2] if end[2]
    if regex.match REGEX_INTERPOLATION
      str = regex.substring(1).split('/')[0]
      str = str.replace REGEX_ESCAPE, (escaped) -> '\\' + escaped
      @tokens = @tokens.concat [['(', '('], ['NEW', 'new'], ['IDENTIFIER', 'RegExp'], ['CALL_START', '(']]
      @interpolateString "\"#{str}\"", escapeQuotes: yes
      @tokens.splice @tokens.length, 0, [',', ','], ['STRING', "\"#{flags}\""] if flags
      @tokens.splice @tokens.length, 0, [')', ')'], [')', ')']
    else
      @token 'REGEX', regex
    @i += regex.length
    true
#

Matches a token in which which the passed delimiter pairs must be correctly balanced (ie. strings, JS literals).

  balancedToken: (delimited...) ->
    @balancedString @chunk, delimited
#

Matches newlines, indents, and outdents, and determines which is which. If we can detect that the current line is continued onto the the next line, then the newline is suppressed:

elements
  .each( ... )
  .map( ... )

Keeps track of the level of indentation, because a single outdent token can close multiple indents, so we need to know how far in we happen to be.

  lineToken: ->
    return false unless indent = @match MULTI_DENT, 1
    @line += count indent, "\n"
    @i    += indent.length
    prev = @prev(2)
    size = indent.match(LAST_DENTS).reverse()[0].match(LAST_DENT)[1].length
    nextCharacter = @match NEXT_CHARACTER, 1
    noNewlines = nextCharacter is '.' or nextCharacter is ',' or @unfinished()
    if size - @indebt is @indent
      return @suppressNewlines() if noNewlines
      return @newlineToken indent
    else if size > @indent
      if noNewlines
        @indebt = size - @indent
        return @suppressNewlines()
      diff = size - @indent + @outdebt
      @token 'INDENT', diff
      @indents.push diff
      @outdebt = @indebt = 0
    else
      @indebt = 0
      @outdentToken @indent - size, noNewlines
    @indent = size
    true
#

Record an outdent token or multiple tokens, if we happen to be moving back inwards past several recorded indents.

  outdentToken: (moveOut, noNewlines, close) ->
    while moveOut > 0
      len = @indents.length - 1
      if @indents[len] is undefined
        moveOut = 0
      else if @indents[len] is @outdebt
        moveOut -= @outdebt
        @outdebt = 0
      else if @indents[len] < @outdebt
        @outdebt -= @indents[len]
        moveOut -= @indents[len]
      else
        dent = @indents.pop()
        dent -= @outdebt
        moveOut -= dent
        @outdebt = 0
        @token 'OUTDENT', dent
    @outdebt -= moveOut if dent
    @token 'TERMINATOR', "\n" unless @tag() is 'TERMINATOR' or noNewlines
    true
#

Matches and consumes non-meaningful whitespace. Tag the previous token as being "spaced", because there are some cases where it makes a difference.

  whitespaceToken: ->
    return false unless space = @match WHITESPACE, 1
    prev = @prev()
    prev.spaced = true if prev
    @i += space.length
    true
#

Generate a newline token. Consecutive newlines get merged together.

  newlineToken: (newlines) ->
    @token 'TERMINATOR', "\n" unless @tag() is 'TERMINATOR'
    true
#

Use a \ at a line-ending to suppress the newline. The slash is removed here once its job is done.

  suppressNewlines: ->
    @tokens.pop() if @value() is "\\"
    true
#

We treat all other single characters as a token. Eg.: ( ) , . ! Multi-character operators are also literal tokens, so that Jison can assign the proper order of operations. There are some symbols that we tag specially here. ; and newlines are both treated as a TERMINATOR, we distinguish parentheses that indicate a method call from regular parentheses, and so on.

