this should be a highlighted keyword.

foo = asdf: 1
#     ^ marked as .variable.assignment.coffee

Should probably also mark it as .property (and maybe drop .variable?)

Everything after do keyword is function except function declaration, and it should be highlighted accordingly

Hi :)

I initially posted this issue over at github-linguist/linguist#2529, and they referred me over here:

I came across this when I was looking through a README (https://github.com/zooniverse/Panoptes-Front-End):

<ChangeListener target={auth}>{=>
  <PromiseRenderer promise={auth.checkCurrent()}>{(user) =>
    if user?
      <p>User is {user.display_name}</p>
    else
      <p>Nobody's signed in.</p>
  }</PromiseRenderer>
}</ChangeListener>

The single quote starts a string that doesn't end. I had a go at escaping it and I'm not sure that you can:

<ChangeListener target={auth}>{=>
  <PromiseRenderer promise={auth.checkCurrent()}>{(user) =>
    if user?
      <p>User is {user.display_name}</p>
    else
      <p>Nobody\'s signed in.</p>
  }</PromiseRenderer>
}</ChangeListener>

Is this a bug, or am I missing something incredibly obvious?

All the best,

Mike

I am not sure what I am doing wrong here, but here is a video.

the name of of a function has two capture groups.
looking at the grammar there seems to be 4 top level capture groups which makes me think it was intentional. (also took a look at the javascript grammar and this seems to be in line with function expressions there too =/ but different to function declarations which only capture the name in one group)

class SomeClass
  initialize: (param) -> null

  <span class="meta function coffee">
    <span class="entity name function coffee">
      "initializ"
      <span class="entity name function coffee">e</span>
   </span>
  </span>

screenshot of the dom:

screenshot of the grammar:

class is definitely a keyword.

hey kev,

awesome job so far on the cs language plugin and atom in general!

i just wanted to document the following scenario:

i have the following coffeescript:

module.exports =
  build_dir: 'build'
  compile_dir: 'bin'
  app_files:
    js: [ 'src/**/*.js', '!src/**/*.spec.js', '!src/assets/**/*.js' ]
    json: [ 'src/**/*.json' ]
    jsunit: [ 'src/**/*.spec.js' ]
    coffee: [ 'src/**/*.coffee', '!src/**/*.spec.coffee' ]
    coffeeunit: [ 'src/**/*.spec.coffee' ]
    atpl: [ 'src/app/**/*.tpl.html' ]
    ctpl: [ 'src/common/**/*.tpl.html' ]
    html: [ 'src/index.html' ]
    less: 'src/less/main.less'
  test_files:
    js: [ 'vendor/angular-mocks/angular-mocks.js' ]
  vendor_files:
    js: [
      'vendor/angular/angular.js'
      'vendor/angular-bootstrap/ui-bootstrap-tpls.min.js'
      'vendor/bower-angular-placeholders/angular-placeholders.js'
      'vendor/angular-ui-router/release/angular-ui-router.js'
      'vendor/angular-ui-utils/modules/route/route.js'
      # tk
      'vendor/jquery/dist/jquery.js'
      'vendor/bootstrap/dist/js/bootstrap.js'
      #'vendor/bootstrap/js/collapse.js'
      #'vendor/bootstrap/js/transition.js'
      'vendor/angular-flash/dist/angular-flash.js'
      'vendor/lodash/dist/lodash.js'
    ]
    css: []
    assets: [ 'vendor/font-awesome/fonts/*' ]
    fonts: [ 'vendor/font-awesome/fonts/*' ]

and when i run auto-indent on it, i get the following (notice how test_files lines up under less) non-equivalent coffeescript:

