Apache spark 这个RDD中的空白是从哪里来的？

其目的是转换驻留在文件中的整数：

1 2 3
4 5 6
7 8 9

分为三个数组，因此可以执行数学运算

预期的

[[1, 2, 3], [4, 5, 6], [7, 8, 9]]

实际值

[[u'1', u' ', u'2', u' ', u'3'], [u'4', u' ', u'5', u' ', u'6'], [u'7', u' ', u'8', u' ', u'9']]

代码

txt = sc.textFile("integers.txt")
print txt.collect()
#[u'1 2 3', u'4 5 6', u'7 8 9']

pairs = txt.map(lambda x: x.split(' '))
print pairs.collect()
#[[u'1', u'2', u'3'], [u'4', u'5', u'6'], [u'7', u'8', u'9']]

pairs = txt.map(lambda x: [s for s in x])
print pairs.collect()
#[[u'1', u' ', u'2', u' ', u'3'], [u'4', u' ', u'5', u' ', u'6'], [u'7', u' ', u'8', u' ', u'9']]

---

问题似乎在于数字是unicode格式，而不是int格式。 您可以通过将它们强制转换为int来解决此问题（请参见）

pairs = txt.map(lambda x: x.split(' '))
// this return every concatenated character that separated by space ' ', which kind of similar to following function (lamda also aware of newline from file)
def AFunc(aString):
   returnArray = []
   tempString = ""
   foreach(char in aString)
      if char == ' ':
         if tempString != "":
            returnArray.append(tempString)
            tempString = ""
      else:
         tempString += char
   return returnArray


// ..
pairs = txt.map(lambda x: [s for s in x])
// this return every character in a string, which kind of similar to following function (lamda also aware of newline from file)
def BFunc(aString):
   returnArray = []
   foreach(char in aString):
     returnArray.append(char)
   return returnArray

>>> pairs = txt.map(lambda x: x.split(' '))
>>> print pairs.collect()
[[u'1', u'2', u'3'], [u'4', u'5', u'6'], [u'7', u'8', u'9']]

>>> pairs2 = pairs.map(lambda x: [int(s) for s in x])
>>> print pairs2.collect()
[[1, 2, 3], [4, 5, 6], [7, 8, 9]]
>>>