  literalToken: ->
    match = @chunk.match OPERATOR
    value = match and match[1]
    space = match and match[2]
    @tagParameters() if value and value.match CODE
    value or= @chunk.substr 0, 1
    @i += value.length
    spaced = (prev = @prev()) and prev.spaced
    tag = value
    if value is '='
      @assignmentError() if include JS_FORBIDDEN, @value()
      if @value() in ['or', 'and']
        @tokens.splice(@tokens.length - 1, 1, ['COMPOUND_ASSIGN', CONVERSIONS[@value()] + '=', prev[2]])
        return true
    if value is ';'                         then tag = 'TERMINATOR'
    else if include(LOGIC, value)           then tag = 'LOGIC'
    else if include(MATH, value)            then tag = 'MATH'
    else if include(COMPARE, value)         then tag = 'COMPARE'
    else if include(COMPOUND_ASSIGN, value) then tag = 'COMPOUND_ASSIGN'
    else if include(UNARY, value)           then tag = 'UNARY'
    else if include(SHIFT, value)           then tag = 'SHIFT'
    else if include(CALLABLE, @tag()) and not spaced
      if value is '('
        prev[0] = 'FUNC_EXIST' if prev[0] is '?'
        tag = 'CALL_START'
      else if value is '['
        tag = 'INDEX_START'
        @tag 1, 'INDEX_SOAK'  if @tag() is '?'
        @tag 1, 'INDEX_PROTO' if @tag() is '::'
    @token tag, value
    true
#

Token Manipulators

#

As we consume a new IDENTIFIER, look at the previous token to determine if it's a special kind of accessor. Return true if any type of accessor is the previous token.

  tagAccessor: ->
    return false if (not prev = @prev()) or (prev and prev.spaced)
    accessor = if prev[1] is '::'
      @tag 1, 'PROTOTYPE_ACCESS'
    else if prev[1] is '.' and not (@value(2) is '.')
      if @tag(2) is '?'
        @tag(1, 'SOAK_ACCESS')
        @tokens.splice(-2, 1)
      else
        @tag 1, 'PROPERTY_ACCESS'
    else
      prev[0] is '@'
    if accessor then 'accessor' else false
#

Sanitize a heredoc or herecomment by escaping internal double quotes and erasing all external indentation on the left-hand side.

  sanitizeHeredoc: (doc, options) ->
    indent = options.indent
    return doc if options.herecomment and not include doc, '\n'
    unless options.herecomment
      while (match = HEREDOC_INDENT.exec(doc)) isnt null
        attempt = if match[2]? then match[2] else match[3]
        indent = attempt if not indent? or attempt.length < indent.length
    indent or= ''
    doc = doc.replace(new RegExp("^" + indent, 'gm'), '')
    return doc if options.herecomment
    doc = doc.replace(/^\n/, '')
    doc.replace(MULTILINER, "\\n")
       .replace(new RegExp(options.quote, 'g'), "\\#{options.quote}")
#

A source of ambiguity in our grammar used to be parameter lists in function definitions versus argument lists in function calls. Walk backwards, tagging parameters specially in order to make things easier for the parser.

  tagParameters: ->
    return if @tag() isnt ')'
    i = 0
    loop
      i += 1
      tok = @prev i
      return if not tok
      switch tok[0]
        when 'IDENTIFIER'       then tok[0] = 'PARAM'
        when ')'                then tok[0] = 'PARAM_END'
        when '(', 'CALL_START'  then return tok[0] = 'PARAM_START'
    true
#

Close up all remaining open blocks at the end of the file.

  closeIndentation: ->
    @outdentToken @indent
#

The error for when you try to use a forbidden word in JavaScript as an identifier.

  identifierError: (word) ->
    throw new Error "SyntaxError: Reserved word \"#{word}\" on line #{@line + 1}"
#

The error for when you try to assign to a reserved word in JavaScript, like "function" or "default".

  assignmentError: ->
    throw new Error "SyntaxError: Reserved word \"#{@value()}\" on line #{@line + 1} can't be assigned"
#

Matches a balanced group such as a single or double-quoted string. Pass in a series of delimiters, all of which must be nested correctly within the contents of the string. This method allows us to have strings within interpolations within strings, ad infinitum.

  balancedString: (str, delimited, options) ->
    options or= {}
    slash = delimited[0][0] is '/'
    levels = []
    i = 0
    while i < str.length
      if levels.length and starts str, '\\', i
        i += 1
      else
        for pair in delimited
          [open, close] = pair
          if levels.length and starts(str, close, i) and levels[levels.length - 1] is pair
            levels.pop()
            i += close.length - 1
            i += 1 unless levels.length
            break
          else if starts str, open, i
            levels.push(pair)
            i += open.length - 1
            break
      break if not levels.length or slash and starts str, '\n', i
      i += 1
    if levels.length
      return false if slash
      throw new Error "SyntaxError: Unterminated #{levels.pop()[0]} starting on line #{@line + 1}"
    if not i then false else str.substring(0, i)
#

Expand variables and expressions inside double-quoted strings using ECMA Harmony's interpolation syntax for substitution of bare variables as well as arbitrary expressions.

