嵌套循环的pyspark程序

Question

我是PySpark的新手,我想了解如何在PySpark中编写多个嵌套for循环,下面是粗略的高级示例.任何帮助将不胜感激.

for ( i=0;i<10;i++)
   for ( j=0;j<10;j++)
       for ( k=0;k<10;k++)
          { 
           print "i"."j"."k"
}

Answer 1

在非分布式设置中,使用foreach组合器重写for循环,但由于Spark性质,map并且flatMap是更好的选择:

from __future__ import print_function
a_loop = lambda x: ((x, y) for y in xrange(10))
print_me = lambda ((x, y), z): print("{0}.{1}.{2}".format(x, y, z)))

(sc.
    parallelize(xrange(10)).
    flatMap(a_loop).
    flatMap(a_loop).
    foreach(print_me)

使用itertools.product:

from itertools import product
sc.parallelize(product(xrange(10), repeat=3)).foreach(print)