-1

我有一些我一直在使用的简单文本处理脚本，我想将这些脚本翻译成 Ruby，以熟悉该语言。

这是我无法运行的第一个脚本：

#!/usr/bin/env ruby

@text = ARGF.read

@replacements = [{:from=>"—", :to=>". "}, {:from=>"ffl", :to=>"ffl"}, {:from=>"ffi", :to=>"ffi"}, {:from=>"fi", :to=>"fi"}, {:from=>"fl", :to=>"fl"}, {:from=>"ff", :to=>"ff"}, {:from=>"æ", :to=>"ae"}, {:from=>"é", :to=>"e"}, {:from=>"Ç", :to=>"s"}, {:from=>"ü", :to=>"u"}, {:from=>"â", :to=>"a"}, {:from=>"ä", :to=>"a"}, {:from=>"à", :to=>"a"}, {:from=>"å", :to=>"a"}, {:from=>"ç", :to=>"s"}, {:from=>"ê", :to=>"e"}, {:from=>"ë", :to=>"e"}, {:from=>"è", :to=>"e"}, {:from=>"ï", :to=>"i"}, {:from=>"î", :to=>"i"}, {:from=>"ì", :to=>"i"}, {:from=>"Ä", :to=>"a"}, {:from=>"Å", :to=>"a"}, {:from=>"É", :to=>"e"}, {:from=>"ô", :to=>"oh"}, {:from=>"ö", :to=>"oe"}, {:from=>"ò", :to=>"o"}, {:from=>"û", :to=>"uu"}, {:from=>"ù", :to=>"u"}, {:from=>"ÿ", :to=>"o"}, {:from=>"Ö", :to=>"o"}, {:from=>"Ü", :to=>"u"}, {:from=>"á", :to=>"ah"}, {:from=>"í", :to=>"ee"}, {:from=>"ó", :to=>"oh"}, {:from=>"ú", :to=>"uu"}, {:from=>"ñ", :to=>"ny"}, {:from=>"Ñ", :to=>"ny"}] 

@replacements.each do |pair|
  @text.gsub!(/#{pair[:from]}/, pair[:to])
end

puts @text

这是我得到的错误：

/home/alec/.bei/under-boac:5: invalid multibyte char (US-ASCII)
/home/alec/.bei/under-boac:5: invalid multibyte char (US-ASCII)
/home/alec/.bei/under-boac:5: syntax error, unexpected $end, expecting '}'
@replacements = [{:from=>"—", :to=>". "}, {:from=>"ffl"...
                            ^

我的部分内容基于“ Ruby 中 STDIN 的最佳实践？ ”。

4

4 回答 4

4

这是您的基本代码，已重新格式化以提高可读性：

@replacements = [
  { :from => "—", :to => ". "  },
  { :from => "ffl", :to => "ffl" },
  { :from => "ffi", :to => "ffi" },
  { :from => "fi", :to => "fi"  },
  { :from => "fl", :to => "fl"  },
  { :from => "ff", :to => "ff"  },
  { :from => "æ", :to => "ae"  },
  { :from => "é", :to => "e"   },
  { :from => "Ç", :to => "s"   },
  { :from => "ü", :to => "u"   },
  { :from => "â", :to => "a"   },
  { :from => "ä", :to => "a"   },
  { :from => "à", :to => "a"   },
  { :from => "å", :to => "a"   },
  { :from => "ç", :to => "s"   },
  { :from => "ê", :to => "e"   },
  { :from => "ë", :to => "e"   },
  { :from => "è", :to => "e"   },
  { :from => "ï", :to => "i"   },
  { :from => "î", :to => "i"   },
  { :from => "ì", :to => "i"   },
  { :from => "Ä", :to => "a"   },
  { :from => "Å", :to => "a"   },
  { :from => "É", :to => "e"   },
  { :from => "ô", :to => "oh"  },
  { :from => "ö", :to => "oe"  },
  { :from => "ò", :to => "o"   },
  { :from => "û", :to => "uu"  },
  { :from => "ù", :to => "u"   },
  { :from => "ÿ", :to => "o"   },
  { :from => "Ö", :to => "o"   },
  { :from => "Ü", :to => "u"   },
  { :from => "á", :to => "ah"  },
  { :from => "í", :to => "ee"  },
  { :from => "ó", :to => "oh"  },
  { :from => "ú", :to => "uu"  },
  { :from => "ñ", :to => "ny"  },
  { :from => "Ñ", :to => "ny"  }
] 

