Skip to content
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.

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

feat: support to read tinyint #22

Merged
merged 3 commits into from
Nov 5, 2023
Merged
Show file tree
Hide file tree
Changes from 2 commits
Commits
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
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,7 @@ Read [Apache ORC](https://orc.apache.org/) in Rust.
| Float, Double | ✓ | | f32, f64 | Float32, Float64 |
| String, Char, and VarChar | ✓ | | string | Utf8 |
| Boolean | ✓ | | bool | Boolean |
| TinyInt | | | | |
| TinyInt | | | i8 | Int8 |
| Binary | ✓ | | Vec\<u8\> | Binary |
| Decimal | ✗ | | | |
| Date | ✓ | | chrono::NaiveDate | Date32 |
Expand Down
9 changes: 8 additions & 1 deletion src/arrow_reader.rs
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@
use chrono::{Datelike, NaiveDate, NaiveDateTime};
use snafu::{OptionExt, ResultExt};

use self::column::tinyint::new_i8_iter;
use self::column::Column;
use crate::arrow_reader::column::binary::new_binary_iterator;
use crate::arrow_reader::column::boolean::new_boolean_iter;
Expand Down Expand Up @@ -117,6 +118,7 @@
Int64(NullableIterator<i64>),
Int32(NullableIterator<i64>),
Int16(NullableIterator<i64>),
Int8(NullableIterator<i8>),
Boolean(NullableIterator<bool>),
Float32(NullableIterator<f32>),
Float64(NullableIterator<f64>),
Expand Down Expand Up @@ -207,6 +209,7 @@
impl_decode_next_batch_cast!(i64, Int64Type);
impl_decode_next_batch_cast!(i32, Int32Type);
impl_decode_next_batch_cast!(i16, Int16Type);
impl_decode_next_batch!(i8);
impl_decode_next_batch!(f32);
impl_decode_next_batch!(f64);

Expand Down Expand Up @@ -238,6 +241,10 @@
Some(array) => fields.push(array),
None => break,
},
Decoder::Int8(decoder) => match decode_next_batch_i8(decoder, chunk)? {
Some(array) => fields.push(array),
None => break,

Check warning on line 246 in src/arrow_reader.rs

View check run for this annotation

Codecov / codecov/patch

src/arrow_reader.rs#L246

Added line #L246 was not covered by tests
},
Decoder::Float32(decoder) => match decode_next_batch_f32(decoder, chunk)? {
Some(array) => fields.push(array),
None => break,
Expand Down Expand Up @@ -328,7 +335,7 @@
for col in &stripe.columns {
let decoder = match col.kind() {
crate::proto::r#type::Kind::Boolean => Decoder::Boolean(new_boolean_iter(col)?),
crate::proto::r#type::Kind::Byte => todo!(),
crate::proto::r#type::Kind::Byte => Decoder::Int8(new_i8_iter(col)?),
crate::proto::r#type::Kind::Short => Decoder::Int16(new_i64_iter(col)?),
crate::proto::r#type::Kind::Int => Decoder::Int32(new_i64_iter(col)?),
crate::proto::r#type::Kind::Long => Decoder::Int64(new_i64_iter(col)?),
Expand Down
1 change: 1 addition & 0 deletions src/arrow_reader/column.rs
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@ pub mod int;
pub mod present;
pub mod string;
pub mod timestamp;
pub mod tinyint;

#[derive(Debug)]
pub struct Column {
Expand Down
26 changes: 26 additions & 0 deletions src/arrow_reader/column/tinyint.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,26 @@
use snafu::OptionExt;

use crate::arrow_reader::column::present::new_present_iter;
use crate::arrow_reader::column::{Column, NullableIterator};
use crate::error::{InvalidColumnSnafu, Result};
use crate::proto::stream::Kind;
use crate::reader::decode::byte_rle::ByteRleIter;

pub fn new_i8_iter(column: &Column) -> Result<NullableIterator<i8>> {
let present = new_present_iter(column)?.collect::<Result<Vec<_>>>()?;
let rows: usize = present.iter().filter(|&p| *p).count();

let iter = column
.stream(Kind::Data)
.transpose()?
.map(|reader| {
Box::new(ByteRleIter::new(reader, rows).map(|value| value.map(|value| value as i8)))
as _
})
.context(InvalidColumnSnafu { name: &column.name })?;

Ok(NullableIterator {
present: Box::new(present.into_iter()),
iter,
})
}
1 change: 1 addition & 0 deletions src/reader/decode.rs
Original file line number Diff line number Diff line change
@@ -1,4 +1,5 @@
pub mod boolean_rle;
pub mod byte_rle;
pub mod float;
pub mod rle_v2;
mod util;
Expand Down
127 changes: 127 additions & 0 deletions src/reader/decode/byte_rle.rs
Original file line number Diff line number Diff line change
@@ -0,0 +1,127 @@
use crate::error::Result;
use std::io::Read;

use super::util::read_u8;

const MAX_LITERAL_SIZE: usize = 128;
const MIN_REPEAT_SIZE: usize = 3;

pub struct ByteRleIter<R: Read> {
reader: R,
literals: [u8; MAX_LITERAL_SIZE],
next_byte: Option<u8>,
num_literals: usize,
used: usize,
repeat: bool,
min_repeat_size: usize,
remaining: usize,
}

