KevinMidboe/linguist
1
0
Fork 0
mirror of https://github.com/KevinMidboe/linguist.git synced 2025-12-08 20:38:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' of https://github.com/github/linguist

Browse Source 
Conflicts:
	grammars.yml
This commit is contained in:
Brahn Partridge
2015-02-17 12:56:19 +01:00
parent 43ee45d9b6 6d94ddb114
commit 83b03df98b
468 changed files with 37765 additions and 2227 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
lib/linguist.rb
View File

@@ -4,4 +4,5 @@ require 'linguist/heuristics'
require 'linguist/language'
require 'linguist/repository'
require 'linguist/samples'
require 'linguist/shebang'
require 'linguist/version'

25
lib/linguist/blob_helper.rb
View File

@@ -236,6 +236,21 @@ module Linguist
name =~ 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?
name =~ 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

19
lib/linguist/classifier.rb
View File

@@ -3,6 +3,25 @@ require 'linguist/tokenizer'
module Linguist
# Language bayesian classifier.
class Classifier
# Public: Use the classifier to detect language of the blob.
#
# blob - An object that quacks like a blob.
# possible_languages - Array of Language objects
#
# Examples
#
# Classifier.call(FileBlob.new("path/to/file"), [
# Language["Ruby"], Language["Python"]
# ])
#
# Returns an Array of Language objects, most probable first.
def self.call(blob, possible_languages)
language_names = possible_languages.map(&:name)
classify(Samples.cache, blob.data, language_names).map do |name, _|
Language[name] # Return the actual Language objects
end
end
# Public: Train classifier that data is a certain language.
#
# db - Hash classifier database object

19
lib/linguist/documentation.yml Normal file
View File

@@ -0,0 +1,19 @@
# 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 Conventions ##
- ^docs?/
- ^Documentation/
- (^|/)CONTRIBUTING(\.|$)
- (^|/)COPYING(\.|$)
- (^|/)INSTALL(\.|$)
- (^|/)LICEN[CS]E(\.|$)
- (^|/)README(\.|$)

18
lib/linguist/generated.rb
View File

@@ -51,20 +51,20 @@ module Linguist
#
# Return true or false
def generated?
minified_files? ||
compiled_coffeescript? ||
xcode_file? ||
generated_parser? ||
generated_net_docfile? ||
generated_net_designer_file? ||
generated_postscript? ||
generated_protocol_buffer? ||
generated_jni_header? ||
composer_lock? ||
node_modules? ||
godeps? ||
vcr_cassette? ||
generated_by_zephir?
generated_by_zephir? ||
minified_files? ||
compiled_coffeescript? ||
generated_parser? ||
generated_net_docfile? ||
generated_postscript? ||
generated_protocol_buffer? ||
generated_jni_header? ||
vcr_cassette?
end
# Internal: Is the blob an Xcode file?