module.exports =
  build_dir: 'build'
  compile_dir: 'bin'
  app_files:
    js: [ 'src/**/*.js', '!src/**/*.spec.js', '!src/assets/**/*.js' ]
    json: [ 'src/**/*.json' ]
    jsunit: [ 'src/**/*.spec.js' ]
    coffee: [ 'src/**/*.coffee', '!src/**/*.spec.coffee' ]
    coffeeunit: [ 'src/**/*.spec.coffee' ]
    atpl: [ 'src/app/**/*.tpl.html' ]
    ctpl: [ 'src/common/**/*.tpl.html' ]
    html: [ 'src/index.html' ]
    less: 'src/less/main.less'
    test_files:
      js: [ 'vendor/angular-mocks/angular-mocks.js' ]
      vendor_files:
        js: [
          'vendor/angular/angular.js'
          'vendor/angular-bootstrap/ui-bootstrap-tpls.min.js'
          'vendor/bower-angular-placeholders/angular-placeholders.js'
          'vendor/angular-ui-router/release/angular-ui-router.js'
          'vendor/angular-ui-utils/modules/route/route.js'
          # tk
          'vendor/jquery/dist/jquery.js'
          'vendor/bootstrap/dist/js/bootstrap.js'
          #'vendor/bootstrap/js/collapse.js'
          #'vendor/bootstrap/js/transition.js'
          'vendor/angular-flash/dist/angular-flash.js'
          'vendor/lodash/dist/lodash.js'
        ]
        css: []
        assets: [ 'vendor/font-awesome/fonts/*' ]
        fonts: [ 'vendor/font-awesome/fonts/*' ]

regards,
tony.

This may be a deliberate decision, and it may also be acceptable behaviour, but I thought it was an oddity so I'll report it anyway.

The left hand side of a standard assignment is normally highlighted blue-ish, but it isn't in the case of an or= assignment.

Repro steps:

1. Change the default editor tab length to anything other than 2. (I use 4 globally)
2. Try to change Coffee tab length to 2.

Actual: The setting is removed and then replaced by whatever value is set in #1
Expected: The setting remains at 2.

I was looking for a regex to identify function names and saw this:

{
  'captures':
    '1':
      'name': 'entity.name.function.coffee'
    '2':
      'name': 'entity.name.function.coffee'
    '3':
      'name': 'variable.parameter.function.coffee'
    '4':
      'name': 'storage.type.function.coffee'
  'match': '(?x)\n\t\t\t\t(?<=^|\\s)\n\t\t\t\t(?=@?[a-zA-Z\\$_])\n\t\t\t\t(\n\t\t\t\t\t@?[a-zA-Z\\$_](\\w|\\$|:|\\.)*\\s*\n\t\t\t\t\t(?=[:=](\\s*\\(.*\\))?\\s*([=-]>))\n\t\t\t\t)\n\t\t\t'
  'name': 'meta.function.coffee'
}

Now the regEx is ported from textmate and equals:

(?x)
(?<=^|\s)
(?=@?[a-zA-Z\$_])
(
    @?[a-zA-Z\$_](\w|\$|:|\.)*\s*
    (?=[:=](\s*\(.*\))?\s*([=-]>))
)

Trying to use it though threw me errors for the content of line 1 and line 2. Upon research it seems that

• (?<=^|\s) lookbehinds are not supported by javascript.
• (?x) I dont even know what this is supposed to do?

Removing those makes the regex work perfectly fine. Bug or is atom implementing lookbehinds somewhere else?

Pressing enter after the following line indents and it shouldn't

element.classList

@test: (a) ->
  return false unless true
#        ^^^^^ this is being marked as `entity.name.function`.  The `true` is correct though.

[email protected]

Consider this:

module.exports=
class Commit extends Treeish

  initialize: (
    ref
    repo
    @tree
    @parents
    @author
    @authoredDate
    @committer
    @committedDate
    @gpgsig
    @message
  ) -> super(ref, repo)

  @parse: (raw, repo) ->
    throw new Error('No valid git repo!!!') unless repo?.isGitRepo
    throw new Error('No raw data!!!') unless (typeof(raw) is 'string')

Your page is rendering it correctly but atom is not:

If the @ is removed, the function name is highlighted correctly.

obj.func @test

Using standard (not literate) coffeescript, if you type a line that starts with form, the next line will be incorrectly indented.

atom 1.5.0, language-coffee-script 0.46.0, reproduces in safe mode

Most of the issues are in the way of parsing variable names.

