持續更新。
--------------------C++篇------------------------分佈式
auto results = from(ids) | get<0>() | as<vector>();
本質上和e.g.Pyspark的分佈式計算的底層思想是一致的。函數
--------------------Python篇-----------------------spa
subprocess:
Popen, e.g.rest
process = Popen(['cat', 'test.py'], stdout=PIPE, stderr=PIPE)
至關於執行了cat test.py這個命令
而後能夠用communicate函數來read,e.g. stdout, stderr = process.communicate(); print stdoutcode
yield & generators
e.g.ip
def foo(): for i in range(0,100): yield i*i generator = foo() for i in generator: print(i)