line-detector 0.1 → 0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 2cb9c2cf09d692bc7f86377e5968e9d84dec521e
-  data.tar.gz: 0a2a951cdc2d1ccb5b4448a2d84f1178597da4c4
+  metadata.gz: cc7429071fc4b90421939e9f2efa635af6706ffd
+  data.tar.gz: 62d454bedea6b374d17ff18aa29aa5b7148ca624
 SHA512:
-  metadata.gz: dcc51273493ad34388934add829d171db2d2c3f34ad5a5637e161c9c6b88e606d9f1949d54a5a4468317e24f06b835143c27346f3deeafaa656fd6d3db47d30f
-  data.tar.gz: e98a57f9fa0f44f05a3987551ebaf95890db6e7318418e13e27ad97f8e46a65e01a8101a2f5efa46ce2af2ec094bc404804fecc8a3cb405c5160440b76506181
+  metadata.gz: d1ef12f759a63cf88182305b5eb334d7fff9accf0fe08968084e7ac9a35be8844962a7ab473793b741477dc26b4887dc59b714255a8fd8ec5c78c343448834ed
+  data.tar.gz: 0097e756cd16d68987f7a56fabb23f4cde41050de9d63dc86408f97d8b86c6634a35654cb423311c06b9188c07a39b3b79f2a2cd5d32e228a38287ffe95d110a

data/lib/line-detector.rb

@@ -1,5 +1,4 @@
 require 'rubygems'
-require 'ptools'
 
 require 'version'
 
@@ -7,34 +6,78 @@ require 'version'
 # LineDetector - line ending detector
 #
 module LineDetector
-  # Assumes input is not multi-line
-  def self.detect_line_ending_of_line(line)
-    if line =~ /\r\n/
-      :crlf
-    elsif line =~ /\n/
-      :lf
-    elsif line =~ /\r/
-      :cr
-    else
-      :none
-    end
-  end
+  EOL_CHARACTERS = [
+    "\r",
+    "\n",
+    "\v",
+    "\f",
+    "\u2028",
+    "\u2029",
+    "\u0085"
+  ]
+
+  EOL2NAME = {
+    "\r\n" => :crlf,
+    "\n\r" => :lfcr,
+    "\n" => :lf,
+    "\r" => :cr,
+    "\v" => :vt,
+    "\f" => :ff,
+    "\u2028" => :ls,
+    "\u2029" => :ps,
+    "\u0085" => :nel
+  }
+
+  NAME2EOL = EOL2NAME.invert
 
+  #
+  # Detect line ending format of arbitrary text
+  #
+  # If text uses multiple line ending formats,
+  # Returns :mix.
+  #
   def self.detect_line_ending_of_text(text)
-    line_endings = text.each_line.map do |line|
-      detect_line_ending_of_line(line)
-    end.uniq
+    line_endings = text.split(/[^#{EOL_CHARACTERS.join('')}]/)
+      .reject { |ending| ending == '' }
+      .map { |ending| ending.gsub(/(.+?)(\1)+/m, '\1') }
+      .uniq
 
-    if line_endings == []
+    len = line_endings.length
+
+    if len == 0
       :none
-    elsif line_endings.length == 1
-      line_endings.first
+    elsif len == 1
+      EOL2NAME[line_endings.first] || :unknown
     else
       :mix
     end
   end
 
+  #
+  # Detect line ending format of a file
+  #
+  # Assumes file is a text file.
+  #
   def self.detect_line_ending_of_file(filename)
-    detect_line_ending_of_text(open(filename))
+    detect_line_ending_of_text(open(filename).read)
+  end
+
+  #
+  # A more capable version of String#lines,
+  # that handles some of the more obscure line ending formats.
+  #
+  # If line ending format cannot be determined,
+  # returns :unknown.
+  #
+  def self.lines(text)
+    line_ending = detect_line_ending_of_text(text)
+
+    if line_ending == :unknown
+      line_ending
+    elsif line_ending == :none
+      [text]
+    else
+      text.split(NAME2EOL[line_ending])
+    end
   end
 end

data/lib/version.rb

@@ -2,5 +2,5 @@
 # LineDetector
 #
 module LineDetector
-  VERSION = '0.1'
+  VERSION = '0.2'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: line-detector
 version: !ruby/object:Gem::Version
-  version: '0.1'
+  version: '0.2'
 platform: ruby
 authors:
 - Andrew Pennebaker
@@ -10,20 +10,6 @@ bindir: bin
 cert_chain: []
 date: 2014-09-04 00:00:00.000000000 Z
 dependencies:
-- !ruby/object:Gem::Dependency
-  name: ptools
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: '1.2'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ~>
-      - !ruby/object:Gem::Version
-        version: '1.2'
 - !ruby/object:Gem::Dependency
   name: rake
   requirement: !ruby/object:Gem::Requirement