Load this example into atom. When I say something is not tagged I mean its scope is only .source.coffee.

someFunc = (arg1, arg2, arg3=1, arg4...) ->
  aVariable = 'ok'
  anArray = new Array()
  anArray.slice()           # anArray not tagged
  anArray.otherMethod()     # anArray, otherMethod not tagged
  randomGlobal.prop.omg()   # not tagged

  functionCall(one)         # functionCall not tagged
  functionCall one          # functionCall not tagged

  anObject =
    key1: 'val1'
    key2: (one, two) ->

# Totally messes this line up
grade = (student, period=(if b? then 7 else 6), messages={"A": "Excellent"}) ->
  if student.excellentWork  # not tagged
    "A+"
  else if student.okayStuff # not tagged
    if student.triedHard then "B" else "B-" # student.triedHard not tagged

class Animal extends Being
  constructor: (@name) ->   # @name not tagged the same as @name in the body, should be?
    @name += ' is cool'

# Ok
math =
  root:   Math.sqrt
  square: square
  cube:   (x) -> x * square x

# Ok
heredoc = """
  CoffeeScript subst test #{ 010 + 0xf / 0b10 + "nested string #{ /\n/ }"}
"""

Giving these symbols correct scopes will greatly improve the support in the coming SymbolProvider in the soon-to-be-bundled-in-core autocomplete-plus.

variable.parameter should match param1 , param2
Now: (param1, param2)

f1 = (param1, param2) -> x * x

A function without arguments wrapped in parentheses will not be processed as a function.

This is a function:

dimensionNames: ( () ->
    ....
).property('dimensions.@each')

This isn't:

dimensionNames: ( ->
    ....
).property('dimensions.@each')

CoffeeScript supports string interpolation in strings with double quotes. If I use it in a string that ends with a colon, coloring breaks - see screenshot below.

foo = [
    { value: 1 }
    { value: "#{var} :" }
    { value: "2" }
]

I’m guessing there’s a non-greedy match going on here, as only the second parameter is highlighted when everything from (up to - ought to be.

foo: (bar, baz = bob()) ->

Hi,

I noticed that if I have a line like this in a coffeescript file:

values.map((v) -> v.value.toLowerCase().replace(/ /g, '_')).join('_').replace(/[^0-9a-z_]/g, '')

...the remaining code in the file will not be highlighted correctly. (in fact, it appears that GitHub gets it wrong too)

I am running language-coffee-script version 0.46.1.

I noticed the Javascript language package comes with a nice regexp grammar.

It would be super rad to have that in CoffeeScript, too!

Maybe the JS grammar could even serve as a base, with modifications to support CS features, like block syntax:

/// ^ (
  ?: [-=]>             # function
   | [-+*/%<>&|^!?=]=  # compound assign / compare
   | >>>=?             # zero-fill right shift
   | ([-+:])\1         # doubles
   | ([&|<>])\2=?      # logic / shift
   | \?\.              # soak access
   | \.{2,3}           # range or splat
) ///

and interpolation:

blah = '.*'
///#{blah}///

Thoughts?

@50Wliu, any chance you can bump this package to 0.43.0?

Fuzzy finder is logging errors for any broken symlinks in the project

ENOENT: no such file or directory, stat '/Users/kevin/github/atom/node_modules/fs-plus/node_modules/.bin/rimraf'
    at Error (native)
    at fs.statSync (fs.js:846:18)
    at Object.fs.statSync (ATOM_SHELL_ASAR.js:206:16)
    at Object.realpathSync (fs.js:1456:12)
    at Object.fs.realpathSync (ATOM_SHELL_ASAR.js:243:29)
    at /Applications/Atom.app/Contents/Resources/app/node_modules/fuzzy-finder/lib/load-paths-handler.js:90:20
    at FSReqWrap.oncomplete (fs.js:77:15)

Syntax Highlighting seems off here, as these expressions are equal.
I build atom directly from the repository.
Well a fork actually, created 4th of december.
Couldn't find this in issues for this filter: is:issue is:open syntax Highlighting coffee

Example:

# correct, this has class inline function
obj.method () ->
# wrong, class is entity name function coffee, but this is a call
obj.method ->

# both lines are identical and valid,
#  the second one is actually more idomatic

"#" + Tab will execute no action

Not sure if conflicting with another plugin.