289
lib/linguist/heuristics.rb
View File

@@ -1,158 +1,233 @@
module Linguist
# A collection of simple heuristics that can be used to better analyze languages.
class Heuristics
ACTIVE = true
# Public: Given an array of String language names,
# apply heuristics against the given data and return an array
# of matching languages, or nil.
# Public: Use heuristics to detect language of the blob.
#
# data - Array of tokens or String data to analyze.
# languages - Array of language name Strings to restrict to.
# blob - An object that quacks like a blob.
# possible_languages - Array of Language objects
#
# Returns an array of Languages or []
def self.find_by_heuristics(data, languages)
if active?
result = []
# Examples
#
# Heuristics.call(FileBlob.new("path/to/file"), [
# Language["Ruby"], Language["Python"]
# ])
#
# Returns an Array of languages, or empty if none matched or were inconclusive.
def self.call(blob, languages)
data = blob.data
if languages.all? { |l| ["Perl", "Prolog"].include?(l) }
result = disambiguate_pl(data)
end
if languages.all? { |l| ["ECL", "Prolog"].include?(l) }
result = disambiguate_ecl(data)
end
if languages.all? { |l| ["IDL", "Prolog"].include?(l) }
result = disambiguate_pro(data)
end
if languages.all? { |l| ["Common Lisp", "OpenCL"].include?(l) }
result = disambiguate_cl(data)
end
if languages.all? { |l| ["Hack", "PHP"].include?(l) }
result = disambiguate_hack(data)
end
if languages.all? { |l| ["Scala", "SuperCollider"].include?(l) }
result = disambiguate_sc(data)
end
if languages.all? { |l| ["AsciiDoc", "AGS Script"].include?(l) }
result = disambiguate_asc(data)
end
if languages.all? { |l| ["FORTRAN", "Forth"].include?(l) }
result = disambiguate_f(data)
end
return result
@heuristics.each do |heuristic|
return Array(heuristic.call(data)) if heuristic.matches?(languages)
end
[] # No heuristics matched
end
# Internal: Define a new heuristic.
#
# languages - String names of languages to disambiguate.
# heuristic - Block which takes data as an argument and returns a Language or nil.
#
# Examples
#
# disambiguate "Perl", "Prolog" do |data|
# if data.include?("use strict")
# Language["Perl"]
# elsif data.include?(":-")
# Language["Prolog"]
# end
# end
#
def self.disambiguate(*languages, &heuristic)
@heuristics << new(languages, &heuristic)
end
# Internal: Array of defined heuristics
@heuristics = []
# Internal
def initialize(languages, &heuristic)
@languages = languages
@heuristic = heuristic
end
# Internal: Check if this heuristic matches the candidate languages.
def matches?(candidates)
candidates.any? && candidates.all? { |l| @languages.include?(l.name) }
end
# Internal: Perform the heuristic
def call(data)
@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"]
elsif /^\s*(# |include|require)\b/.match(data)
Language["BitBake"]
end
end
# .h extensions are ambiguous between C, C++, and Objective-C.
# We want to shortcut look for Objective-C _and_ now C++ too!
#
# Returns an array of Languages or []
def self.disambiguate_c(data)
matches = []
if data.include?("@interface")
matches << Language["Objective-C"]
elsif data.include?("#include <cstdint>")
matches << Language["C++"]
disambiguate "C#", "Smalltalk" do |data|
if /![\w\s]+methodsFor: /.match(data)
Language["Smalltalk"]
elsif /^\s*namespace\s*[\w\.]+\s*{/.match(data) || /^\s*\/\//.match(data)
Language["C#"]
end
matches
end
def self.disambiguate_pl(data)
matches = []
if data.include?("use strict")
matches << Language["Perl"]
disambiguate "Objective-C", "C++", "C" do |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))
Language["C++"]
end
end
disambiguate "Perl", "Perl6", "Prolog" do |data|
if data.include?("use v6")
Language["Perl6"]
elsif data.match(/use strict|use\s+v?5\./)
Language["Perl"]