@replacements.each do |pair|
  @text.gsub!( /#{ pair[:from] }/, pair[:to] )
end

这可以简化，哈希组合成一个大的，顺便说一句，应该使用作为哈希：

# encoding: utf-8

@replacements = {
  "—" => ". "  ,
  "ffl" => "ffl" ,
  "ffi" => "ffi" ,
  "fi" => "fi"  ,
  "fl" => "fl"  ,
  "ff" => "ff"  ,
  "æ" => "ae"  ,
  "é" => "e"   ,
  "Ç" => "s"   ,
  "ü" => "u"   ,
  "â" => "a"   ,
  "ä" => "a"   ,
  "à" => "a"   ,
  "å" => "a"   ,
  "ç" => "s"   ,
  "ê" => "e"   ,
  "ë" => "e"   ,
  "è" => "e"   ,
  "ï" => "i"   ,
  "î" => "i"   ,
  "ì" => "i"   ,
  "Ä" => "a"   ,
  "Å" => "a"   ,
  "É" => "e"   ,
  "ô" => "oh"  ,
  "ö" => "oe"  ,
  "ò" => "o"   ,
  "û" => "uu"  ,
  "ù" => "u"   ,
  "ÿ" => "o"   ,
  "Ö" => "o"   ,
  "Ü" => "u"   ,
  "á" => "ah"  ,
  "í" => "ee"  ,
  "ó" => "oh"  ,
  "ú" => "uu"  ,
  "ñ" => "ny"  ,
  "Ñ" => "ny"  
}
@replacements.each do |k,v|
  @text.gsub!(k, v)
end

注意：使用“编码注释”来帮助 Ruby 理解字符的编码。

但是使用相同的哈希，该循环可以减少到这个，它运行得非常快：

@text.gsub!(Regexp.union(@replacements.keys), @replacements)
于 2013-01-09T20:34:55.810 回答
2

如果您在源文件中使用非 7 位 ASCII，则需要添加标头来标识使用的字符集。例子：

#!/usr/bin/env ruby
# encoding: utf-8

此外，您在这里做的事情效率极低，如果您构造正确的正则表达式，当您可以在 O(1) 中执行此操作时，需要对文件进行 O(N) 传递：

replace_map = Hash[@replacements.collect { |r| [ r[:from], r[:to] ] }]
replace_regex = Regexp.new("(#{replace_map.keys.collect { |r| Regexp.escape(r) }.join('|')})")

@text.gsub!(replace_regexp) do |s|
  replace_map[s[1]]
end

:from使用键/值映射而不是那里的奇怪/:to配对会容易得多。

于 2013-01-09T20:17:13.460 回答
2 
#encoding: utf-8

@text = "öbñ"
@replacements = [{:from=>"—", :to=>". "}, {:from=>"ffl", :to=>"ffl"}, {:from=>"ffi", :to=>"ffi"}, {:from=>"fi", :to=>"fi"}, {:from=>"fl", :to=>"fl"}, {:from=>"ff", :to=>"ff"}, {:from=>"æ", :to=>"ae"}, {:from=>"é", :to=>"e"}, {:from=>"Ç", :to=>"s"}, {:from=>"ü", :to=>"u"}, {:from=>"â", :to=>"a"}, {:from=>"ä", :to=>"a"}, {:from=>"à", :to=>"a"}, {:from=>"å", :to=>"a"}, {:from=>"ç", :to=>"s"}, {:from=>"ê", :to=>"e"}, {:from=>"ë", :to=>"e"}, {:from=>"è", :to=>"e"}, {:from=>"ï", :to=>"i"}, {:from=>"î", :to=>"i"}, {:from=>"ì", :to=>"i"}, {:from=>"Ä", :to=>"a"}, {:from=>"Å", :to=>"a"}, {:from=>"É", :to=>"e"}, {:from=>"ô", :to=>"oh"}, {:from=>"ö", :to=>"oe"}, {:from=>"ò", :to=>"o"}, {:from=>"û", :to=>"uu"}, {:from=>"ù", :to=>"u"}, {:from=>"ÿ", :to=>"o"}, {:from=>"Ö", :to=>"o"}, {:from=>"Ü", :to=>"u"}, {:from=>"á", :to=>"ah"}, {:from=>"í", :to=>"ee"}, {:from=>"ó", :to=>"oh"}, {:from=>"ú", :to=>"uu"}, {:from=>"ñ", :to=>"ny"}, {:from=>"Ñ", :to=>"ny"}] 

# Hammer the @replacements into one Hash like {"—"=>". ", "ffl"=>"ffl"}:
from_to = Hash[@replacements.map{|h| h.values}]
# Generate one Regular expression to catch all keys:
re = Regexp.union(from_to.keys)
# Let gsub do the work in one pass:
@text.gsub!(re, from_to)

这与@tadman 的代码大致相同。#encoding: utf-8应该可以解决您的问题；其余的行阻止扫描文本 38 次。

于 2013-01-09T20:26:55.703 回答
0

你甚至可以简化：

@replacements=Hash[*%w[— .  ffl ffl ffi ffi fi fi fl fl ff ff æ ae é e Ç s ü u â a ä a à a å a ç s ê e ë e è e ï i î i ì i Ä a Å a É e ô oh ö oe ò o û uu ù u ÿ o Ö o Ü u á ah í ee ó oh ú uu ñ ny Ñ ny]]
于 2014-07-09T16:24:15.167 回答