如果新数据可以是数组或标量，并且如果它是数组，则希望防止新数据嵌套，splat操作符非常棒!它为标量返回一个标量，为数组返回一个已解压的参数列表。

1.9.3-p551 :020 > a = [1, 2]
 => [1, 2] 
1.9.3-p551 :021 > b = [3, 4]
 => [3, 4] 
1.9.3-p551 :022 > c = 5
 => 5 
1.9.3-p551 :023 > a.object_id
 => 6617020 
1.9.3-p551 :024 > a.push *b
 => [1, 2, 3, 4] 
1.9.3-p551 :025 > a.object_id
 => 6617020 
1.9.3-p551 :026 > a.push *c
 => [1, 2, 3, 4, 5] 
1.9.3-p551 :027 > a.object_id
 => 6617020 

我发现更容易推入或追加数组，然后将它们压平，如下所示:

somearray = ["some", "thing"]
anotherarray = ["another", "thing"]
somearray.push anotherarray # => ["some", "thing", ["another", "thing"]]
#or
somearray << anotherarray # => ["some", "thing", ["another", "thing"]]
somearray.flatten!  # => ["some", "thing", "another", "thing"]
somearray # => ["some", "thing", "another", "thing"]

Somearray = ["some"， "thing"]

Anotherarray = ["another"， "thing"]

Somearray + anotherarray

你可以使用+运算符!

irb(main):001:0> a = [1,2]
=> [1, 2]
irb(main):002:0> b = [3,4]
=> [3, 4]
irb(main):003:0> a + b
=> [1, 2, 3, 4]

你可以在这里阅读关于数组类的所有内容: http://ruby-doc.org/core/classes/Array.html

详细阐述@Pilcrow的答案，对于大型数组唯一合适的答案是concat(+)，因为它速度快，并且在循环中操作时不会分配一个新对象来进行垃圾回收。

下面是基准:

require 'benchmark'

huge_ary_1 = Array.new(1_000_000) { rand(5_000_000..30_000_00) }

huge_ary_2 = Array.new(1_000_000) { rand(35_000_000..55_000_00) }

Benchmark.bm do |bm|
  p '-------------------CONCAT ----------------'
  bm.report { huge_ary_1.concat(huge_ary_2) }

  p '------------------- PUSH ----------------'
  bm.report { huge_ary_1.push(*huge_ary_2)  }
end

结果:

       user     system      total        real
"-------------------CONCAT ----------------"
  0.000000   0.000000   0.000000 (  0.009388)
"------------------- PUSH ----------------"
  example/array_concat_vs_push.rb:13:in `block (2 levels) in <main>': stack level too deep (SystemStackError)

正如你所看到的，当数组足够大时，使用push会抛出ERROR:堆栈级别太深(SystemStackError)。

只是另一种方法。

[somearray, anotherarray].flatten
=> ["some", "thing", "another", "thing"]