"Hello #{name.capitalize()}."

If it encounters an interpolation, this method will recursively create a new Lexer, tokenize the interpolated contents, and merge them into the token stream.

  interpolateString: (str, options) ->
    options or= {}
    if str.length < 3 or not starts str, '"'
      @token 'STRING', str
    else
      lexer   = new Lexer
      tokens  = []
      quote   = str.substring 0, 1
      [i, pi] = [1, 1]
      while i < str.length - 1
        if starts str, '\\', i
          i += 1
        else if expr = @balancedString(str.substring(i), [['#{', '}']])
          tokens.push ['STRING', quote + str.substring(pi, i) + quote] if pi < i
          inner = expr.substring(2, expr.length - 1)
          if inner.length
            inner = inner.replace new RegExp('\\\\' + quote, 'g'), quote if options.heredoc
            nested = lexer.tokenize "(#{inner})", line: @line
            (tok[0] = ')') for tok, idx in nested when tok[0] is 'CALL_END'
            nested.pop()
            tokens.push ['TOKENS', nested]
          else
            tokens.push ['STRING', quote + quote]
          i += expr.length - 1
          pi = i + 1
        i += 1
      tokens.push ['STRING', quote + str.substring(pi, i) + quote] if pi < i and pi < str.length - 1
      tokens.unshift ['STRING', '""'] unless tokens[0][0] is 'STRING'
      interpolated = tokens.length > 1
      @token '(', '(' if interpolated
      for token, i in tokens
        [tag, value] = token
        if tag is 'TOKENS'
          @tokens = @tokens.concat value
        else if tag is 'STRING' and options.escapeQuotes
          escaped = value.substring(1, value.length - 1).replace(/"/g, '\\"')
          @token tag, "\"#{escaped}\""
        else
          @token tag, value
        @token '+', '+' if i < tokens.length - 1
      @token ')', ')' if interpolated
      tokens
#

Helpers

#

Add a token to the results, taking note of the line number.

  token: (tag, value) ->
    @tokens.push [tag, value, @line]
#

Peek at a tag in the current token stream.

  tag: (index, newTag) ->
    return unless tok = @prev index
    return tok[0] = newTag if newTag?
    tok[0]
#

Peek at a value in the current token stream.

  value: (index, val) ->
    return unless tok = @prev index
    return tok[1] = val if val?
    tok[1]
#

Peek at a previous token, entire.

  prev: (index) ->
    @tokens[@tokens.length - (index or 1)]
#

Attempt to match a string against the current chunk, returning the indexed match if successful, and false otherwise.

  match: (regex, index) ->
    return false unless m = @chunk.match regex
    if m then m[index] else false
#

Are we in the midst of an unfinished expression?

  unfinished: ->
    prev = @prev(2)
    @value() and @value().match and @value().match(NO_NEWLINE) and
      prev and (prev[0] isnt '.') and not @value().match(CODE) and
      not @chunk.match ASSIGNED
#

Constants

#

Keywords that CoffeeScript shares in common with JavaScript.

JS_KEYWORDS = [
  "if", "else",
  "true", "false",
  "new", "return",
  "try", "catch", "finally", "throw",
  "break", "continue",
  "for", "in", "while",
  "delete", "instanceof", "typeof",
  "switch", "super", "extends", "class",
  "this", "null", "debugger"
]
#

CoffeeScript-only keywords, which we're more relaxed about allowing. They can't be used standalone, but you can reference them as an attached property.

COFFEE_ALIASES =  ["and", "or", "is", "isnt", "not"]
COFFEE_KEYWORDS = COFFEE_ALIASES.concat [
  "then", "unless", "until", "loop",
  "yes", "no", "on", "off",
  "of", "by", "where", "when"
]
#

The list of keywords that are reserved by JavaScript, but not used, or are used by CoffeeScript internally. We throw an error when these are encountered, to avoid having a JavaScript error at runtime.

