github-linguist 7.6.0 → 7.10.0

data/grammars/text.xml.svg.json

@@ -50,7 +50,7 @@
     },
     "attrValueCSS": {
       "begin": "(?i)(?\u003c=style)\\G\\s*(=)",
-      "end": "(?=\\s*(?:[%?/]?\u003e))|([^\\s\"'\u003e]+)|(?\u003c=[\"'])",
+      "end": "(?=\\s*(?:[%?/]?\u003e))|(?\u003c=[\"'])([^\\s\u003e]*)|(?:\\G|^)\\s*([^\\s\"'\u003e]+)",
       "patterns": [
         {
           "match": "(?:\\G|^)\\s+(?!/?\u003e)"
@@ -129,6 +129,9 @@
       },
       "endCaptures": {
         "1": {
+          "name": "invalid.illegal.syntax.xml.svg"
+        },
+        "2": {
           "name": "string.unquoted.xml.svg",
           "patterns": [
             {
@@ -140,7 +143,7 @@
     },
     "attrValuePlain": {
       "begin": "\\s*(=)",
-      "end": "(?=\\s*(?:[%?/]?\u003e))|([^\\s\"'\u003e]+)|(?\u003c=[\"'])",
+      "end": "(?=\\s*(?:[%?/]?\u003e))|(?\u003c=[\"'])([^\\s\u003e]*)|(?:\\G|^)\\s*([^\\s\"'\u003e]+)",
       "patterns": [
         {
           "match": "(?:\\G|^)\\s+(?!/?\u003e)"
@@ -193,6 +196,9 @@
       },
       "endCaptures": {
         "1": {
+          "name": "invalid.illegal.syntax.xml.svg"
+        },
+        "2": {
           "name": "string.unquoted.xml.svg"
         }
       }

data/grammars/version

@@ -1 +1 @@
-7.6.0
+7.10.0

data/lib/linguist/VERSION

@@ -1 +1 @@
-7.6.0
+7.10.0

data/lib/linguist/blob_helper.rb

@@ -270,8 +270,10 @@ module Linguist
           # also--importantly--without having to duplicate many (potentially
           # large) strings.
           begin
-
-            data.split(encoded_newlines_re, -1)
+            # `data` is split after having its last `\n` removed by
+            # chomp (if any). This prevents the creation of an empty
+            # element after the final `\n` character on POSIX files.
+            data.chomp.split(encoded_newlines_re, -1)
           rescue Encoding::ConverterNotFoundError
             # The data is not splittable in the detected encoding.  Assume it's
             # one big line.

data/lib/linguist/file_blob.rb

@@ -35,7 +35,7 @@ module Linguist
     #
     # Returns a String.
     def data
-      @data ||= File.read(@fullpath)
+      @data ||= File.read(@fullpath, :encoding => "ASCII-8BIT")
     end
 
     # Public: Get byte size

data/lib/linguist/generated.rb

@@ -62,7 +62,9 @@ module Linguist
       node_modules? ||
       go_vendor? ||
       go_lock? ||
+      esy_lock? ||
       npm_shrinkwrap_or_package_lock? ||
+      generated_yarn_plugnplay? ||
       godeps? ||
       generated_by_zephir? ||
       minified_files? ||
@@ -91,7 +93,11 @@ module Linguist
       generated_grpc_cpp? ||
       generated_dart? ||
       generated_perl_ppport_header? ||
-      generated_gamemakerstudio?
+      generated_gamemakerstudio? ||
+      generated_gimp? ||
+      generated_visualstudio6? ||
+      generated_haxe? ||
+      generated_jooq?
     end
 
     # Internal: Is the blob an Xcode file?
@@ -118,7 +124,14 @@ module Linguist
       !!name.match(/(^|\/)Carthage\/Build\//)
     end
 
