basicjoin.txt

# cat > R.txt
k1,v1
k1,v2
k2,v3
k2,v4
k3,v7
k3,v8
k3,v9

# cat > S.txt
k1,v11
k1,v22
k1,v33
k2,v55
k4,v77
k5,v88

# ./pyspark
Python 2.6.9 (unknown, Sep  9 2014, 15:05:12)
[GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.39)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
Welcome to
      ____              __
     / __/__  ___ _____/ /__
    _\ \/ _ \/ _ `/ __/  '_/
   /__ / .__/\_,_/_/ /_/\_\   version 1.2.0
      /_/

Using Python version 2.6.9 (unknown, Sep  9 2014 15:05:12)
SparkContext available as sc.
>>> R = sc.textFile("R.txt");
>>> R.collect()
[u'k1,v1', 
 u'k1,v2', 
 u'k2,v3', 
 u'k2,v4', 
 u'k3,v7', 
 u'k3,v8', 
 u'k3,v9']
 
>>> S = sc.textFile("S.txt");
>>> S.collect()
[u'k1,v11', 
 u'k1,v22', 
 u'k1,v33', 
 u'k2,v55', 
 u'k4,v77', 
 u'k5,v88'
]

>>> r1 = R.map(lambda s: s.split(","))
>>> r1.collect()
[
 [u'k1', u'v1'], 
 [u'k1', u'v2'], 
 [u'k2', u'v3'], 
 [u'k2', u'v4'], 
 [u'k3', u'v7'], 
 [u'k3', u'v8'], 
 [u'k3', u'v9']
]
>>> r2 = r1.flatMap(lambda s: [(s[0], s[1])])
>>> r2.collect()
[
 (u'k1', u'v1'), 
 (u'k1', u'v2'), 
 (u'k2', u'v3'), 
 (u'k2', u'v4'), 
 (u'k3', u'v7'), 
 (u'k3', u'v8'), 
 (u'k3', u'v9')
]
>>>
>>> s1 = S.map(lambda s: s.split(","))
>>> s1.collect()
[
 [u'k1', u'v11'], 
 [u'k1', u'v22'], 
 [u'k1', u'v33'], 
 [u'k2', u'v55'], 
 [u'k4', u'v77'], 
 [u'k5', u'v88']
]
>>> s2 = s1.flatMap(lambda s: [(s[0], s[1])])
>>> s2.collect()
[
 (u'k1', u'v11'), 
 (u'k1', u'v22'), 
 (u'k1', u'v33'), 
 (u'k2', u'v55'), 
 (u'k4', u'v77'), 
 (u'k5', u'v88')
]
>>> RjoinedS = r2.join(s2)
>>> RjoinedS.collect()
[
 (u'k2', (u'v3', u'v55')), 
 (u'k2', (u'v4', u'v55')), 
 (u'k1', (u'v1', u'v11')), 
 (u'k1', (u'v1', u'v22')), 
 (u'k1', (u'v1', u'v33')), 
 (u'k1', (u'v2', u'v11')), 
 (u'k1', (u'v2', u'v22')), 
 (u'k1', (u'v2', u'v33'))
]
>>>