Merge branch 'master' into specflow

lib/linguist.rb

@@ -6,3 +6,15 @@ require 'linguist/repository'
 require 'linguist/samples'
 require 'linguist/shebang'
 require 'linguist/version'
+
+class << Linguist
+  attr_accessor :instrumenter
+
+  def instrument(*args, &bk)
+    if instrumenter
+      instrumenter.instrument(*args, &bk)
+    else
+      yield if block_given?
+    end
+  end
+end

lib/linguist/blob_helper.rb

@@ -236,6 +236,21 @@ module Linguist
       path =~ VendoredRegexp ? true : false
     end
 
+    documentation_paths = YAML.load_file(File.expand_path("../documentation.yml", __FILE__))
+    DocumentationRegexp = Regexp.new(documentation_paths.join('|'))
+
+    # Public: Is the blob in a documentation directory?
+    #
+    # Documentation files are ignored by language statistics.
+    #
+    # See "documentation.yml" for a list of documentation conventions that match
+    # this pattern.
+    #
+    # Return true or false
+    def documentation?
+      path =~ DocumentationRegexp ? true : false
+    end
+
     # Public: Get each line of data
     #
     # Requires Blob#data
@@ -317,5 +332,15 @@ module Linguist
     def tm_scope
       language && language.tm_scope
     end
+
+    DETECTABLE_TYPES = [:programming, :markup].freeze
+
+    # Internal: Should this blob be included in repository language statistics?
+    def include_in_language_stats?
+      !vendored? &&
+      !documentation? &&
+      !generated? &&
+      language && DETECTABLE_TYPES.include?(language.type)
+    end
   end
 end

lib/linguist/documentation.yml

@@ -0,0 +1,23 @@
+# Documentation files and directories are excluded from language
+# statistics.
+#
+# Lines in this file are Regexps that are matched against the file
+# pathname.
+#
+# Please add additional test coverage to
+# `test/test_blob.rb#test_documentation` if you make any changes.
+
+## Documentation directories ##
+
+- ^docs?/
+- (^|/)[Dd]ocumentation/
+- (^|/)javadoc/
+- ^man/
+
+## Documentation files ##
+
+- (^|/)CONTRIBUTING(\.|$)
+- (^|/)COPYING(\.|$)
+- (^|/)INSTALL(\.|$)
+- (^|/)LICEN[CS]E(\.|$)
+- (^|/)README(\.|$)

lib/linguist/file_blob.rb

@@ -74,7 +74,7 @@ module Linguist
     #
     # Returns an Array
     def extensions
-      basename, *segments = name.split(".")
+      basename, *segments = name.downcase.split(".")
 
       segments.map.with_index do |segment, index|
         "." + segments[index..-1].join(".")

lib/linguist/generated.rb

@@ -59,10 +59,13 @@ module Linguist
       godeps? ||
       generated_by_zephir? ||
       minified_files? ||
+      source_map? ||
       compiled_coffeescript? ||
       generated_parser? ||
       generated_net_docfile? ||
       generated_postscript? ||
+      compiled_cython_file? ||
+      generated_protocol_buffer_go? ||
       generated_protocol_buffer? ||
       generated_jni_header? ||
       vcr_cassette?
@@ -95,6 +98,20 @@ module Linguist
       end
     end
 
+    # Internal: Is the blob a generated source map?
+    #
+    # Source Maps usually have .css.map or .js.map extensions. In case they
+    # are not following the name convention, detect them based on the content.
+    #
+    # Returns true or false.
+    def source_map?
+      return false unless extname.downcase == '.map'
+
+      name =~ /(\.css|\.js)\.map$/i ||                 # Name convention
+      lines[0] =~ /^{"version":\d+,/ ||                # Revision 2 and later begin with the version number
+      lines[0] =~ /^\/\*\* Begin line maps\. \*\*\/{/  # Revision 1 begins with a magic comment
+    end
+
     # Internal: Is the blob of JS generated by CoffeeScript?
     #
     # CoffeeScript is meant to output JS that would be difficult to
@@ -214,6 +231,13 @@ module Linguist
         creator.include?("ImageMagick")
     end
 