This is the error i get when the atom tries to syntax-hilight a .litcoffee file. This problem goes away if i disable language-coffee-script

Atom Version: 1.2.0
System: Mac OS X 10.11.1
Thrown From: Atom Core

Stack Trace

Uncaught Error: end pattern with unmatched parenthesis

At /opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/scanner.js:31

Error: end pattern with unmatched parenthesis
    at Error (native)
    at Scanner.module.exports.Scanner.createScanner (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/scanner.js:31:24)
    at Scanner.module.exports.Scanner.getScanner (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/scanner.js:37:31)
    at Scanner.module.exports.Scanner.findNextMatch (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/scanner.js:56:22)
    at Rule.module.exports.Rule.findNextMatch (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/rule.js:99:28)
    at Rule.module.exports.Rule.getNextTags (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/rule.js:141:21)
    at Grammar.module.exports.Grammar.tokenizeLine (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/first-mate/lib/grammar.js:150:44)
    at TokenizedBuffer.module.exports.TokenizedBuffer.buildTokenizedLineForRowWithText (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/src/tokenized-buffer.js:546:28)
    at TokenizedBuffer.module.exports.TokenizedBuffer.buildTokenizedLineForRow (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/src/tokenized-buffer.js:532:19)
    at TokenizedBuffer.module.exports.TokenizedBuffer.tokenizeNextChunk (/opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/src/tokenized-buffer.js:292:43)
    at /opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/src/tokenized-buffer.js:269:26
    at /opt/homebrew-cask/Caskroom/atom/latest/Atom.app/Contents/Resources/app.asar/node_modules/underscore-plus/node_modules/underscore/underscore.js:666:47

Commands

  7x -2:22.1.0 blur (atom-text-editor.editor.vim-mode-plus.normal-mode.is-focused)
     -2:19.5.0 core:confirm (atom-text-editor.editor.mini.is-focused)
     -2:15.6.0 core:select-all (atom-text-editor.editor.mini.is-focused)
     -2:13.9.0 core:confirm (atom-text-editor.editor.mini.is-focused)
     -2:11.4.0 core:select-all (atom-text-editor.editor.mini.is-focused)
     -2:10.3.0 core:confirm (atom-text-editor.editor.mini.is-focused)
  7x -2:08.6.0 blur (atom-text-editor.editor.mini.is-focused)
     -2:04.6.0 core:close (atom-text-editor.editor.vim-mode-plus.normal-mode.is-focused)
  3x -2:04.5.0 blur (atom-text-editor.editor.vim-mode-plus.normal-mode.is-focused)
  3x -2:00.8.0 core:close (ol.tree-view.full-menu.list-tree.has-collapsable-children.focusable-panel)
 33x -1:51.9.0 blur (ol.tree-view.full-menu.list-tree.has-collapsable-children.focusable-panel)
     -1:18.6.0 core:confirm (atom-text-editor.editor.mini.is-focused)
  2x -0:40.7.0 blur (atom-text-editor.editor.mini.is-focused)
     -0:37.5.0 core:backspace (atom-text-editor.editor.mini.is-focused)
     -0:37.3.0 core:confirm (atom-text-editor.editor.mini.is-focused)
 10x -0:26 blur (atom-text-editor.editor.mini.is-focused)

Config

{
  "core": {
    "disabledPackages": [
      "minimap",
      "minimap-autohide",
      "minimap-git-diff",
      "minimap-selection",
      "minimap-pigments",
      "last-cursor-position",
      "tidy-tabs",
      "color-tabs",
      "tab-switcher",
      "terminal-panel",
      "tree-view-git-status",
      "pane-split-moves-tab",
      "preview-plus",
      "source-preview",
      "vim-mode",
      "recent-files-fuzzy-finder"
    ],
    "themes": [
      "native-ui",
      "base16-tomorrow-dark-theme"
    ]
  },
  "editor": {
    "invisibles": {},
    "fontFamily": "Monaco",
    "lineHeight": 1.2,
    "softWrap": true,
    "fontSize": 15,
    "preferredLineLength": 120
  }
}

Installed Packages

# User
antialiased, v0.1.0
api-blueprint-preview, v0.6.0
autocomplete-ctags, v0.3.2
autocomplete-modules, v1.3.1
autocomplete-python, v0.10.3
git-log, v0.4.1
language-api-blueprint, v1.0.2
language-cjsx, v0.3.0
language-docker, v1.1.3
language-groovy, v0.4.0
language-javascript-jsx, v0.3.7
markdown-preview-plus, v2.2.2
markdown-scroll-sync, v0.4.2
native-ui, v0.8.1
quick-highlight, v0.3.7
tabs-to-spaces, v0.11.1
vim-mode-plus, v0.2.0

# Dev
No dev packages

It's a small issue, but != should really be treated in the exact same way as ==.

