Github user davies commented on a diff in the pull request:
https://github.com/apache/spark/pull/12057#discussion_r58109906
--- Diff: python/pyspark/worker.py ---
@@ -59,7 +59,54 @@ def read_command(serializer, file):
def chain(f, g):
"""chain two function together """
- return lambda x: g(f(x))
+ return lambda *a: g(f(*a))
+
+
+def wrap_udf(f, return_type):
+ if return_type.needConversion():
+ toInternal = return_type.toInternal
+ return lambda *a: toInternal(f(*a))
+ else:
+ return lambda *a: f(*a)
+
+
+def read_single_udf(pickleSer, infile):
+ num_arg = read_int(infile)
+ arg_offsets = [read_int(infile) for i in range(num_arg)]
+ row_func = None
+ for i in range(read_int(infile)):
+ f, return_type = read_command(pickleSer, infile)
+ if row_func is None:
+ row_func = f
+ else:
+ row_func = chain(row_func, f)
+ # the last returnType will be the return type of UDF
+ return arg_offsets, wrap_udf(row_func, return_type)
+
+
+def read_udfs(pickleSer, infile):
+ num_udfs = read_int(infile)
+ if num_udfs == 1:
+ # fast path for single UDF
+ _, udf = read_single_udf(pickleSer, infile)
+ mapper = lambda a: udf(*a)
+ else:
+ udfs = {}
+ call_udf = []
+ for i in range(num_udfs):
+ arg_offsets, udf = read_single_udf(pickleSer, infile)
+ udfs['f%d' % i] = udf
+ args = ["a[%d]" % o for o in arg_offsets]
+ call_udf.append("f%d(%s)" % (i, ", ".join(args)))
+ # Create function like this:
+ # lambda a: (f0(a0), f1(a1, a2), f2(a3))
+ mapper_str = "lambda a: (%s)" % (", ".join(call_udf))
+ mapper = eval(mapper_str, udfs)
+
+ func = lambda _, it: map(mapper, it)
+ ser = BatchedSerializer(PickleSerializer(), 100)
--- End diff --
Before this patch, we use AutoBatchedSerialzier, which could hold thousands
of rows (holding more rows in JVM, may cause OOM).
The `100` is used on Java side.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