+    def generated_protocol_buffer_go?
+      return false unless extname == '.go'
+      return false unless lines.count > 1
+
+      return lines[0].include?("Code generated by protoc-gen-go")
+    end
+
     # Internal: Is the blob a C++, Java or Python source file generated by the
     # Protocol Buffer compiler?
     #
@@ -274,5 +298,18 @@ module Linguist
       # VCR Cassettes have "recorded_with: VCR" in the second last line.
       return lines[-2].include?("recorded_with: VCR")
     end
+
+    # Internal: Is this a compiled C/C++ file from Cython?
+    #
+    # Cython-compiled C/C++ files typically contain:
+    # /* Generated by Cython x.x.x on ... */
+    # on the first line.
+    #
+    # Return true or false
+    def compiled_cython_file?
+      return false unless ['.c', '.cpp'].include? extname
+      return false unless lines.count > 1
+      return lines[0].include?("Generated by Cython")
+    end
   end
 end

lib/linguist/heuristics.rb

@@ -33,7 +33,7 @@ module Linguist
     #     disambiguate "Perl", "Prolog" do |data|
     #       if data.include?("use strict")
     #         Language["Perl"]
-    #       elsif data.include?(":-")
+    #       elsif /^[^#]+:-/.match(data)
     #         Language["Prolog"]
     #       end
     #     end
@@ -61,6 +61,9 @@ module Linguist
       @heuristic.call(data)
     end
 
+    # Common heuristics
+    ObjectiveCRegex = /^[ \t]*@(interface|class|protocol|property|end|synchronised|selector|implementation)\b/
+
     disambiguate "BitBake", "BlitzBasic" do |data|
       if /^\s*; /.match(data) || data.include?("End Function")
         Language["BlitzBasic"]
@@ -78,7 +81,7 @@ module Linguist
     end
 
     disambiguate "Objective-C", "C++", "C" do |data|
-      if (/^[ \t]*@(interface|class|protocol|property|end|synchronised|selector|implementation)\b/.match(data))
+      if ObjectiveCRegex.match(data)
         Language["Objective-C"]
       elsif (/^\s*#\s*include <(cstdint|string|vector|map|list|array|bitset|queue|stack|forward_list|unordered_map|unordered_set|(i|o|io)stream)>/.match(data) ||
         /^\s*template\s*</.match(data) || /^[ \t]*try/.match(data) || /^[ \t]*catch\s*\(/.match(data) || /^[ \t]*(class|(using[ \t]+)?namespace)\s+\w+/.match(data) || /^[ \t]*(private|public|protected):$/.match(data) || /std::\w+/.match(data))
@@ -89,29 +92,42 @@ module Linguist
     disambiguate "Perl", "Perl6", "Prolog" do |data|
       if data.include?("use v6")
         Language["Perl6"]
-      elsif data.include?("use strict")
+      elsif data.match(/use strict|use\s+v?5\./)
         Language["Perl"]
-      elsif data.include?(":-")
+      elsif /^[^#]+:-/.match(data)
         Language["Prolog"]
       end
     end
 
     disambiguate "ECL", "Prolog" do |data|
-      if data.include?(":-")
+      if /^[^#]+:-/.match(data)
         Language["Prolog"]
       elsif data.include?(":=")
         Language["ECL"]
       end
     end
 
-    disambiguate "IDL", "Prolog" do |data|
-      if data.include?(":-")
+    disambiguate "IDL", "Prolog", "INI", "QMake" do |data|
+      if /^[^#]+:-/.match(data)
         Language["Prolog"]
-      else
+      elsif data.include?("last_client=")
+        Language["INI"]
+      elsif data.include?("HEADERS") && data.include?("SOURCES")
+        Language["QMake"]
+      elsif /^\s*function[ \w,]+$/.match(data)
         Language["IDL"]
       end
     end
 
+    disambiguate "GAP", "Scilab" do |data|
+      if (data.include?("gap> "))
+        Language["GAP"]
+      # Heads up - we don't usually write heuristics like this (with no regex match)
+      else
+        Language["Scilab"]
+      end
+    end
+
     disambiguate "Common Lisp", "OpenCL", "Cool" do |data|
       if data.include?("(defun ")
         Language["Common Lisp"]
@@ -138,25 +154,51 @@ module Linguist
       end
     end
 