-    # Internal: Is the blob minified files?
+    # Internal: Does extname indicate a filetype which is commonly minified?
+    #
+    # Returns true or false.
+    def maybe_minified?
+      ['.js', '.css'].include? extname.downcase
+    end
+
+    # Internal: Is the blob a minified file?
     #
     # Consider a file minified if the average line length is
     # greater then 110c.
@@ -127,32 +140,29 @@ module Linguist
     #
     # Returns true or false.
     def minified_files?
-      return unless ['.js', '.css'].include? extname
-      if lines.any?
+      if maybe_minified? and lines.any?
         (lines.inject(0) { |n, l| n += l.length } / lines.length) > 110
       else
         false
       end
     end
 
-    # Internal: Does the blob contain a source map reference?
+    # Internal: Does the blob contain a source-map reference?
     #
-    # We assume that if one of the last 2 lines starts with a source map
+    # We assume that if one of the last 2 lines starts with a source-map
     # reference, then the current file was generated from other files.
     #
     # We use the last 2 lines because the last line might be empty.
     #
-    # We only handle JavaScript, no CSS support yet.
-    #
     # Returns true or false.
     def has_source_map?
-      return false unless extname.downcase == '.js'
-      lines.last(2).any? { |line| line.start_with?('//# sourceMappingURL') }
+      return false unless maybe_minified?
+      lines.last(2).any? { |l| l.match(/^\/[*\/][\#@] source(?:Mapping)?URL|sourceURL=/) }
     end
 
-    # Internal: Is the blob a generated source map?
+    # Internal: Is the blob a generated source-map?
     #
-    # Source Maps usually have .css.map or .js.map extensions. In case they
+    # 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.
@@ -293,10 +303,10 @@ module Linguist
       return false unless extname == '.go'
       return false unless lines.count > 1
 
-      return lines[0].include?("Code generated by")
+      return lines.first(40).any? { |l| l.include? "Code generated by" }
     end
 
-    PROTOBUF_EXTENSIONS = ['.py', '.java', '.h', '.cc', '.cpp', '.m', '.rb']
+    PROTOBUF_EXTENSIONS = ['.py', '.java', '.h', '.cc', '.cpp', '.m', '.rb', '.php']
 
     # Internal: Is the blob a C++, Java or Python source file generated by the
     # Protocol Buffer compiler?
@@ -306,7 +316,7 @@ module Linguist
       return false unless PROTOBUF_EXTENSIONS.include?(extname)
       return false unless lines.count > 1
 
-      return lines[0].include?("Generated by the protocol buffer compiler.  DO NOT EDIT!")
+      return lines.first(3).any? { |l| l.include?("Generated by the protocol buffer compiler.  DO NOT EDIT!") }
     end
 
     # Internal: Is the blob a Javascript source file generated by the
@@ -363,11 +373,25 @@ module Linguist
       !!name.match(/(Gopkg|glide)\.lock/)
     end
 
+    # Internal: Is the blob a generated esy lock file?
+    #
+    # Returns true or false.
+    def esy_lock?
+      !!name.match(/(^|\/)(\w+\.)?esy.lock$/)
+    end
+
     # Internal: Is the blob a generated npm shrinkwrap or package lock file?
     #
     # Returns true or false.
     def npm_shrinkwrap_or_package_lock?
-      name.match(/npm-shrinkwrap\.json/) || name.match(/package-lock\.json/)
+      !!name.match(/npm-shrinkwrap\.json/) || !!name.match(/package-lock\.json/)
+    end
+
+    # Internal: Is the blob a generated Yarn Plug'n'Play?
+    #
+    # Returns true or false.
+    def generated_yarn_plugnplay?
+      !!name.match(/(^|\/)\.pnp\.(c|m)?js$/)
     end
 
     # Internal: Is the blob part of Godeps/,
@@ -426,7 +450,7 @@ module Linguist
     #
     # Returns true or false.
     def pipenv_lock?
-      name.match(/Pipfile\.lock/)
+      !!name.match(/Pipfile\.lock/)
     end
 
