self._number_left = length//chunksize + bool(length % chunksize)
def _set(self, i, success_result):
+ self._number_left -= 1
success, result = success_result
- if success:
+ if success and self._success:
self._value[i*self._chunksize:(i+1)*self._chunksize] = result
- self._number_left -= 1
if self._number_left == 0:
if self._callback:
self._callback(self._value)
del self._cache[self._job]
self._event.set()
else:
- self._success = False
- self._value = result
- if self._error_callback:
- self._error_callback(self._value)
- del self._cache[self._job]
- self._event.set()
+ if not success and self._success:
+ # only store first exception
+ self._success = False
+ self._value = result
+ if self._number_left == 0:
+ # only consider the result ready once all jobs are done
+ if self._error_callback:
+ self._error_callback(self._value)
+ del self._cache[self._job]
+ self._event.set()
#
# Class whose instances are returned by `Pool.imap()`
def mul(x, y):
return x*y
+def raise_large_valuerror(wait):
+ time.sleep(wait)
+ raise ValueError("x" * 1024**2)
+
class SayWhenError(ValueError): pass
def exception_throwing_generator(total, when):
with self.assertRaises(RuntimeError):
p.apply(self._test_wrapped_exception)
+ def test_map_no_failfast(self):
+ # Issue #23992: the fail-fast behaviour when an exception is raised
+ # during map() would make Pool.join() deadlock, because a worker
+ # process would fill the result queue (after the result handler thread
+ # terminated, hence not draining it anymore).
+
+ t_start = time.time()
+
+ with self.assertRaises(ValueError):
+ with self.Pool(2) as p:
+ try:
+ p.map(raise_large_valuerror, [0, 1])
+ finally:
+ time.sleep(0.5)
+ p.close()
+ p.join()
+
+ # check that we indeed waited for all jobs
+ self.assertGreater(time.time() - t_start, 0.9)
+
def raising():
raise KeyError("key")
Library
-------
+- Issue #23992: multiprocessing: make MapResult not fail-fast upon exception.
+
- Issue #26243: Support keyword arguments to zlib.compress(). Patch by Aviv
Palivoda.