-    disambiguate "AsciiDoc", "AGS Script" do |data|
-      Language["AsciiDoc"] if /^=+(\s|\n)/.match(data)
+    disambiguate "AsciiDoc", "AGS Script", "Public Key" do |data|
+      if /^[=-]+(\s|\n)|{{[A-Za-z]/.match(data)
+        Language["AsciiDoc"]
+      elsif /^(\/\/.+|((import|export)\s+)?(function|int|float|char)\s+((room|repeatedly|on|game)_)?([A-Za-z]+[A-Za-z_0-9]+)\s*[;\(])/.match(data)
+        Language["AGS Script"]
+      elsif /^-----BEGIN/.match(data)
+        Language["Public Key"]
+      end
     end
 
-    disambiguate "FORTRAN", "Forth" do |data|
+    disambiguate "FORTRAN", "Forth", "Formatted" do |data|
       if /^: /.match(data)
         Language["Forth"]
-      elsif /^([c*][^a-z]|      (subroutine|program)\s|!)/i.match(data)
+      elsif /^([c*][^a-z]|      (subroutine|program)\s|\s*!)/i.match(data)
         Language["FORTRAN"]
       end
     end
 
-    disambiguate "F#", "Forth", "GLSL" do |data|
+    disambiguate "F#", "Forth", "GLSL", "Filterscript" do |data|
       if /^(: |new-device)/.match(data)
         Language["Forth"]
       elsif /^\s*(#light|import|let|module|namespace|open|type)/.match(data)
         Language["F#"]
-      elsif /^\s*(#include|#pragma|precision|uniform|varying|void)/.match(data)
+      elsif /^\s*(#version|precision|uniform|varying|vec[234])/.match(data)
         Language["GLSL"]
+      elsif /#include|#pragma\s+(rs|version)|__attribute__/.match(data)
+        Language["Filterscript"]
+      end
+    end
+
+    disambiguate "Limbo", "M", "MUF", "Mathematica", "Matlab", "Mercury", "Objective-C" do |data|
+      if ObjectiveCRegex.match(data)
+        Language["Objective-C"]
+      elsif data.include?(":- module")
+        Language["Mercury"]
+      elsif /^: /.match(data)
+        Language["MUF"]
+      elsif /^\s*;/.match(data)
+        Language["M"]
+      elsif /^\s*\(\*/.match(data)
+        Language["Mathematica"]
+      elsif /^\s*%/.match(data)
+        Language["Matlab"]
+      elsif /^\w+\s*:\s*module\s*{/.match(data)
+        Language["Limbo"]
       end
     end
 
@@ -172,6 +214,14 @@ module Linguist
       end
     end
 
+    disambiguate "Common Lisp", "NewLisp" do |data|
+      if /^\s*\((defun|in-package|defpackage) /.match(data)
+        Language["Common Lisp"]
+      elsif /^\s*\(define /.match(data)
+        Language["NewLisp"]
+      end
+    end
+
     disambiguate "TypeScript", "XML" do |data|
       if data.include?("<TS ")
         Language["XML"]
@@ -189,5 +239,55 @@ module Linguist
         Language["Text"]
       end
     end
+
+    disambiguate "PLSQL", "SQLPL", "PLpgSQL", "SQL" do |data|
+      if /^\\i\b|AS \$\$|LANGUAGE '+plpgsql'+/i.match(data) || /SECURITY (DEFINER|INVOKER)/i.match(data) || /BEGIN( WORK| TRANSACTION)?;/i.match(data)
+        #Postgres
+        Language["PLpgSQL"]
+      elsif /(alter module)|(language sql)|(begin( NOT)+ atomic)/i.match(data)  || /signal SQLSTATE '[0-9]+'/i.match(data)
+        #IBM db2
+        Language["SQLPL"]
+      elsif /pragma|\$\$PLSQL_|XMLTYPE|sysdate|systimestamp|\.nextval|connect by|AUTHID (DEFINER|CURRENT_USER)/i.match(data) || /constructor\W+function/i.match(data)
+        #Oracle
+        Language["PLSQL"]
+      elsif ! /begin|boolean|package|exception/i.match(data)
+        #Generic SQL
+        Language["SQL"]
+      end
+    end
+
+    disambiguate "D", "DTrace", "Makefile" do |data|
+      if /^module /.match(data)
+        Language["D"]
+      elsif /^((dtrace:::)?BEGIN|provider |#pragma (D (option|attributes)|ident)\s)/.match(data)
+        Language["DTrace"]
+      elsif /(\/.*:( .* \\)$| : \\$|^ : |: \\$)/.match(data)
+        Language["Makefile"]
+      end
+    end
+
+    disambiguate "OCaml", "Standard ML" do |data|
+      if /(^\s*module)|let rec |match\s+(\S+\s)+with/.match(data)
+        Language["OCaml"]
+      elsif /=> |case\s+(\S+\s)+of/.match(data)
+        Language["Standard ML"]
+      end
+    end
+
+    disambiguate "NL", "NewLisp" do |data|
+      if /^g3 /.match(data)
+        Language["NL"]
+      else
+        Language["NewLisp"]
+      end
+    end
+
+    disambiguate "Rust", "RenderScript" do |data|
+      if data.include?("^(use |fn |mod |pub |macro_rules|impl|#!?\[)")
+        Language["Rust"]
+      elsif /#include|#pragma\s+(rs|version)|__attribute__/.match(data)
+        Language["RenderScript"]
+      end
+    end
   end
 end