     # Internal: Is it a KiCAD or GFortran module file?
@@ -591,6 +615,38 @@ module Linguist
              lines[0] =~ /^\d\.\d\.\d.+\|\{/
     end
 
+    # Internal: Is this a generated GIMP C image file?
+    #
+    # GIMP saves C sources with one of two comment forms:
+    # * `/* GIMP RGB C-Source image dump (<filename>.c) */` (C source export)
+    # * `/*  GIMP header image file format (RGB): <filename>.h  */` (Header export)
+    #
+    # Return true or false
+    def generated_gimp?
+      return false unless ['.c', '.h'].include? extname
+      return false unless lines.count > 0
+      return lines[0].match(/\/\* GIMP [a-zA-Z0-9\- ]+ C\-Source image dump \(.+?\.c\) \*\//) ||
+             lines[0].match(/\/\*  GIMP header image file format \([a-zA-Z0-9\- ]+\)\: .+?\.h  \*\//)
+    end
+
+    # Internal: Is this a generated Microsoft Visual Studio 6.0 build file?
+    #
+    # Return true or false
+    def generated_visualstudio6?
+      return false unless extname.downcase == '.dsp'
+      lines.first(3).any? { |l| l.include? '# Microsoft Developer Studio Generated Build File' }
+    end
+
+    HAXE_EXTENSIONS = ['.js', '.py', '.lua', '.cpp', '.h', '.java', '.cs', '.php']
+
+    # Internal: Is this a generated Haxe-generated source file?
+    #
+    # Return true or false
+    def generated_haxe?
+      return false unless HAXE_EXTENSIONS.include?(extname)
+      return lines.first(3).any? { |l| l.include?("Generated by Haxe") }
+    end
+
     # Internal: Is this a generated HTML file?
     #
     # HTML documents generated by authoring tools often include a
@@ -621,23 +677,35 @@ module Linguist
       return false if matches.empty?
       return matches.map {|x| extract_html_meta(x) }.any? do |attr|
         attr["name"].to_s.downcase == 'generator' &&
-        attr["content"].match(/^
-          ( org \s+ mode
-          | j?latex2html
-          | groff
-          | makeinfo
-          | texi2html
-          ) \b
-        /ix)
+        [attr["content"], attr["value"]].any? do |cv|
+          !cv.nil? &&
+          cv.match(/^
+            ( org \s+ mode
+            | j?latex2html
+            | groff
+            | makeinfo
+            | texi2html
+            | ronn
+            ) \b
+          /ix)
+        end
       end
     end
 
+    # Internal: Is this a generated jOOQ file?
+    #
+    # Return true or false
+    def generated_jooq?
+      return false unless extname.downcase == '.java'
+      lines.first(2).any? { |l| l.include? 'This file is generated by jOOQ.' }
+    end
+
     # Internal: Extract a Hash of name/content pairs from an HTML <meta> tag
     def extract_html_meta(match)
       (match.last.sub(/\/\Z/, "").strip.scan(/
-        (?<=^|\s)        # Check for preceding whitespace
-        (name|content)   # Attribute names we're interested in
-        \s* = \s*        # Key-value separator
+        (?<=^|\s)              # Check for preceding whitespace
+        (name|content|value)   # Attribute names we're interested in
+        \s* = \s*              # Key-value separator
 