RESERVED = [
  "case", "default", "do", "function", "var", "void", "with",
  "const", "let", "enum", "export", "import", "native",
  "__hasProp", "__extends", "__slice"
]
#

The superset of both JavaScript keywords and reserved words, none of which may be used as identifiers or properties.

JS_FORBIDDEN = JS_KEYWORDS.concat RESERVED
#

Token matching regexes.

IDENTIFIER    = /^([a-zA-Z\$_](\w|\$)*)/
NUMBER        = /^(((\b0(x|X)[0-9a-fA-F]+)|((\b[0-9]+(\.[0-9]+)?|\.[0-9]+)(e[+\-]?[0-9]+)?)))\b/i
HEREDOC       = /^("{6}|'{6}|"{3}([\s\S]*?)\n?([ \t]*)"{3}|'{3}([\s\S]*?)\n?([ \t]*)'{3})/
OPERATOR      = /^(-[\-=>]?|\+[+=]?|[*&|\/%=<>^:!?]+)([ \t]*)/
WHITESPACE    = /^([ \t]+)/
COMMENT       = /^(###([^#][\s\S]*?)(###[ \t]*\n|(###)?$)|(\s*#(?!##[^#])[^\n]*)+)/
CODE          = /^((-|=)>)/
MULTI_DENT    = /^((\n([ \t]*))+)(\.)?/
LAST_DENTS    = /\n([ \t]*)/g
LAST_DENT     = /\n([ \t]*)/
#

Regex-matching-regexes.

REGEX_START         = /^\/([^\/])/
REGEX_INTERPOLATION = /([^\\]#\{.*[^\\]\})/
REGEX_END           = /^(([imgy]{1,4})\b|\W|$)/
REGEX_ESCAPE        = /\\[^\$]/g
#

Token cleaning regexes.

JS_CLEANER      = /(^`|`$)/g
MULTILINER      = /\n/g
NO_NEWLINE      = /^([+\*&|\/\-%=<>!.\\][<>=&|]*|and|or|is|isnt|not|delete|typeof|instanceof)$/
HEREDOC_INDENT  = /(\n+([ \t]*)|^([ \t]+))/g
ASSIGNED        = /^\s*(([a-zA-Z\$_@]\w*|["'][^\r\n]+?["']|\d+)[ \t]*?[:=][^:=])/
NEXT_CHARACTER  = /^\s*(\S)/
#

Compound assignment tokens.

COMPOUND_ASSIGN = ['-=', '+=', '/=', '*=', '%=', '||=', '&&=', '?=', '<<=', '>>=', '>>>=', '&=', '^=', '|=']
#

Unary tokens.

UNARY   = ['UMINUS', 'UPLUS', '!', '!!', '~', 'TYPEOF', 'DELETE']
#

Logical tokens.

LOGIC   = ['&', '|', '^', '&&', '||']
#

Bit-shifting tokens.

SHIFT   = ['<<', '>>', '>>>']
#

Comparison tokens.

COMPARE = ['<=', '<', '>', '>=']
#

Mathmatical tokens.

MATH    = ['*', '/', '%']
#

Tokens which a regular expression will never immediately follow, but which a division operator might.

See: http://www.mozilla.org/js/language/js20-2002-04/rationale/syntax.html#regular-expressions

Our list is shorter, due to sans-parentheses method calls.

NOT_REGEX = ['NUMBER', 'REGEX', '++', '--', 'FALSE', 'NULL', 'TRUE', ']']
#

Tokens which could legitimately be invoked or indexed. A opening parentheses or bracket following these tokens will be recorded as the start of a function invocation or indexing operation.

CALLABLE = ['IDENTIFIER', 'SUPER', ')', ']', '}', 'STRING', '@', 'THIS', '?', '::']
#

Tokens that, when immediately preceding a WHEN, indicate that the WHEN occurs at the start of a line. We disambiguate these from trailing whens to avoid an ambiguity in the grammar.

LINE_BREAK = ['INDENT', 'OUTDENT', 'TERMINATOR']
#

Conversions from CoffeeScript operators into JavaScript ones.

CONVERSIONS =
  'and':  '&&'
  'or':   '||'
  'is':   '=='
  'isnt': '!='
  'not':  '!'
  '===':  '=='