如何替换哈希中包含的值？

Question

I have an array of hashes:

arr = [
  {:key1=>"foo one bar", :key2=>"two", :key3=>"three"},
  {:key1=>"four", :key2=>"five", :key3=>"six"},
  {:key1=>"seven", :key2=>"eight", :key3=>"nine"}
]

and I would like to search through and replace the value of :key1 with "replaced" if :key CONTAINS "one".

The resultant array would then read:

arr = [
  {:key1=>"replaced", ;key2=>"two", :key3=>"three"},
  {:key1=>"four", ;key2=>"five", :key3=>"six"},
  {:key1=>"seven", ;key2=>"eight", :key3=>"nine"}
]

I tried using include? from my previous post and replies in " Replace one matched value in a hash with another " but to no avail.

These will get an EXACT match:

arr.each{|item| item[:key1] = "replaced" if item[:key1]=="one"}

as will:

p arr.each {|x| x[:key1] = "replaced" if x.values_at(:key1) == "one"}

but I need it to just INCLUDE the string term.

Answer 1

arr = [
  {:key1=>"foo one bar", :key2=>"two", :key3=>"three"},
  {:key1=>"four", :key2=>"five", :key3=>"six"},
  {:key1=>"seven", :key2=>"eight", :key3=>"nine"}
]

arr.each {|x| x[:key1] = 'replaced' if x[:key1].include? 'one'}
p arr

Output:

[{:key1=>"replaced", :key2=>"two", :key3=>"three"}, {:key1=>"four", :key2=>"five", :key3=>"six"}, {:key1=>"seven", :key2=>"eight", :key3=>"nine"}]

Answer 2

您可以使用=~查找字符串中包含单词one的模式，例如：

arr.each{|item| item[:key1] = "replaced" if item[:key1] =~ /one/ }

Answer 3

A couple tricks can help create tight, but understandable, code. Starting with:

arr = [
  {:key1=>"foo one bar", :key2=>"two",   :key3=>"three"},
  {:key1=>"four",        :key2=>"five",  :key3=>"six"},
  {:key1=>"seven",       :key2=>"eight", :key3=>"nine"}
]

I can use:

arr.map!{ |h|
  h[:key1][/\b one \b/x] ?
    h.merge(:key1 => 'replaced') :
    h
}

Which, after running, looks like:

[
    [0] {
        :key1 => "replaced",
        :key2 => "two",
        :key3 => "three"
    },
    [1] {
        :key1 => "four",
        :key2 => "five",
        :key3 => "six"
    },
    [2] {
        :key1 => "seven",
        :key2 => "eight",
        :key3 => "nine"
    }
]

map! , like map , needs a return value otherwise it'd return a nil in that position in the iteration. A ternary statement fits this nicely. Keeping ternary statements looking nice in Ruby is more difficult than in a language like Perl, which I think is a large reason so many people avoid them. We can format a ternary using trailing operators though, which helps us break the statements into separate lines, improving readability.