Merge branch 'master' into add-proximity-test

lib/linguist/languages.yml

@@ -28,6 +28,7 @@
 
 ABAP:
   type: programming
+  color: "#E8274B"
   extensions:
   - .abap
   ace_mode: abap
@@ -337,6 +338,7 @@ Boo:
 
 Brainfuck:
   type: programming
+  color: "#2F2530"
   extensions:
   - .b
   - .bf
@@ -643,6 +645,7 @@ Creole:
 
 Crystal:
   type: programming
+  color: "#76EEF5"
   extensions:
   - .cr
   ace_mode: ruby
@@ -869,6 +872,10 @@ Erlang:
   - .es
   - .escript
   - .hrl
+  filenames:
+  - rebar.config
+  - rebar.config.lock
+  - rebar.lock
   ace_mode: erlang
   interpreters:
   - escript
@@ -1564,6 +1571,7 @@ Kit:
 
 Kotlin:
   type: programming
+  color: "#EA4DFA"
   extensions:
   - .kt
   - .ktm
@@ -1781,6 +1789,7 @@ MUF:
 
 Makefile:
   type: programming
+  color: "#427819"
   aliases:
   - bsdmake
   - make
@@ -2930,6 +2939,7 @@ Shell:
   - .fcgi
   - .ksh
   - .tmux
+  - .tool
   - .zsh
   interpreters:
   - bash

lib/linguist/shebang.rb

@@ -23,17 +23,20 @@ module Linguist
       # First line must start with #!
       return unless shebang && shebang.start_with?("#!")
 
-      # Get the parts of the shebang without the #!
-      tokens = shebang.sub(/^#!\s*/, '').strip.split(' ')
+      s = StringScanner.new(shebang)
 
       # There was nothing after the #!
-      return if tokens.empty?
+      return unless path = s.scan(/^#!\s*\S+/)
 
-      # Get the name of the interpreter
-      script = File.basename(tokens.first)
+      # Keep going
+      script = path.split('/').last
 
-      # Get next argument if interpreter was /usr/bin/env
-      script = tokens[1] if script == 'env'
+      # if /usr/bin/env type shebang then walk the string
+      if script == 'env'
+        s.scan(/\s+/)
+        s.scan(/.*=[^\s]+\s+/) # skip over variable arguments e.g. foo=bar
+        script = s.scan(/\S+/)
+      end
 
       # Interpreter was /usr/bin/env with no arguments
       return unless script
@@ -41,6 +44,9 @@ module Linguist
       # "python2.6" -> "python2"
       script.sub! /(\.\d+)$/, ''
 
+      # #! perl -> perl
+      script.sub! /^#!\s*/, ''
+
       # Check for multiline shebang hacks that call `exec`
       if script == 'sh' &&
         data.lines.first(5).any? { |l| l.match(/exec (\w+).+\$0.+\$@/) }

lib/linguist/tokenizer.rb

@@ -22,8 +22,10 @@ module Linguist
     # Start state on token, ignore anything till the next newline
     SINGLE_LINE_COMMENTS = [
       '//', # C
+      '--', # Ada, Haskell, AppleScript
       '#',  # Ruby
       '%',  # Tex
+      '"',  # Vim
     ]
 
     # Start state on opening token, ignore anything until the closing
@@ -130,6 +132,9 @@ module Linguist
     #   extract_shebang("#!/usr/bin/env node")
     #   # => "node"
     #
+    #   extract_shebang("#!/usr/bin/env A=B foo=bar awk -f")
+    #   # => "awk"
+    #
     # Returns String token or nil it couldn't be parsed.
     def extract_shebang(data)
       s = StringScanner.new(data)
@@ -138,6 +143,7 @@ module Linguist
         script = path.split('/').last
         if script == 'env'
           s.scan(/\s+/)
+          s.scan(/.*=[^\s]+\s+/)
           script = s.scan(/\S+/)
         end
         script = script[/[^\d]+/, 0] if script