         # Attribute value
         ( "[^"]+"        # name="value"

data/lib/linguist/heuristics.yml

@@ -58,6 +58,10 @@ disambiguations:
     pattern: '^[=-]+(\s|\n)|{{[A-Za-z]'
   - language: AGS Script
     pattern: '^(\/\/.+|((import|export)\s+)?(function|int|float|char)\s+((room|repeatedly|on|game)_)?([A-Za-z]+[A-Za-z_0-9]+)\s*[;\(])'
+- extensions: ['.asm']
+  rules:
+  - language: Motorola 68K Assembly
+    named_pattern: m68k
 - extensions: ['.asy']
   rules:
   - language: LTspice Symbol
@@ -113,6 +117,12 @@ disambiguations:
     #  : dependency
     # path/file.ext1 : some/path/../file.ext2
     pattern: '([\/\\].*:\s+.*\s\\$|: \\$|^[ %]:|^[\w\s\/\\.]+\w+\.\w+\s*:\s+[\w\s\/\\.]+\w+\.\w+)'
+- extensions: ['.dsp']
+  rules:
+  - language: Microsoft Developer Studio Project
+    pattern: '# Microsoft Developer Studio Generated Build File'
+  - language: Faust
+    pattern: '\bprocess\s*[(=]|\b(library|import)\s*\(\s*"|\bdeclare\s+(name|version|author|copyright|license)\s+"'
 - extensions: ['.ecl']
   rules:
   - language: ECLiPSe
@@ -173,6 +183,8 @@ disambiguations:
   - language: Game Maker Language
 - extensions: ['.gs']
   rules:
+  - language: GLSL
+    pattern: '^#version\s+[0-9]+\b'
   - language: Gosu
     pattern: '^uses java\.'
 - extensions: ['.h']
@@ -185,6 +197,12 @@ disambiguations:
   rules:
   - language: Hack
     pattern: '<\?hh'
+- extensions: ['.i']
+  rules:
+  - language: Motorola 68K Assembly
+    named_pattern: m68k
+  - language: SWIG
+    pattern: '^[ \t]*%[a-z_]+\b|^%[{}]$'
 - extensions: ['.ice']
   rules:
   - language: JSON
@@ -192,10 +210,19 @@ disambiguations:
   - language: Slice
 - extensions: ['.inc']
   rules:
+  - language: Motorola 68K Assembly
+    named_pattern: m68k
   - language: PHP
     pattern: '^<\?(?:php)?'
   - language: SourcePawn
     pattern: '^public\s+(?:SharedPlugin(?:\s+|:)__pl_\w+\s*=(?:\s*{)?|(?:void\s+)?__pl_\w+_SetNTVOptional\(\)(?:\s*{)?)'
+  - language: NASL
+    pattern:
+    - '^\s*include\s*\(\s*(?:"|'')[\\/\w\-\.:\s]+\.(?:nasl|inc)\s*(?:"|'')\s*\)\s*;'
+    - '^\s*(?:global|local)_var\s+(?:\w+(?:\s*=\s*[\w\-"'']+)?\s*)(?:,\s*\w+(?:\s*=\s*[\w\-"'']+)?\s*)*+\s*;'
+    - '^\s*namespace\s+\w+\s*{'
+    - '^\s*object\s+\w+\s*(?:extends\s+\w+(?:::\w+)?)?\s*{'
+    - '^\s*(?:public\s+|private\s+|\s*)function\s+\w+\s*\([\w\s,]*\)\s*{'
   - language: POV-Ray SDL
     pattern: '^\s*#(declare|local|macro|while)\s'
 - extensions: ['.l']
@@ -237,6 +264,10 @@ disambiguations:
     pattern: '^\s*%'
   - language: Limbo
     pattern: '^\w+\s*:\s*module\s*{'
+- extensions: ['.mask']
+  rules:
+  - language: Unity3d Asset
+    pattern: 'tag:unity3d.com'
 - extensions: ['.md']
   rules:
   - language: Markdown
@@ -285,6 +316,19 @@ disambiguations:
   - language: NL
     pattern: '^(b|g)[0-9]+ '
   - language: NewLisp
+- extensions: ['.odin']
+  rules:
+  - language: Object Data Instance Notation
