Skip to content

Minor: Add a parquet row_filter test, reduce some test boiler plate #7522

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

Merged
merged 5 commits into from
May 20, 2025
Merged

Minor: Add a parquet row_filter test, reduce some test boiler plate #7522

Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
db849a7
Add single row filter test
alamb May 16, 2025
0e7c6ee
DRI
alamb May 16, 2025
f049e4c
Update parquet/src/arrow/async_reader/mod.rs
alamb May 19, 2025
6ca1468
Merge remote-tracking branch 'apache/main' into alamb/more_row_filter…
alamb May 19, 2025
ea66581
Improve comments for other tests
alamb May 19, 2025
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
170 changes: 95 additions & 75 deletions parquet/src/arrow/async_reader/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1124,6 +1124,16 @@ mod tests {
requests: Arc<Mutex<Vec<Range<usize>>>>,
}

impl TestReader {
fn new(data: Bytes) -> Self {
Self {
data,
metadata: Default::default(),
requests: Default::default(),
}
}
}

impl AsyncFileReader for TestReader {
fn get_bytes(&mut self, range: Range<u64>) -> BoxFuture<'_, Result<Bytes>> {
let range = range.clone();
Expand Down Expand Up @@ -1156,11 +1166,7 @@ mod tests {
let path = format!("{testdata}/alltypes_plain.parquet");
let data = Bytes::from(std::fs::read(path).unwrap());

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let requests = async_reader.requests.clone();
let builder = ParquetRecordBatchStreamBuilder::new(async_reader)
Expand Down Expand Up @@ -1209,11 +1215,7 @@ mod tests {
let path = format!("{testdata}/alltypes_plain.parquet");
let data = Bytes::from(std::fs::read(path).unwrap());

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let requests = async_reader.requests.clone();
let builder = ParquetRecordBatchStreamBuilder::new(async_reader)
Expand Down Expand Up @@ -1270,11 +1272,7 @@ mod tests {
let path = format!("{testdata}/alltypes_tiny_pages_plain.parquet");
let data = Bytes::from(std::fs::read(path).unwrap());

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let options = ArrowReaderOptions::new().with_page_index(true);
let builder = ParquetRecordBatchStreamBuilder::new_with_options(async_reader, options)
Expand Down Expand Up @@ -1339,11 +1337,7 @@ mod tests {

assert_eq!(metadata.num_row_groups(), 1);

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let builder = ParquetRecordBatchStreamBuilder::new(async_reader)
.await
Expand Down Expand Up @@ -1380,11 +1374,7 @@ mod tests {
let path = format!("{testdata}/alltypes_tiny_pages_plain.parquet");
let data = Bytes::from(std::fs::read(path).unwrap());

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let options = ArrowReaderOptions::new().with_page_index(true);
let builder = ParquetRecordBatchStreamBuilder::new_with_options(async_reader, options)
Expand Down Expand Up @@ -1458,11 +1448,7 @@ mod tests {

let selection = RowSelection::from(selectors);

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let options = ArrowReaderOptions::new().with_page_index(true);
let builder = ParquetRecordBatchStreamBuilder::new_with_options(async_reader, options)
Expand Down Expand Up @@ -1524,11 +1510,7 @@ mod tests {

let selection = RowSelection::from(selectors);

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let options = ArrowReaderOptions::new().with_page_index(true);
let builder = ParquetRecordBatchStreamBuilder::new_with_options(async_reader, options)
Expand All @@ -1555,6 +1537,70 @@ mod tests {

#[tokio::test]
async fn test_row_filter() {
let a = StringArray::from_iter_values(["a", "b", "b", "b", "c", "c"]);
let b = StringArray::from_iter_values(["1", "2", "3", "4", "5", "6"]);
let data = RecordBatch::try_from_iter([
("a", Arc::new(a) as ArrayRef),
("b", Arc::new(b) as ArrayRef),
])
.unwrap();

let mut buf = Vec::with_capacity(1024);
let mut writer = ArrowWriter::try_new(&mut buf, data.schema(), None).unwrap();
writer.write(&data).unwrap();
writer.close().unwrap();

let data: Bytes = buf.into();
let metadata = ParquetMetaDataReader::new()
.parse_and_finish(&data)
.unwrap();
let parquet_schema = metadata.file_metadata().schema_descr_ptr();

let test = TestReader::new(data);
let requests = test.requests.clone();

let a_scalar = StringArray::from_iter_values(["b"]);
let a_filter = ArrowPredicateFn::new(
ProjectionMask::leaves(&parquet_schema, vec![0]),
move |batch| eq(batch.column(0), &Scalar::new(&a_scalar)),
);

let filter = RowFilter::new(vec![Box::new(a_filter)]);

let mask = ProjectionMask::leaves(&parquet_schema, vec![0, 1]);
let stream = ParquetRecordBatchStreamBuilder::new(test)
.await
.unwrap()
.with_projection(mask.clone())
.with_batch_size(1024)
.with_row_filter(filter)
.build()
.unwrap();

let batches: Vec<_> = stream.try_collect().await.unwrap();
assert_eq!(batches.len(), 1);

let batch = &batches[0];
assert_eq!(batch.num_columns(), 2);

