Skip to content
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

Remove duplicate psp_okey column from arrow updates #1044

Merged
merged 3 commits into from
May 13, 2020
Merged

Remove duplicate psp_okey column from arrow updates #1044

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
1a3bef7
add arrow update tests
sc1f May 11, 2020
39220c3
WIP: fix arrow segfaults
sc1f May 12, 2020
9dd545e
remove duplicate psp_okey
sc1f May 12, 2020
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
5 changes: 3 additions & 2 deletions cpp/perspective/src/cpp/emscripten.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1064,7 +1064,8 @@ namespace binding {

// Always use the `Table` column names and data types on up
if (table_initialized && is_update) {
auto schema = gnode->get_output_schema();
auto gnode_output_schema = gnode->get_output_schema();
auto schema = gnode_output_schema.drop({"psp_okey"});
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It is amazing to me that this coerces to std::set<std::string> ..

Since column_names is simply iterated on 10loc below this, it may be cheaper to simple continue on "psp_okey", since drop() iterates through the set and reconstructs both name and type vectors from scratch.

column_names = schema.columns();
data_types = schema.types();

Expand Down Expand Up @@ -1099,7 +1100,7 @@ namespace binding {
}

// Updated data types need to reflect in new data table
auto new_schema = gnode->get_output_schema();
auto new_schema = gnode->get_output_schema().drop({"psp_okey"});
data_types = new_schema.types();
}
} else {
Expand Down
5 changes: 3 additions & 2 deletions python/perspective/perspective/src/table.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,7 +62,8 @@ std::shared_ptr<Table> make_table_py(t_val table, t_data_accessor accessor,

// Always use the `Table` column names and data types on update.
if (table_initialized && is_update) {
auto schema = gnode->get_output_schema();
auto gnode_output_schema = gnode->get_output_schema();
auto schema = gnode_output_schema.drop({"psp_okey"});
column_names = schema.columns();
data_types = schema.types();

Expand Down Expand Up @@ -97,7 +98,7 @@ std::shared_ptr<Table> make_table_py(t_val table, t_data_accessor accessor,
}
}
// Make sure promoted types are used to construct data table
auto new_schema = gnode->get_output_schema();
auto new_schema = gnode->get_output_schema().drop({"psp_okey"});
data_types = new_schema.types();
} else {
column_names = arrow_loader.names();
Expand Down
50 changes: 50 additions & 0 deletions python/perspective/perspective/tests/table/test_update_arrow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,8 +7,11 @@
#

import os
import random
import uuid
import pyarrow as pa
from datetime import date, datetime
from pytest import mark
from perspective.table import Table

SOURCE_STREAM_ARROW = os.path.join(os.path.dirname(__file__), "arrow", "int_float_str.arrow")
Expand Down Expand Up @@ -476,3 +479,50 @@ def test_update_arrow_column_order_int(self, util):
assert tbl.view().to_dict() == {
name: data[0] for name in names
}

def test_update_arrow_thread_safe_int_index(self, util):
data = [["a", "b", "c"] for i in range(10)]
data += [[1, 2, 3]]
names = ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "uid"]
arrow = util.make_arrow(names, data)
tbl = Table(arrow, index="uid")

for i in range(100):
idx = (1, 2, 3)[random.randint(0, 2)]
update_data = [[str(uuid.uuid4()) + str(random.randint(100, 1000000000))], [idx]]
update_names = [names[random.randint(0, 9)], "uid"]
update_arrow = util.make_arrow(update_names, update_data)
tbl.update(update_arrow)

assert tbl.size() == 3

def test_update_arrow_thread_safe_datetime_index(self, util):
data = [["a", "b", "c"] for i in range(10)]
data += [[datetime(2020, 1, 15, 12, 17), datetime(2020, 1, 15, 12, 18), datetime(2020, 1, 15, 12, 19)]]
names = ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "uid"]
arrow = util.make_arrow(names, data)
tbl = Table(arrow, index="uid")

for i in range(100):
idx = (datetime(2020, 1, 15, 12, 17), datetime(2020, 1, 15, 12, 18), datetime(2020, 1, 15, 12, 19))[random.randint(0, 2)]
update_data = [[str(uuid.uuid4()) + str(random.randint(100, 1000000000))], [idx]]
update_names = [names[random.randint(0, 9)], "uid"]
update_arrow = util.make_arrow(update_names, update_data)
tbl.update(update_arrow)

assert tbl.size() == 3

def test_update_arrow_thread_safe_str_index(self, util):
data = [["a", "b", "c"] for i in range(11)]
names = ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "uid"]
arrow = util.make_arrow(names, data)
tbl = Table(arrow, index="uid")

for i in range(100):
idx = ("a", "b", "c")[random.randint(0, 2)]
update_data = [[str(uuid.uuid4()) + str(random.randint(100, 1000000000))], [idx]]
update_names = [names[random.randint(0, 9)], "uid"]
update_arrow = util.make_arrow(update_names, update_data)
tbl.update(update_arrow)

assert tbl.size() == 3