parquet_variant/
utils.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.
17use std::{array::TryFromSliceError, ops::Range, str};
18
19use arrow_schema::ArrowError;
20
21use std::cmp::Ordering;
22use std::fmt::Debug;
23use std::slice::SliceIndex;
24
25/// Helper for reporting integer overflow errors in a consistent way.
26pub(crate) fn overflow_error(msg: &str) -> ArrowError {
27    ArrowError::InvalidArgumentError(format!("Integer overflow computing {msg}"))
28}
29
30#[inline]
31pub(crate) fn slice_from_slice<I: SliceIndex<[u8]> + Clone + Debug>(
32    bytes: &[u8],
33    index: I,
34) -> Result<&I::Output, ArrowError> {
35    bytes.get(index.clone()).ok_or_else(|| {
36        ArrowError::InvalidArgumentError(format!(
37            "Tried to extract byte(s) {index:?} from {}-byte buffer",
38            bytes.len(),
39        ))
40    })
41}
42
43/// Helper to safely slice bytes with offset calculations.
44///
45/// Equivalent to `slice_from_slice(bytes, (base_offset + range.start)..(base_offset + range.end))`
46/// but using checked addition to prevent integer overflow panics on 32-bit systems.
47#[inline]
48pub(crate) fn slice_from_slice_at_offset(
49    bytes: &[u8],
50    base_offset: usize,
51    range: Range<usize>,
52) -> Result<&[u8], ArrowError> {
53    let start_byte = base_offset
54        .checked_add(range.start)
55        .ok_or_else(|| overflow_error("slice start"))?;
56    let end_byte = base_offset
57        .checked_add(range.end)
58        .ok_or_else(|| overflow_error("slice end"))?;
59    slice_from_slice(bytes, start_byte..end_byte)
60}
61
62pub(crate) fn array_from_slice<const N: usize>(
63    bytes: &[u8],
64    offset: usize,
65) -> Result<[u8; N], ArrowError> {
66    slice_from_slice_at_offset(bytes, offset, 0..N)?
67        .try_into()
68        .map_err(|e: TryFromSliceError| ArrowError::InvalidArgumentError(e.to_string()))
69}
70
71pub(crate) fn first_byte_from_slice(slice: &[u8]) -> Result<u8, ArrowError> {
72    slice
73        .first()
74        .copied()
75        .ok_or_else(|| ArrowError::InvalidArgumentError("Received empty bytes".to_string()))
76}
77
78/// Helper to get a &str from a slice at the given offset and range, or an error if it contains invalid UTF-8 data.
79#[inline]
80pub(crate) fn string_from_slice(
81    slice: &[u8],
82    offset: usize,
83    range: Range<usize>,
84) -> Result<&str, ArrowError> {
85    let offset_buffer = slice_from_slice_at_offset(slice, offset, range)?;
86
87    //Use simdutf8 by default
88    #[cfg(feature = "simdutf8")]
89    {
90        simdutf8::basic::from_utf8(offset_buffer).map_err(|_| {
91            // Use simdutf8::compat to return details about the decoding error
92            let e = simdutf8::compat::from_utf8(offset_buffer).unwrap_err();
93            ArrowError::InvalidArgumentError(format!("encountered non UTF-8 data: {e}"))
94        })
95    }
96
97    //Use std::str if simdutf8 is not enabled
98    #[cfg(not(feature = "simdutf8"))]
99    str::from_utf8(offset_buffer)
100        .map_err(|_| ArrowError::InvalidArgumentError("invalid UTF-8 string".to_string()))
101}
102
103/// Performs a binary search over a range using a fallible key extraction function; a failed key
104/// extraction immediately terminats the search.
105///
106/// This is similar to the standard library's `binary_search_by`, but generalized to ranges instead
107/// of slices.
108///
109/// # Arguments
110/// * `range` - The range to search in
111/// * `target` - The target value to search for
112/// * `key_extractor` - A function that extracts a comparable key from slice elements.
113///   This function can fail and return None.
114///
115/// # Returns
116/// * `Some(Ok(index))` - Element found at the given index
117/// * `Some(Err(index))` - Element not found, but would be inserted at the given index
118/// * `None` - Key extraction failed
119pub(crate) fn try_binary_search_range_by<F>(
120    range: Range<usize>,
121    cmp: F,
122) -> Option<Result<usize, usize>>
123where
124    F: Fn(usize) -> Option<Ordering>,
125{
126    let Range { mut start, mut end } = range;
127    while start < end {
128        let mid = start + (end - start) / 2;
129        match cmp(mid)? {
130            Ordering::Equal => return Some(Ok(mid)),
131            Ordering::Greater => end = mid,
132            Ordering::Less => start = mid + 1,
133        }
134    }
135
136    Some(Err(start))
137}
138
139/// Verifies the expected size of type T, for a type that should only grow if absolutely necessary.
140#[allow(unused)]
141pub(crate) const fn expect_size_of<T>(expected: usize) {
142    let size = std::mem::size_of::<T>();
143    if size != expected {
144        let _ = [""; 0][size];
145    }
146}
parquet_variant/utils.rs

parquet_variant/
utils.rs
