arrow_json/reader/
run_end_array.rs

1// Licensed to the Apache Software Foundation (ASF) under one
2// or more contributor license agreements.  See the NOTICE file
3// distributed with this work for additional information
4// regarding copyright ownership.  The ASF licenses this file
5// to you under the Apache License, Version 2.0 (the
6// "License"); you may not use this file except in compliance
7// with the License.  You may obtain a copy of the License at
8//
9//   http://www.apache.org/licenses/LICENSE-2.0
10//
11// Unless required by applicable law or agreed to in writing,
12// software distributed under the License is distributed on an
13// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
14// KIND, either express or implied.  See the License for the
15// specific language governing permissions and limitations
16// under the License.
17
18use std::marker::PhantomData;
19use std::ops::Range;
20use std::slice::from_ref;
21use std::sync::Arc;
22
23use arrow_array::types::RunEndIndexType;
24use arrow_array::{ArrayRef, RunArray, UInt32Array, new_empty_array};
25use arrow_buffer::{ArrowNativeType, RunEndBuffer, ScalarBuffer};
26use arrow_ord::partition::partition;
27use arrow_schema::{ArrowError, DataType};
28use arrow_select::take::take;
29
30use crate::reader::tape::Tape;
31use crate::reader::{ArrayDecoder, DecoderContext};
32
33pub struct RunEndEncodedArrayDecoder<R> {
34    data_type: DataType,
35    decoder: Box<dyn ArrayDecoder>,
36    values_nullable: bool,
37    phantom: PhantomData<R>,
38}
39
40impl<R: RunEndIndexType> RunEndEncodedArrayDecoder<R> {
41    pub fn new(
42        ctx: &DecoderContext,
43        data_type: &DataType,
44        is_nullable: bool,
45    ) -> Result<Self, ArrowError> {
46        let DataType::RunEndEncoded(_, values_field) = data_type else {
47            unreachable!()
48        };
49        let values_nullable = values_field.is_nullable() && is_nullable;
50        let decoder = ctx.make_decoder(values_field, values_nullable)?;
51
52        Ok(Self {
53            data_type: data_type.clone(),
54            decoder,
55            values_nullable,
56            phantom: Default::default(),
57        })
58    }
59}
60
61impl<R: RunEndIndexType + Send> ArrayDecoder for RunEndEncodedArrayDecoder<R> {
62    fn decode(&mut self, tape: &Tape<'_>, pos: &[u32]) -> Result<ArrayRef, ArrowError> {
63        let len = pos.len();
64        if len == 0 {
65            return Ok(new_empty_array(&self.data_type));
66        }
67
68        let flat_array = self.decoder.decode(tape, pos)?;
69        if !self.values_nullable && flat_array.logical_null_count() != 0 {
70            return Err(ArrowError::JsonError(
71                "Encountered nulls in non-nullable values of RunEndEncoded".to_string(),
72            ));
73        }
74
75        let partitions = partition(from_ref(&flat_array))?;
76        let size = partitions.len();
77        let mut run_ends = Vec::with_capacity(size);
78        let mut indices = Vec::with_capacity(size);
79
80        for Range { start, end } in partitions.ranges() {
81            let run_end = R::Native::from_usize(end).ok_or_else(|| {
82                ArrowError::JsonError(format!(
83                    "Run end value {end} exceeds {:?} range",
84                    R::DATA_TYPE
85                ))
86            })?;
87            run_ends.push(run_end);
88            indices.push(start);
89        }
90
91        let indices = UInt32Array::from_iter_values(indices.into_iter().map(|i| i as u32));
92        let values = take(flat_array.as_ref(), &indices, None)?;
93
94        // SAFETY: run_ends are strictly increasing with the last value equal to len
95        let run_ends = unsafe { RunEndBuffer::new_unchecked(ScalarBuffer::from(run_ends), 0, len) };
96
97        // SAFETY: run_ends are valid and values has the same length as run_ends
98        let array =
99            unsafe { RunArray::<R>::new_unchecked(self.data_type.clone(), run_ends, values) };
100        Ok(Arc::new(array))
101    }
102}
arrow_json/reader/run_end_array.rs

arrow_json/reader/
run_end_array.rs