// Filter should have kept only rows with "b" in column 0
assert_eq!(
batch.column(0).as_ref(),
&StringArray::from_iter_values(["b", "b", "b"])
);
assert_eq!(
batch.column(1).as_ref(),
&StringArray::from_iter_values(["2", "3", "4"])
);

// Should only have made 2 requests:
// * First request fetches data for evaluating the predicate
// * Second request fetches data for evaluating the projection
assert_eq!(requests.lock().unwrap().len(), 2);
}

#[tokio::test]
async fn test_two_row_filters() {
let a = StringArray::from_iter_values(["a", "b", "b", "b", "c", "c"]);
let b = StringArray::from_iter_values(["1", "2", "3", "4", "5", "6"]);
let c = Int32Array::from_iter(0..6);
Expand All @@ -1576,11 +1622,7 @@ mod tests {
.unwrap();
let parquet_schema = metadata.file_metadata().schema_descr_ptr();

let test = TestReader {
data,
metadata: Default::default(),
requests: Default::default(),
};
let test = TestReader::new(data);
let requests = test.requests.clone();

let a_scalar = StringArray::from_iter_values(["b"]);
Expand Down Expand Up @@ -1623,6 +1665,9 @@ mod tests {
assert_eq!(val, 3);

// Should only have made 3 requests
// * First request fetches data for evaluating the first predicate
// * Second request fetches data for evaluating the second predicate
// * Third request fetches data for evaluating the projection
assert_eq!(requests.lock().unwrap().len(), 3);
}

Expand Down Expand Up @@ -1653,11 +1698,7 @@ mod tests {

assert_eq!(metadata.num_row_groups(), 2);

let test = TestReader {
data,
metadata: Default::default(),
requests: Default::default(),
};
let test = TestReader::new(data);

let stream = ParquetRecordBatchStreamBuilder::new(test.clone())
.await
Expand Down Expand Up @@ -1744,11 +1785,7 @@ mod tests {

assert_eq!(metadata.num_row_groups(), 1);

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let a_filter =
ArrowPredicateFn::new(ProjectionMask::leaves(&parquet_schema, vec![1]), |batch| {
Expand Down Expand Up @@ -1812,11 +1849,7 @@ mod tests {

assert_eq!(metadata.num_row_groups(), 1);

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let requests = async_reader.requests.clone();
let (_, fields) = parquet_to_arrow_schema_and_fields(
Expand Down Expand Up @@ -1882,11 +1915,7 @@ mod tests {
let path = format!("{testdata}/alltypes_plain.parquet");
let data = Bytes::from(std::fs::read(path).unwrap());

let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let builder = ParquetRecordBatchStreamBuilder::new(async_reader)
.await
Expand Down Expand Up @@ -2025,11 +2054,7 @@ mod tests {
let testdata = arrow::util::test_util::parquet_test_data();
let path = format!("{testdata}/data_index_bloom_encoding_stats.parquet");
let data = Bytes::from(std::fs::read(path).unwrap());
let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());
let builder = ParquetRecordBatchStreamBuilder::new(async_reader)
.await
.unwrap();
Expand All @@ -2052,11 +2077,7 @@ mod tests {
}

async fn test_get_row_group_column_bloom_filter(data: Bytes, with_length: bool) {
let async_reader = TestReader {
data: data.clone(),
metadata: Default::default(),
requests: Default::default(),
};
let async_reader = TestReader::new(data.clone());

let mut builder = ParquetRecordBatchStreamBuilder::new(async_reader)
.await
Expand Down Expand Up @@ -2195,11 +2216,7 @@ mod tests {
.unwrap();
let parquet_schema = metadata.file_metadata().schema_descr_ptr();

let test = TestReader {
data,
metadata: Default::default(),
requests: Default::default(),
};
let test = TestReader::new(data);
let requests = test.requests.clone();

let a_scalar = StringArray::from_iter_values(["b"]);
Expand Down Expand Up @@ -2250,6 +2267,9 @@ mod tests {
assert_eq!(val, 3);

// Should only have made 3 requests
// * First request fetches data for evaluating the first predicate
// * Second request fetches data for evaluating the second predicate
// * Third request fetches data for evaluating the projection
assert_eq!(requests.lock().unwrap().len(), 3);
}

Expand Down
Loading