tustvold commented on a change in pull request #1596:
URL: https://github.com/apache/arrow-datafusion/pull/1596#discussion_r787618177
##########
File path: datafusion/src/physical_plan/sorts/sort.rs
##########
@@ -15,47 +15,432 @@
// specific language governing permissions and limitations
// under the License.
-//! Defines the SORT plan
+//! Sort that deals with an arbitrary size of the input.
+//! It will do in-memory sorting if it has enough memory budget
+//! but spills to disk if needed.
use crate::error::{DataFusionError, Result};
+use crate::execution::memory_manager::{
+ ConsumerType, MemoryConsumer, MemoryConsumerId, MemoryManager,
+};
use crate::execution::runtime_env::RuntimeEnv;
-use crate::physical_plan::common::AbortOnDropSingle;
+use crate::physical_plan::common::{batch_byte_size, IPCWriter,
SizedRecordBatchStream};
use crate::physical_plan::expressions::PhysicalSortExpr;
use crate::physical_plan::metrics::{
- BaselineMetrics, ExecutionPlanMetricsSet, MetricsSet, RecordOutput,
+ BaselineMetrics, Count, ExecutionPlanMetricsSet, MetricsSet, Time,
};
+use
crate::physical_plan::sorts::sort_preserving_merge::SortPreservingMergeStream;
+use crate::physical_plan::sorts::SortedStream;
+use crate::physical_plan::stream::RecordBatchReceiverStream;
use crate::physical_plan::{
- common, DisplayFormatType, Distribution, ExecutionPlan, Partitioning,
+ common, DisplayFormatType, Distribution, EmptyRecordBatchStream,
ExecutionPlan,
+ Partitioning, SendableRecordBatchStream, Statistics,
};
-use crate::physical_plan::{RecordBatchStream, SendableRecordBatchStream,
Statistics};
+use arrow::array::ArrayRef;
pub use arrow::compute::SortOptions;
use arrow::compute::{lexsort_to_indices, take, SortColumn, TakeOptions};
use arrow::datatypes::SchemaRef;
use arrow::error::Result as ArrowResult;
+use arrow::ipc::reader::FileReader;
use arrow::record_batch::RecordBatch;
-use arrow::{array::ArrayRef, error::ArrowError};
use async_trait::async_trait;
-use futures::stream::Stream;
-use futures::Future;
-use pin_project_lite::pin_project;
+use futures::lock::Mutex;
Review comment:
Oh I thought it was, std::Mutex then. I more just meant a non-async
Mutex :grin:
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