Steps to reproduce:

1. Create new file.
2. Enter the following code.

state_enums = [ 'AL', 'AK', 'AZ', 'AR', 'CA', 'CO', 'CT', 'DE', 'DC', 'FL', 'GA', 'GU', 'HI', 'ID', 'IL', 'IN', 'IA', 'KS', 'KY', 'LA', 'ME', 'MP', 'MD', 'MA', 'MI', 'MN', 'MS', 'MO', 'MT', 'NE', 'NV', 'NH', 'NJ', 'NM', 'NY', 'NC', 'ND', 'OH', 'OK', 'OR', 'PA', 'PR', 'RI', 'AS', 'SC', 'SD', 'TN', 'TX', 'UT', 'VT', 'VI', 'VA', 'WA', 'WV', 'WI', 'WY' ]
state_enums.push('ON', 'QC', 'NS', 'NB', 'MB', 'BC', 'PE', 'AB', 'NL') # Canada`

1. Now indent the code once by selecting it all and doing cmd/ctrl + ]

What you should see is the code loses it's highlighting after 20 or so index entries but this seems to only be the case when it is indented. Outdenting restores the highlighting.

Sticky is a function but it's not highlighted

The following file is rendered as such in Atom:

As can be seen, the source blocks are not highlighted. Anyone encountering this too?
Please note that the markdown parts are highlighted correctly in the same file.
Atom 0.201

This is an edge case, but the following valid CoffeeScript produces regex highlighting between the /'s in the conditional.

a = 0
if (a /= 1) < (1 / 2)
  return 1 / 2

Not sure if this is too specific to request (as it is related to React.js), but it would be nice if when using CoffeeScripts backticks to escape code it would properly highlight HTML-like syntax.

Before:

After:

If this is way out of scope, forgive me!

This expression is valid in CoffeeScript, but is highlighted very weird:

# Wrong:
a = {} if a[0] == 0

# Correct:
a = {} if a[0] != 0

Including code block after a list breaks highlighting. See screenshot (source code on the left, markdown preview on the right):

The example text:

Some text in `.litcoffee` file with functions in it.

    a = ->
      m = 'indented function'
      return 'properly formatted ' + m

1. Example
2. List


    b = ->
      m = 'indendet function'
      return 'not highlighted ' + m

1. Another
2. List

With some text after it.

    c = ->
      m = 'indendet function'
      return 'properly highlighted ' + m

Save as file.litcoffee and open in atom.io. Function is not highlighted properly in editor, however markdown preview is fine.

[email protected], [email protected], on OSX Yosmite

I'd be happy to setup a PR with a spec for this, and try fixing it, but I'm not sure how to run the tests.

If you use backticks (`) to escape raw JavaScript in your Coffeescript file, it seems to break the syntax highlighting / parsing.

The following line causes this:

`import Foo from 'bar'`

It seems that the interior of the backticks is parsed as JavaScript (cool!), but since ES6 supports backticks for string templating, it sees the the ending backtick as the start of an ES6 template string, rather than the terminating escape character in Coffeescript (not cool).

Interestingly, this only seems to appear in Atom v0.189.0. The prior version (v0.188.0) doesn't show this problem. Not sure if this issue belongs here or in the Atom repo, let me know.

I have a RegEx for TLDs first thing in a coffeescript file that breaks the color styling for the rest of the file.

