[
https://issues.apache.org/jira/browse/ARROW-2328?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16428582#comment-16428582
]
ASF GitHub Bot commented on ARROW-2328:
---------------------------------------
Adriandorr commented on a change in pull request #1784: ARROW-2328: [C++] Fixed
and unit tested feather writing with slice
URL: https://github.com/apache/arrow/pull/1784#discussion_r179815275
##########
File path: cpp/src/arrow/ipc/feather-test.cc
##########
@@ -406,6 +406,125 @@ TEST_F(TestTableWriter, PrimitiveNullRoundTrip) {
}
}
+TEST_F(TestTableWriter, SliceRoundTrip) {
+ std::shared_ptr<RecordBatch> batch;
+ ASSERT_OK(MakeIntBatchSized(300, &batch));
+ batch = batch->Slice(100, 100);
+
+ ASSERT_OK(writer_->Append("f0", *batch->column(0)));
+ ASSERT_OK(writer_->Append("f1", *batch->column(1)));
+ Finish();
+
+ std::shared_ptr<Column> col;
+ ASSERT_OK(reader_->GetColumn(0, &col));
+ ASSERT_TRUE(col->data()->chunk(0)->Equals(batch->column(0)));
+ ASSERT_EQ("f0", col->name());
+
+ ASSERT_OK(reader_->GetColumn(1, &col));
+ ASSERT_TRUE(col->data()->chunk(0)->Equals(batch->column(1)));
+ ASSERT_EQ("f1", col->name());
+}
+
+TEST_F(TestTableWriter, SliceStringsRoundTrip) {
+ std::shared_ptr<RecordBatch> batch;
+ ASSERT_OK(MakeStringTypesRecordBatchWithNulls(&batch, false));
+ batch = batch->Slice(320, 30);
+
+ ASSERT_OK(writer_->Append("f0", *batch->column(0)));
+ ASSERT_OK(writer_->Append("f1", *batch->column(1)));
+ Finish();
+
+ std::shared_ptr<Column> col;
+ ASSERT_OK(reader_->GetColumn(0, &col));
+ SCOPED_TRACE(col->data()->chunk(0)->ToString() + "\n" +
batch->column(0)->ToString());
+ ASSERT_TRUE(col->data()->chunk(0)->Equals(batch->column(0)));
+ ASSERT_EQ("f0", col->name());
+
+ ASSERT_OK(reader_->GetColumn(1, &col));
+ ASSERT_TRUE(col->data()->chunk(0)->Equals(batch->column(1)));
+ ASSERT_EQ("f1", col->name());
+}
+
+TEST_F(TestTableWriter, SliceStringsWithNullsRoundTrip) {
+ std::shared_ptr<RecordBatch> batch;
+ ASSERT_OK(MakeStringTypesRecordBatchWithNulls(&batch, true));
+ batch = batch->Slice(320, 30);
+
+ ASSERT_OK(writer_->Append("f0", *batch->column(0)));
+ ASSERT_OK(writer_->Append("f1", *batch->column(1)));
+ Finish();
+
+ std::shared_ptr<Column> col;
+ ASSERT_OK(reader_->GetColumn(0, &col));
+ SCOPED_TRACE(col->data()->chunk(0)->ToString() + "\n" +
batch->column(0)->ToString());
+ ASSERT_TRUE(col->data()->chunk(0)->Equals(batch->column(0)));
+ ASSERT_EQ("f0", col->name());
+
+ ASSERT_OK(reader_->GetColumn(1, &col));
+ ASSERT_TRUE(col->data()->chunk(0)->Equals(batch->column(1)));
+ ASSERT_EQ("f1", col->name());
+}
+
+TEST_F(TestTableWriter, SliceAtNonEightOffsetStringsWithNullsRoundTrip) {
Review comment:
I think the problem with loops in unit tests is that you only see the first
failure. That said I should be able to convert these tests to one parameterised
test, which hopefully also answers the above concerns..
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> Writing a slice with feather ignores the offset
> -----------------------------------------------
>
> Key: ARROW-2328
> URL: https://issues.apache.org/jira/browse/ARROW-2328
> Project: Apache Arrow
> Issue Type: Bug
> Components: C++
> Affects Versions: 0.8.0
> Reporter: Adrian
> Priority: Major
> Labels: pull-request-available
> Fix For: 0.10.0
>
> Original Estimate: 0.5h
> Remaining Estimate: 0.5h
>
> Writing a slice from row n of length m of an array to feather would write the
> first m rows, instead of the rows starting at n.
> The null bitmap also ends up misaligned. Also tested and fixed in the pull
> request below.
> I've created a pull request with tests and fix here:
> [Pullrequest#1766|https://github.com/apache/arrow/pull/1766]
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