lib/linguist/language.rb

@@ -11,6 +11,7 @@ require 'linguist/samples'
 require 'linguist/file_blob'
 require 'linguist/blob_helper'
 require 'linguist/strategy/filename'
+require 'linguist/strategy/modeline'
 require 'linguist/shebang'
 
 module Linguist
@@ -31,13 +32,6 @@ module Linguist
     # Valid Languages types
     TYPES = [:data, :markup, :programming, :prose]
 
-    # Names of non-programming languages that we will still detect
-    #
-    # Returns an array
-    def self.detectable_markup
-      ["CSS", "Less", "Sass", "SCSS", "Stylus", "TeX"]
-    end
-
     # Detect languages by a specific type
     #
     # type - A symbol that exists within TYPES
@@ -79,7 +73,7 @@ module Linguist
           raise ArgumentError, "Extension is missing a '.': #{extension.inspect}"
         end
 
-        @extension_index[extension] << language
+        @extension_index[extension.downcase] << language
       end
 
       language.interpreters.each do |interpreter|
@@ -94,8 +88,9 @@ module Linguist
     end
 
     STRATEGIES = [
-      Linguist::Strategy::Filename,
+      Linguist::Strategy::Modeline,
       Linguist::Shebang,
+      Linguist::Strategy::Filename,
       Linguist::Heuristics,
       Linguist::Classifier
     ]
@@ -110,19 +105,31 @@ module Linguist
       # Bail early if the blob is binary or empty.
       return nil if blob.likely_binary? || blob.binary? || blob.empty?
 
-      # Call each strategy until one candidate is returned.
-      STRATEGIES.reduce([]) do |languages, strategy|
-        candidates = strategy.call(blob, languages)
-        if candidates.size == 1
-          return candidates.first
-        elsif candidates.size > 1
-          # More than one candidate was found, pass them to the next strategy.
-          candidates
-        else
-          # No candiates were found, pass on languages from the previous strategy.
-          languages
+      Linguist.instrument("linguist.detection", :blob => blob) do
+        # Call each strategy until one candidate is returned.
+        languages = []
+        returning_strategy = nil
+
+        STRATEGIES.each do |strategy|
+          returning_strategy = strategy
+          candidates = Linguist.instrument("linguist.strategy", :blob => blob, :strategy => strategy, :candidates => languages) do
+            strategy.call(blob, languages)
+          end
+          if candidates.size == 1
+            languages = candidates
+            break
+          elsif candidates.size > 1
+            # More than one candidate was found, pass them to the next strategy.
+            languages = candidates
+          else
+            # No candidates, try the next strategy
+          end
         end
-      end.first
+
+        Linguist.instrument("linguist.detected", :blob => blob, :strategy => returning_strategy, :language => languages.first)
+
+        languages.first
+      end
     end
 