/^(http(?:s?)\:\/\/|~\/|\/)?(?:\w+:\w+@)?([A-Za-z0-9-]{1,63}\.)+(COMMUNITY|DIRECTORY|EDUCATION|EQUIPMENT|INSTITUTE|MARKETING|SOLUTIONS|XN--J1AMH|XN--L1ACC|BARGAINS|BOUTIQUE|BUILDERS|CATERING|CLEANING|CLOTHING|COMPUTER|DEMOCRAT|DIAMONDS|GRAPHICS|HOLDINGS|LIGHTING|PARTNERS|PLUMBING|TRAINING|VENTURES|XN--P1AI|ACADEMY|CAREERS|COMPANY|CRUISES|DOMAINS|EXPOSED|FLIGHTS|FLORIST|GALLERY|GUITARS|HOLIDAY|KITCHEN|RECIPES|RENTALS|REVIEWS|SHIKSHA|SINGLES|SUPPORT|SYSTEMS|AGENCY|BERLIN|CAMERA|CENTER|COFFEE|CONDOS|DATING|ESTATE|EVENTS|EXPERT|FUTBOL|KAUFEN|LUXURY|MAISON|MONASH|MUSEUM|NAGOYA|PHOTOS|REPAIR|REPORT|SOCIAL|TATTOO|TIENDA|TRAVEL|VIAJES|VILLAS|VISION|VOTING|VOYAGE|BUILD|CARDS|CHEAP|CODES|DANCE|EMAIL|GLASS|HOUSE|NINJA|PARTS|PHOTO|SHOES|SOLAR|TODAY|TOKYO|TOOLS|WATCH|WORKS|AERO|ARPA|ASIA|BIKE|BLUE|BUZZ|CAMP|CLUB|COOL|COOP|FARM|GIFT|GURU|INFO|JOBS|KIWI|LAND|LIMO|LINK|MENU|MOBI|MODA|NAME|PICS|PINK|POST|QPON|RICH|RUHR|SEXY|TIPS|WANG|WIEN|ZONE|BIZ|CAB|CAT|CEO|COM|EDU|GOV|INT|KIM|MIL|NET|ONL|ORG|PRO|RED|TEL|UNO|WED|XXX|AC|AD|AE|AF|AG|AI|AL|AM|AN|AO|AQ|AR|AS|AT|AU|AW|AX|AZ|BA|BB|BD|BE|BF|BG|BH|BI|BJ|BM|BN|BO|BR|BS|BT|BV|BW|BY|BZ|CA|CC|CD|CF|CG|CH|CI|CK|CL|CM|CN|CO|CR|CU|CV|CW|CX|CY|CZ|DE|DJ|DK|DM|DO|DZ|EC|EE|EG|ER|ES|ET|EU|FI|FJ|FK|FM|FO|FR|GA|GB|GD|GE|GF|GG|GH|GI|GL|GM|GN|GP|GQ|GR|GS|GT|GU|GW|GY|HK|HM|HN|HR|HT|HU|ID|IE|IL|IM|IN|IO|IQ|IR|IS|IT|JE|JM|JO|JP|KE|KG|KH|KI|KM|KN|KP|KR|KW|KY|KZ|LA|LB|LC|LI|LK|LR|LS|LT|LU|LV|LY|MA|MC|MD|ME|MG|MH|MK|ML|MM|MN|MO|MP|MQ|MR|MS|MT|MU|MV|MW|MX|MY|MZ|NA|NC|NE|NF|NG|NI|NL|NO|NP|NR|NU|NZ|OM|PA|PE|PF|PG|PH|PK|PL|PM|PN|PR|PS|PT|PW|PY|QA|RE|RO|RS|RU|RW|SA|SB|SC|SD|SE|SG|SH|SI|SJ|SK|SL|SM|SN|SO|SR|ST|SU|SV|SX|SY|SZ|TC|TD|TF|TrG|TH|TJ|TK|TL|TM|TN|TO|TP|TR|TT|TV|TW|TZ|UA|UG|UK|US|UY|UZ|VA|VC|VE|VG|VI|VN|VU|WF|WS|YE|YT|ZA|ZM|ZW)(.*)$/mig

it should look more like

Atom: 0.106.0
OSX: 10.9.3
Lang-CS: 0.22.0

Case 1: step into function definition, put the cursor at the beginning of some line, press Enter
Result: current line indent is lost, text starts from the beginning of line

Case 2: cut some line, paste somewhere inside of the same function
Result: the same as above

File contents of:

#{test
#test
#{test

Result:

Coffeescript allows for classes to be created and instanciated as follows:

new class
  constructor: ->
    console.log 'I am a dog'

new class Cat
  constructor: ->
    console.log 'I am a cat'

However this syntax is unrecognised, and is formatted as if class were the constructor function's name.

Coffeescript allows for anonymous subclasses, created with class extends animal, however this syntax is incorrectly parsed as though extends were a class name, when in fact it is a keyword.

When inserting new comments in the code, despite the indentation level it always puts the #at the first column, creating compilation and jshint errors.

I believe that keyword operators such as and|or|isnt should not be the same color as other operators (==|!=|+=) because, while they are still in the same category, they are often used in a different manner.

My proposal is having them receive the same coloring as control keywords.

On the left, the current one. On the right, the proposed change.
Imgur

Have a look here:

The parser seems to assume this snippet contains a regular expression.