I am starting a new process and redirecting its stdout to a text file. I am want to read the progress the processing from the file, however the file stays empty until the process finishes executing. Can you explain this behavior and if there is a work around it ?
p = Process(target=foo, args=(a,))
p.start()
def foo(a):
sys.stdout = open(str(os.getpid()) + ".out", "w")
print("test")
time.sleep(10)
The workaround is to add flush=True to your print statement.
from multiprocessing import Process
import sys
import time
def foo(a):
sys.stdout = open(str(os.getpid()) + ".out", "w")
print("test", flush=True)
time.sleep(10)
if __name__ == '__main__':
p = Process(target=foo, args=(1,))
p.start()
@azelcer Is correct I need to use unbuffered output, however text i/o can' be unbuffered in python so the best way is to use a buffer of size one so the output will be flushed to the file every time a line is printed.
p = Process(target=foo, args=(a,))
p.start()
def foo(a):
sys.stdout = open(str(os.getpid()) + ".out", "w", 1)
print("test")
time.sleep(10)