elsif data.include?(":-")
matches << Language["Prolog"]
Language["Prolog"]
end
matches
end
def self.disambiguate_ecl(data)
matches = []
disambiguate "ECL", "Prolog" do |data|
if data.include?(":-")
matches << Language["Prolog"]
Language["Prolog"]
elsif data.include?(":=")
matches << Language["ECL"]
Language["ECL"]
end
matches
end
def self.disambiguate_pro(data)
matches = []
if (data.include?(":-"))
matches << Language["Prolog"]
disambiguate "IDL", "Prolog" do |data|
if data.include?(":-")
Language["Prolog"]
else
matches << Language["IDL"]
Language["IDL"]
end
matches
end
def self.disambiguate_ts(data)
matches = []
if (data.include?("</translation>"))
matches << Language["XML"]
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
matches << Language["TypeScript"]
Language["Scilab"]
end
matches
end
def self.disambiguate_cl(data)
matches = []
disambiguate "Common Lisp", "OpenCL", "Cool" do |data|
if data.include?("(defun ")
matches << Language["Common Lisp"]
Language["Common Lisp"]
elsif /^class/x.match(data)
Language["Cool"]
elsif /\/\* |\/\/ |^\}/.match(data)
matches << Language["OpenCL"]
Language["OpenCL"]
end
matches
end
def self.disambiguate_r(data)
matches = []
matches << Language["Rebol"] if /\bRebol\b/i.match(data)
matches << Language["R"] if data.include?("<-")
matches
end
def self.disambiguate_hack(data)
matches = []
disambiguate "Hack", "PHP" do |data|
if data.include?("<?hh")
matches << Language["Hack"]
Language["Hack"]
elsif /<?[^h]/.match(data)
matches << Language["PHP"]
Language["PHP"]
end
matches
end
def self.disambiguate_sc(data)
matches = []
if (/\^(this|super)\./.match(data) || /^\s*(\+|\*)\s*\w+\s*{/.match(data) || /^\s*~\w+\s*=\./.match(data))
matches << Language["SuperCollider"]
disambiguate "Scala", "SuperCollider" do |data|
if /\^(this|super)\./.match(data) || /^\s*(\+|\*)\s*\w+\s*{/.match(data) || /^\s*~\w+\s*=\./.match(data)
Language["SuperCollider"]
elsif /^\s*import (scala|java)\./.match(data) || /^\s*val\s+\w+\s*=/.match(data) || /^\s*class\b/.match(data)
Language["Scala"]
end
if (/^\s*import (scala|java)\./.match(data) || /^\s*val\s+\w+\s*=/.match(data) || /^\s*class\b/.match(data))
matches << Language["Scala"]
end
matches
end
def self.disambiguate_asc(data)
matches = []
matches << Language["AsciiDoc"] if /^=+(\s|\n)/.match(data)
matches
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
def self.disambiguate_f(data)
matches = []
disambiguate "FORTRAN", "Forth" do |data|
if /^: /.match(data)
matches << Language["Forth"]
elsif /^([c*][^a-z]| subroutine\s)/i.match(data)
matches << Language["FORTRAN"]
Language["Forth"]
elsif /^([c*][^a-z]| (subroutine|program)\s|\s*!)/i.match(data)
Language["FORTRAN"]
end
matches
end
def self.active?
!!ACTIVE
disambiguate "F#", "Forth", "GLSL" 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)
Language["GLSL"]
end
end
disambiguate "M", "Mathematica", "Matlab", "Mercury", "Objective-C" do |data|
if ObjectiveCRegex.match(data)
Language["Objective-C"]
elsif data.include?(":- module")
Language["Mercury"]
elsif /^\s*;/.match(data)
Language["M"]
elsif /^\s*\(\*/.match(data)
Language["Mathematica"]
elsif /^\s*%/.match(data)
Language["Matlab"]
end
end
disambiguate "Gosu", "JavaScript" do |data|
Language["Gosu"] if /^uses java\./.match(data)
end
disambiguate "LoomScript", "LiveScript" do |data|
if /^\s*package\s*[\w\.\/\*\s]*\s*{/.match(data)
Language["LoomScript"]
else
Language["LiveScript"]
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"]
else
Language["TypeScript"]
end
end
disambiguate "Frege", "Forth", "Text" do |data|
if /^(: |also |new-device|previous )/.match(data)
Language["Forth"]
elsif /^\s*(import|module|package|data|type) /.match(data)
Language["Frege"]
else
Language["Text"]
end
end
end
end