     # Public: Get all Languages
@@ -155,7 +162,7 @@ module Linguist
     #   Language.find_by_alias('cpp')
     #   # => #<Language name="C++">
     #
-    # Returns the Lexer or nil if none was found.
+    # Returns the Language or nil if none was found.
     def self.find_by_alias(name)
       name && @alias_index[name.downcase]
     end
@@ -196,7 +203,7 @@ module Linguist
     # Returns all matching Languages or [] if none were found.
     def self.find_by_extension(extname)
       extname = ".#{extname}" unless extname.start_with?(".")
-      @extension_index[extname]
+      @extension_index[extname.downcase]
     end
 
     # DEPRECATED
@@ -219,7 +226,7 @@ module Linguist
     end
 
 
-    # Public: Look up Language by its name or lexer.
+    # Public: Look up Language by its name.
     #
     # name - The String name of the Language
     #
@@ -243,7 +250,7 @@ module Linguist
     #
     # This list is configured in "popular.yml".
     #
-    # Returns an Array of Lexers.
+    # Returns an Array of Languages.
     def self.popular
       @popular ||= all.select(&:popular?).sort_by { |lang| lang.name.downcase }
     end
@@ -255,7 +262,7 @@ module Linguist
     #
     # This list is created from all the languages not listed in "popular.yml".
     #
-    # Returns an Array of Lexers.
+    # Returns an Array of Languages.
     def self.unpopular
       @unpopular ||= all.select(&:unpopular?).sort_by { |lang| lang.name.downcase }
     end
@@ -375,11 +382,6 @@ module Linguist
     # Returns the name String
     attr_reader :search_term
 
-    # Public: Get Lexer
-    #
-    # Returns the Lexer
-    attr_reader :lexer
-
     # Public: Get the name of a TextMate-compatible scope
     #
     # Returns the scope
@@ -495,16 +497,6 @@ module Linguist
       @searchable
     end
 
-    # Public: Highlight syntax of text
-    #
-    # text    - String of code to be highlighted
-    # options - A Hash of options (defaults to {})
-    #
-    # Returns html String
-    def colorize(text, options = {})
-      lexer.highlight(text, options)
-    end
-
     # Public: Return name as String representation
     def to_s
       name
@@ -548,8 +540,8 @@ module Linguist
 
     if extnames = extensions[name]
       extnames.each do |extname|
-        if !options['extensions'].index { |x| x.end_with? extname }
-          warn "#{name} has a sample with extension (#{extname}) that isn't explicitly defined in languages.yml" unless extname == '.script!'
+        if !options['extensions'].index { |x| x.downcase.end_with? extname.downcase }
+          warn "#{name} has a sample with extension (#{extname.downcase}) that isn't explicitly defined in languages.yml" unless extname == '.script!'
           options['extensions'] << extname
         end
       end
@@ -580,7 +572,6 @@ module Linguist
       :color             => options['color'],
       :type              => options['type'],
       :aliases           => options['aliases'],
-      :lexer             => options['lexer'],
       :tm_scope          => options['tm_scope'],
       :ace_mode          => options['ace_mode'],
       :wrap              => options['wrap'],

lib/linguist/lazy_blob.rb

@@ -4,7 +4,7 @@ require 'rugged'
 
 module Linguist
   class LazyBlob
-    GIT_ATTR = ['linguist-language', 'linguist-vendored']
+    GIT_ATTR = ['linguist-documentation', 'linguist-language', 'linguist-vendored']
     GIT_ATTR_OPTS = { :priority => [:index], :skip_system => true }
     GIT_ATTR_FLAGS = Rugged::Repository::Attributes.parse_opts(GIT_ATTR_OPTS)
 
@@ -39,11 +39,19 @@ module Linguist
       end
     end
 
+    def documentation?
+      if attr = git_attributes['linguist-documentation']
+        boolean_attribute(attr)
+      else
+        super
+      end
+    end
+
     def language
       return @language if defined?(@language)
 
       @language = if lang = git_attributes['linguist-language']
-        Language.find_by_name(lang)
+        Language.find_by_alias(lang)
       else
         super
       end

lib/linguist/popular.yml

@@ -9,21 +9,21 @@
 - CSS
 - Clojure
 - CoffeeScript
-- Common Lisp
-- Diff
-- Emacs Lisp
-- Erlang
+- Go
 - HTML
 - Haskell
 - Java
 - JavaScript
 - Lua
+- Matlab
 - Objective-C
 - PHP
 - Perl
 - Python
+- R
 - Ruby
-- SQL
 - Scala
-- Scheme
 - Shell
+- Swift
+- TeX
+- VimL

lib/linguist/repository.rb

@@ -156,13 +156,8 @@ module Linguist
 
           blob = Linguist::LazyBlob.new(repository, delta.new_file[:oid], new, mode.to_s(8))
 
