chore: use upstream versions of some workarounds (#2057)

* chore: use upstream versions of some workarounds

* docs: update docstring

Co-authored-by: kodiakhq[bot] <49736102+kodiakhq[bot]@users.noreply.github.com>

internal_types/src/schema.rs

@@ -786,12 +786,12 @@ macro_rules! assert_column_eq {
 
 #[cfg(test)]
 mod test {
+    use arrow::compute::SortOptions;
     use InfluxColumnType::*;
     use InfluxFieldType::*;
 
     use super::{builder::SchemaBuilder, *};
     use crate::schema::merge::SchemaMerger;
-    use crate::schema::sort::SortOptions;
 
     fn make_field(
         name: &str,

internal_types/src/schema/sort.rs

@@ -1,5 +1,6 @@
 use std::{fmt::Display, str::FromStr};
 
+use arrow::compute::SortOptions;
 use indexmap::{map::Iter, IndexMap};
 use itertools::Itertools;
 use snafu::Snafu;
@@ -23,24 +24,6 @@ pub enum Error {
 
 pub type Result<T, E = Error> = std::result::Result<T, E>;
 
-/// Temporary - <https://github.com/apache/arrow-rs/pull/425>
-#[derive(Debug, Clone, Copy, Eq, PartialEq)]
-pub struct SortOptions {
-    /// Whether to sort in descending order
-    pub descending: bool,
-    /// Whether to sort nulls first
-    pub nulls_first: bool,
-}
-
-impl Default for SortOptions {
-    fn default() -> Self {
-        Self {
-            descending: false,
-            nulls_first: true,
-        }
-    }
-}
-
 #[derive(Debug, Clone, Copy, Eq, PartialEq)]
 pub struct ColumnSort {
     /// Position of this column in the sort key

query/src/frontend/reorg.rs

@@ -268,8 +268,9 @@ struct ScanPlan<C: QueryChunk + 'static> {
 
 #[cfg(test)]
 mod test {
+    use arrow::compute::SortOptions;
     use arrow_util::assert_batches_eq;
-    use internal_types::schema::{merge::SchemaMerger, sort::SortOptions};
+    use internal_types::schema::merge::SchemaMerger;
 
     use crate::{
         exec::{Executor, ExecutorType},

query/src/provider/deduplicate/algo.rs

@@ -339,21 +339,12 @@ impl RecordBatchDeduplicator {
     }
 
     /// Create a new record batch from offset --> len
-    ///
-    /// <https://github.com/apache/arrow-rs/issues/460> for adding this upstream
     fn slice_record_batch(
         batch: &RecordBatch,
         offset: usize,
         len: usize,
     ) -> ArrowResult<RecordBatch> {
-        let schema = batch.schema();
-        let new_columns: Vec<_> = batch
-            .columns()
-            .iter()
-            .map(|old_column| old_column.slice(offset, len))
-            .collect();
-
-        let batch = RecordBatch::try_new(schema, new_columns)?;
+        let batch = batch.slice(offset, len);
 
         // At time of writing, `concat_batches` concatenates the
         // contents of dictionaries as well; Do a post pass to remove the