127
lib/linguist/language.rb
View File

@@ -10,6 +10,9 @@ require 'linguist/heuristics'
require 'linguist/samples'
require 'linguist/file_blob'
require 'linguist/blob_helper'
require 'linguist/strategy/filename'
require 'linguist/strategy/modeline'
require 'linguist/shebang'
module Linguist
# Language names that are recognizable by GitHub. Defined languages
@@ -29,13 +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
@@ -91,6 +87,14 @@ module Linguist
language
end
STRATEGIES = [
Linguist::Strategy::Modeline,
Linguist::Shebang,
Linguist::Strategy::Filename,
Linguist::Heuristics,
Linguist::Classifier
]
# Public: Detects the Language of the blob.
#
# blob - an object that includes the Linguist `BlobHelper` interface;
@@ -98,61 +102,22 @@ module Linguist
#
# Returns Language or nil.
def self.detect(blob)
name = blob.name.to_s
# Bail early if the blob is binary or empty.
return nil if blob.likely_binary? || blob.binary? || blob.empty?
# A bit of an elegant hack. If the file is executable but extensionless,
# append a "magic" extension so it can be classified with other
# languages that have shebang scripts.
extensions = FileBlob.new(name).extensions
if extensions.empty? && blob.mode && (blob.mode.to_i(8) & 05) == 05
name += ".script!"
end
# Find languages that match based on filename.
possible_languages = find_by_filename(name)
if possible_languages.length == 1
# Simplest and most common case, we can just return the one match based
# on extension
possible_languages.first
# If there is more than one possible language with that extension (or no
# extension at all, in the case of extensionless scripts), we need to
# continue our detection work
else
# Matches possible_languages.length == 0 || possible_languages.length > 0
data = blob.data
# Check if there's a shebang line and use that as authoritative
if (result = find_by_shebang(data)) && !result.empty?
return result.first
# More than one language with that extension. We need to make a choice.
elsif possible_languages.length > 1
# First try heuristics
possible_language_names = possible_languages.map(&:name)
heuristic_languages = Heuristics.find_by_heuristics(data, possible_language_names)
# If there are multiple possible languages returned from heuristics
# then reduce language candidates for Bayesian classifier here.
if heuristic_languages.size > 1
possible_language_names = heuristic_languages.map(&:name)
end
if heuristic_languages.size == 1
return heuristic_languages.first
# Lastly, fall back to the probabilistic classifier.
elsif classified = Classifier.classify(Samples.cache, data, possible_language_names).first
# Return the actual Language object based of the string language name (i.e., first element of `#classify`)
return Language[classified[0]]
end
# 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
end
end
end.first
end
# Public: Get all Languages
@@ -185,7 +150,7 @@ 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
@@ -229,21 +194,27 @@ module Linguist
@extension_index[extname]
end
# Public: Look up Languages by shebang line.
# DEPRECATED
def self.find_by_shebang(data)
@interpreter_index[Shebang.interpreter(data)]
end
# Public: Look up Languages by interpreter.
#
# data - Array of tokens or String data to analyze.
# interpreter - String of interpreter name
#
# Examples
#
# Language.find_by_shebang("#!/bin/bash\ndate;")
# Language.find_by_interpreter("bash")
# # => [#<Language name="Bash">]
#
# Returns the matching Language
def self.find_by_shebang(data)
@interpreter_index[Linguist.interpreter_from_shebang(data)]
def self.find_by_interpreter(interpreter)
@interpreter_index[interpreter]
end
# Public: Look up Language by its name or lexer.
# Public: Look up Language by its name.
#
# name - The String name of the Language
#
@@ -267,7 +238,7 @@ 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
@@ -279,7 +250,7 @@ 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
@@ -293,8 +264,12 @@ module Linguist
# Public: A List of languages compatible with Ace.
#
# TODO: Remove this method in a 5.x release. Every language now needs an ace_mode
# key, so this function isn't doing anything unique anymore.
#
# Returns an Array of Languages.
def self.ace_modes
warn "This method will be deprecated in a future 5.x release. Every language now has an `ace_mode` set."
@ace_modes ||= all.select(&:ace_mode).sort_by { |lang| lang.name.downcase }
end
@@ -395,11 +370,6 @@ 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
@@ -515,16 +485,6 @@ 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
@@ -568,7 +528,7 @@ module Linguist
if extnames = extensions[name]
extnames.each do |extname|
if !options['extensions'].include?(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!'
options['extensions'] << extname
end
@@ -600,7 +560,6 @@ 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'],

602
lib/linguist/languages.yml
View File

File diff suppressed because it is too large Load Diff

10
lib/linguist/lazy_blob.rb
View File

@@ -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)
@@ -37,6 +37,14 @@ 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)

2
lib/linguist/popular.yml
View File

@@ -3,7 +3,6 @@
# This file should only be edited by GitHub staff
- ActionScript
- Bash
- C
- C#
- C++
@@ -27,3 +26,4 @@
- SQL
- Scala
- Scheme
- Shell

9
lib/linguist/repository.rb
View File

@@ -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

44
lib/linguist/samples.rb
View File

@@ -6,6 +6,7 @@ end
require 'linguist/md5'
require 'linguist/classifier'
require 'linguist/shebang'
module Linguist
# Model for accessing classifier training data.
@@ -33,10 +34,6 @@ module Linguist
Dir.entries(ROOT).sort!.each do |category|
next if category == '.' || category == '..'
# Skip text and binary for now
# Possibly reconsider this later
next if category == 'Text' || category == 'Binary'
dirname = File.join(ROOT, category)
Dir.entries(dirname).each do |filename|
next if filename == '.' || filename == '..'
@@ -61,7 +58,7 @@ module Linguist
yield({
:path => path,
:language => category,
:interpreter => Linguist.interpreter_from_shebang(File.read(path)),
:interpreter => Shebang.interpreter(File.read(path)),
:extname => File.extname(filename)
})
end
@@ -114,41 +111,4 @@ module Linguist
db
end
end
# Used to retrieve the interpreter from the shebang line of a file's
# data.
def self.interpreter_from_shebang(data)
lines = data.lines.to_a
if lines.any? && (match = lines[0].match(/(.+)\n?/)) && (bang = match[0]) =~ /^#!/
bang.sub!(/^#! /, '#!')
tokens = bang.split(' ')
pieces = tokens.first.split('/')
if pieces.size > 1
script = pieces.last
else
script = pieces.first.sub('#!', '')
end
script = script == 'env' ? tokens[1] : script
# If script has an invalid shebang, we might get here
return unless script
# "python2.6" -> "python2"
script.sub! $1, '' if script =~ /(\.\d+)$/
# Check for multiline shebang hacks that call `exec`
if script == 'sh' &&
lines[0...5].any? { |l| l.match(/exec (\w+).+\$0.+\$@/) }
script = $1
end
File.basename(script)
else
nil
end
end
end

