Skip to content

[SPARK-22032][PySpark] Speed up StructType conversion #19249

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 6 commits into from
Closed

[SPARK-22032][PySpark] Speed up StructType conversion #19249

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
aa69a72
Update types.py
maver1ck Sep 15, 2017
e4d7f76
PEP8 fix
maver1ck Sep 15, 2017
64afb16
PEP8 fix
maver1ck Sep 15, 2017
b1800ac
Add optimization to 'toInternal' function
maver1ck Sep 17, 2017
8708a9d
PEP8 fix
maver1ck Sep 17, 2017
e9b7798
Comments
maver1ck Sep 17, 2017
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 16 additions & 6 deletions python/pyspark/sql/types.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -483,7 +483,9 @@ def __init__(self, fields=None):
self.names = [f.name for f in fields]
assert all(isinstance(f, StructField) for f in fields),\
"fields should be a list of StructField"
self._needSerializeAnyField = any(f.needConversion() for f in self)
# Precalculated list of fields that need conversion with fromInternal/toInternal functions
self._needConversion = [f.needConversion() for f in self]
self._needSerializeAnyField = any(self._needConversion)

def add(self, field, data_type=None, nullable=True, metadata=None):
"""
Expand Down Expand Up @@ -528,7 +530,9 @@ def add(self, field, data_type=None, nullable=True, metadata=None):
data_type_f = data_type
self.fields.append(StructField(field, data_type_f, nullable, metadata))
self.names.append(field)
self._needSerializeAnyField = any(f.needConversion() for f in self)
# Precalculated list of fields that need conversion with fromInternal/toInternal functions
self._needConversion = [f.needConversion() for f in self]
self._needSerializeAnyField = any(self._needConversion)
return self

def __iter__(self):
Expand Down Expand Up @@ -590,13 +594,17 @@ def toInternal(self, obj):
return

if self._needSerializeAnyField:
# Only calling toInternal function for fields that need conversion
if isinstance(obj, dict):
return tuple(f.toInternal(obj.get(n)) for n, f in zip(self.names, self.fields))
return tuple(f.toInternal(obj.get(n)) if c else obj.get(n)
for n, f, c in zip(self.names, self.fields, self._needConversion))
elif isinstance(obj, (tuple, list)):
return tuple(f.toInternal(v) for f, v in zip(self.fields, obj))
return tuple(f.toInternal(v) if c else v
for f, v, c in zip(self.fields, obj, self._needConversion))
elif hasattr(obj, "__dict__"):
d = obj.__dict__
return tuple(f.toInternal(d.get(n)) for n, f in zip(self.names, self.fields))
return tuple(f.toInternal(d.get(n)) if c else d.get(n)
for n, f, c in zip(self.names, self.fields, self._needConversion))
else:
raise ValueError("Unexpected tuple %r with StructType" % obj)
else:
Expand All @@ -619,7 +627,9 @@ def fromInternal(self, obj):
# it's already converted by pickler
return obj
if self._needSerializeAnyField:
values = [f.fromInternal(v) for f, v in zip(self.fields, obj)]
# Only calling fromInternal function for fields that need conversion
values = [f.fromInternal(v) if c else v
for f, v, c in zip(self.fields, obj, self._needConversion)]
else:
values = obj
return _create_row(self.names, values)
Expand Down