+    pattern: '(?:^|<)\s*[A-Za-z0-9_]+\s*=\s*<'
+  - language: Odin
+    pattern: 'package\s+\w+|\b(?:im|ex)port\s*"[\w:./]+"|\w+\s*::\s*(?:proc|struct)\s*\(|^\s*//\s'
+- extensions: ['.p']
+  rules:
+  - language: Gnuplot
+    pattern:
+    - '^s?plot\b'
+    - '^set\s+(term|terminal|out|output|[xy]tics|[xy]label|[xy]range|style)\b'
+  - language: OpenEdge ABL
 - extensions: ['.php']
   rules:
   - language: Hack
@@ -297,7 +341,7 @@ disambiguations:
     pattern: '^[^#]*:-'
   - language: Perl
     named_pattern: perl5
-  - language: Perl 6
+  - language: Raku
     named_pattern: perl6
 - extensions: ['.plist']
   rules:
@@ -308,7 +352,7 @@ disambiguations:
   rules:
   - language: Perl
     named_pattern: perl5
-  - language: Perl 6
+  - language: Raku
     named_pattern: perl6
   - language: X PixMap
     pattern: '^\s*\/\* XPM \*\/'
@@ -325,6 +369,8 @@ disambiguations:
     pattern: '^\s+\w+\s+=>\s'
 - extensions: ['.pro']
   rules:
+  - language: Proguard
+    pattern: '^-(include\b.*\.pro$|keep\b|keepclassmembers\b|keepattributes\b)'
   - language: Prolog
     pattern: '^[^\[#]+:-'
   - language: INI
@@ -384,6 +430,10 @@ disambiguations:
     pattern: '^(use |fn |mod |pub |macro_rules|impl|#!?\[)'
   - language: RenderScript
     pattern: '#include|#pragma\s+(rs|version)|__attribute__'
+- extensions: ['.s']
+  rules:
+  - language: Motorola 68K Assembly
+    named_pattern: m68k
 - extensions: ['.sc']
   rules:
   - language: SuperCollider
@@ -417,7 +467,7 @@ disambiguations:
   rules:
   - language: Perl
     named_pattern: perl5
-  - language: Perl 6
+  - language: Raku
     named_pattern: perl6
   - language: Turing
     pattern: '^\s*%[ \t]+|^\s*var\s+\w+(\s*:\s*\w+)?\s*:=\s*\w+'
@@ -456,7 +506,7 @@ disambiguations:
   rules:
   - language: Vim script
     pattern: '^UseVimball'
-  - language: Visual Basic
+  - language: VBA
 - extensions: ['.w']
   rules:
   - language: OpenEdge ABL
@@ -465,6 +515,8 @@ disambiguations:
     pattern: '^@(<|\w+\.)'
 - extensions: ['.x']
   rules:
+  - language: DirectX 3D File
+    pattern:  '^xof 030(2|3)(?:txt|bin|tzip|bzip)\b'
   - language: RPC
     pattern: '\b(program|version)\s+\w+\s*{|\bunion\s+\w+\s+switch\s*\('
   - language: Logos
@@ -487,6 +539,14 @@ named_patterns:
   - 'std::\w+'
   fortran: '^(?i:[c*][^abd-z]|      (subroutine|program|end|data)\s|\s*!)'
   key_equals_value: '^[^#!;][^=]*='
+  m68k:
+  - '(?im)\bmoveq(?:\.l)?\s+#(?:\$-?[0-9a-f]{1,3}|%[0-1]{1,8}|-?[0-9]{1,3}),\s*d[0-7]\b'
+  - '(?im)^\s*move(?:\.[bwl])?\s+(?:sr|usp),\s*[^\s]+'
+  - '(?im)^\s*move\.[bwl]\s+.*\b[ad]\d'
+  - '(?im)^\s*movem\.[bwl]\b'
+  - '(?im)^\s*move[mp](?:\.[wl])?\b'
+  - '(?im)^\s*btst\b'
+  - '(?im)^\s*dbra\b'
   objectivec: '^\s*(@(interface|class|protocol|property|end|synchronised|selector|implementation)\b|#import\s+.+\.h[">])'
   perl5: '\buse\s+(?:strict\b|v?5\.)'
   perl6: '^\s*(?:use\s+v6\b|\bmodule\b|\b(?:my\s+)?class\b)'