-          # Skip vendored or generated blobs
-          next if blob.vendored? || blob.generated? || blob.language.nil?
-
-          # Only include programming languages and acceptable markup languages
-          if blob.language.type == :programming || Language.detectable_markup.include?(blob.language.name)
-            file_map[new] = [blob.language.group.name, blob.size]
-          end
+          next unless blob.include_in_language_stats?
+          file_map[new] = [blob.language.group.name, blob.size]
         end
       end
 

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/strategy/modeline.rb

@@ -0,0 +1,30 @@
+module Linguist
+  module Strategy
+    class Modeline
+      EmacsModeline = /-\*-\s*(?:(?!mode)[\w-]+\s*:\s*(?:[\w+-]+)\s*;?\s*)*(?:mode\s*:)?\s*([\w+-]+)\s*(?:;\s*(?!mode)[\w-]+\s*:\s*[\w+-]+\s*)*;?\s*-\*-/i
+      VimModeline = /\/\*\s*vim:\s*set\s*(?:ft|filetype)=(\w+):\s*\*\//i
+
+      # Public: Detects language based on Vim and Emacs modelines
+      #
+      # blob               - An object that quacks like a blob.
+      #
+      # Examples
+      #
+      #   Modeline.call(FileBlob.new("path/to/file"))
+      #
+      # Returns an Array with one Language if the blob has a Vim or Emacs modeline
+      # that matches a Language name or alias. Returns an empty array if no match.
+      def self.call(blob, _ = nil)
+        Array(Language.find_by_alias(modeline(blob.data)))
+      end
+
+      # Public: Get the modeline from the first n-lines of the file
+      #
+      # Returns a String or nil
+      def self.modeline(data)
+        match = data.match(EmacsModeline) || data.match(VimModeline)
+        match[1] if match
+      end
+    end
+  end
+end

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
@@ -33,7 +35,8 @@ module Linguist
       ['<!--', '-->'], # XML
       ['{-', '-}'],    # Haskell
       ['(*', '*)'],    # Coq
-      ['"""', '"""']   # Python
+      ['"""', '"""'],  # Python
+      ["'''", "'''"]   # Python
     ]
 
     START_SINGLE_LINE_COMMENT =  Regexp.compile(SINGLE_LINE_COMMENTS.map { |c|
@@ -129,6 +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)
@@ -137,6 +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

lib/linguist/vendor.yml

@@ -24,6 +24,9 @@
 - (^|/)config.guess$
 - (^|/)config.sub$
 
+# Linters
+- cpplint.py
+
 # Node dependencies
 - node_modules/
 
@@ -40,7 +43,7 @@
 # Minified JavaScript and CSS
 - (\.|-)min\.(js|css)$
 
-#Stylesheets imported from packages
+# Stylesheets imported from packages
 - ([^\s]*)import\.(css|less|scss|styl)$
 
 # Bootstrap css and js
@@ -143,6 +146,9 @@
 
 ## Python ##
 
+# Sphinx
+- (^|/)docs?/_?(build|themes?|templates?|static)/
+
 # django
 - (^|/)admin_media/
 
@@ -221,7 +227,7 @@
 - ^readme$
 
 # Test fixtures
-- ^[Tt]est/fixtures/
+- ^[Tt]ests?/fixtures/
 
 # PhoneGap/Cordova
 - (^|/)cordova([^.]*)\.js$
@@ -233,7 +239,7 @@
 # Vagrant
 - ^Vagrantfile$
 
-# .DS_Store's
+# .DS_Stores
 - .[Dd][Ss]_[Ss]tore$
 
 # R packages
@@ -251,3 +257,9 @@
 # ProGuard
 - proguard.pro
 - proguard-rules.pro
+
+# PuPHPet
+- ^puphpet/
+
+# Android Google APIs
+- (^|/)\.google_apis/

lib/linguist/version.rb

@@ -1,3 +1,3 @@
 module Linguist
-  VERSION = "4.2.6"
+  VERSION = "4.5.4"
 end