53
lib/linguist/shebang.rb Normal file
View File

@@ -0,0 +1,53 @@
module Linguist
class Shebang
# Public: Use shebang to detect language of the blob.
#
# blob - An object that quacks like a blob.
#
# Examples
#
# Shebang.call(FileBlob.new("path/to/file"))
#
# Returns an Array with one Language if the blob has a shebang with a valid
# interpreter, or empty if there is no shebang.
def self.call(blob, _ = nil)
Language.find_by_interpreter interpreter(blob.data)
end
# Public: Get the interpreter from the shebang
#
# Returns a String or nil
def self.interpreter(data)
shebang = data.lines.first
# 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(' ')
# There was nothing after the #!
return if tokens.empty?
# Get the name of the interpreter
script = File.basename(tokens.first)
# Get next argument if interpreter was /usr/bin/env
script = tokens[1] if script == 'env'
# Interpreter was /usr/bin/env with no arguments
return unless script
# "python2.6" -> "python2"
script.sub! /(\.\d+)$/, ''
# Check for multiline shebang hacks that call `exec`
if script == 'sh' &&
data.lines.first(5).any? { |l| l.match(/exec (\w+).+\$0.+\$@/) }
script = $1
end
File.basename(script)
end
end
end

20
lib/linguist/strategy/filename.rb Normal file
View File

@@ -0,0 +1,20 @@
module Linguist
module Strategy
# Detects language based on filename and/or extension
class Filename
def self.call(blob, _)
name = blob.name.to_s
# A bit of an elegant hack. If the file is executable but extensionless,
# append a "magic" extension so it can be classified with other
# languages that have shebang scripts.
extensions = FileBlob.new(name).extensions
if extensions.empty? && blob.mode && (blob.mode.to_i(8) & 05) == 05
name += ".script!"
end
Language.find_by_filename(name)
end
end
end
end

30
lib/linguist/strategy/modeline.rb Normal file
View File

@@ -0,0 +1,30 @@
module Linguist
module Strategy
class Modeline
EmacsModeline = /-\*-\s*mode:\s*(\w+);?\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

3
lib/linguist/tokenizer.rb
View File

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

23
lib/linguist/vendor.yml
View File

@@ -32,6 +32,7 @@
# Erlang bundles
- ^rebar$
- erlang.mk
# Go dependencies
- Godeps/_workspace/
@@ -39,24 +40,27 @@
# Minified JavaScript and CSS
- (\.|-)min\.(js|css)$
# Stylesheets imported from packages
- ([^\s]*)import\.(css|less|scss|styl)$
# Bootstrap css and js
- (^|/)bootstrap([^.]*)\.(js|css)$
- (^|/)bootstrap([^.]*)\.(js|css|less|scss|styl)$
- (^|/)custom\.bootstrap([^\s]*)(js|css|less|scss|styl)$
# Font Awesome
- font-awesome.css
- (^|/)font-awesome\.(css|less|scss|styl)$
# Foundation css
- foundation.css
- (^|/)foundation\.(css|less|scss|styl)$
# Normalize.css
- normalize.css
- (^|/)normalize\.(css|less|scss|styl)$
# Bourbon SCSS
- (^|/)[Bb]ourbon/.*\.css$
- (^|/)[Bb]ourbon/.*\.scss$
# Bourbon css
- (^|/)[Bb]ourbon/.*\.(css|less|scss|styl)$
# Animate.css
- animate.css
- (^|/)animate\.(css|less|scss|styl)$
# Vendored dependencies
- third[-_]?party/
@@ -232,9 +236,6 @@
# .DS_Store's
- .[Dd][Ss]_[Ss]tore$
# Mercury --use-subdirs
- Mercury/
# R packages
- ^vignettes/
- ^inst/extdata/

2
lib/linguist/version.rb
View File

@@ -1,3 +1,3 @@
module Linguist
VERSION = "4.0.3"
VERSION = "4.4.0"
end