impl<R: Read> ByteRleIter<R> {
pub fn new(reader: R, length: usize) -> Self {
Self {
reader,
literals: [0u8; MAX_LITERAL_SIZE],
next_byte: None,
WenyXu marked this conversation as resolved.
Show resolved Hide resolved
num_literals: 0,
used: 0,
repeat: false,
min_repeat_size: MIN_REPEAT_SIZE,
WenyXu marked this conversation as resolved.
Show resolved Hide resolved
remaining: length,
}
}

pub fn into_inner(self) -> R {
self.reader
}

Check warning on line 36 in src/reader/decode/byte_rle.rs

View check run for this annotation

Codecov / codecov/patch

src/reader/decode/byte_rle.rs#L34-L36

Added lines #L34 - L36 were not covered by tests

fn read_byte(&mut self) -> Result<u8> {
if let Some(byt) = self.next_byte.take() {
Ok(byt)

Check warning on line 40 in src/reader/decode/byte_rle.rs

View check run for this annotation

Codecov / codecov/patch

src/reader/decode/byte_rle.rs#L40

Added line #L40 was not covered by tests
} else {
read_u8(&mut self.reader)
}
}

fn read_values(&mut self) -> Result<()> {
let control = self.read_byte()?;
self.used = 0;
if control < 0x80 {
self.repeat = true;
self.num_literals = control as usize + self.min_repeat_size;
let val = self.read_byte()?;
self.literals[0] = val;
} else {
self.repeat = false;
self.num_literals = 0x100 - control as usize;
for i in 0..self.num_literals {
let result = self.read_byte()?;
self.literals[i] = result;
}
}
Ok(())
}
}

impl<R: Read> Iterator for ByteRleIter<R> {
type Item = Result<u8>;

fn next(&mut self) -> Option<Self::Item> {
if self.remaining == 0 {
return None;
}
if self.used == self.num_literals {
match self.read_values() {
Ok(_) => {}
Err(err) => return Some(Err(err)),

Check warning on line 76 in src/reader/decode/byte_rle.rs

View check run for this annotation

Codecov / codecov/patch

src/reader/decode/byte_rle.rs#L76

Added line #L76 was not covered by tests
}
}

let result = if self.repeat {
self.literals[0]
} else {
self.literals[self.used]
};
self.used += 1;
self.remaining -= 1;
Some(Ok(result))
}
}

#[cfg(test)]
mod test {
use super::*;

#[test]
fn reader_test() {
let data = [0x61u8, 0x00];

let data = &mut data.as_ref();

let iter = ByteRleIter::new(data, 100)
.collect::<Result<Vec<_>>>()
.unwrap();

assert_eq!(iter, vec![0; 100]);

let data = [0x01, 0x01];

let data = &mut data.as_ref();

let iter = ByteRleIter::new(data, 4)
.collect::<Result<Vec<_>>>()
.unwrap();

assert_eq!(iter, vec![1; 4]);

let data = [0xfe, 0x44, 0x45];

let data = &mut data.as_ref();

let iter = ByteRleIter::new(data, 2)
.collect::<Result<Vec<_>>>()
.unwrap();

assert_eq!(iter, vec![0x44, 0x45]);
}
}
Binary file modified tests/basic/data/test.orc
Binary file not shown.
5 changes: 4 additions & 1 deletion tests/basic/data/write.py
Original file line number Diff line number Diff line change
Expand Up @@ -22,7 +22,8 @@
"utf8_increase": ["a", "bb", "ccc", "dddd", "eeeee"],
"utf8_decrease": ["eeeee", "dddd", "ccc", "bb", "a"],
"timestamp_simple": [datetime.datetime(2023, 4, 1, 20, 15, 30, 2000), datetime.datetime.fromtimestamp(int('1629617204525777000')/1000000000), datetime.datetime(2023, 1, 1), datetime.datetime(2023, 2, 1), datetime.datetime(2023, 3, 1)],
"date_simple": [datetime.date(2023, 4, 1), datetime.date(2023, 3, 1), datetime.date(2023, 1, 1), datetime.date(2023, 2, 1), datetime.date(2023, 3, 1)]
"date_simple": [datetime.date(2023, 4, 1), datetime.date(2023, 3, 1), datetime.date(2023, 1, 1), datetime.date(2023, 2, 1), datetime.date(2023, 3, 1)],
"tinyint_simple": [-1, None, 1, 127, -127]
}

def infer_schema(data):
Expand All @@ -48,6 +49,8 @@ def infer_schema(data):
dt = "double"
if key.startswith("bigint"):
dt = "bigint"
if key.startswith("tinyint"):
dt = "tinyint"
schema += key + ":" + dt + ","

schema = schema[:-1] + ">"
Expand Down
37 changes: 18 additions & 19 deletions tests/basic/main.rs
Original file line number Diff line number Diff line change
Expand Up @@ -229,15 +229,15 @@ pub fn basic_test_0() {
let reader = new_arrow_reader_root(&path);
let batch = reader.collect::<Result<Vec<_>, _>>().unwrap();

let expected = r#"+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+
| a | b | str_direct | d | e | f | int_short_repeated | int_neg_short_repeated | int_delta | int_neg_delta | int_direct | int_neg_direct | bigint_direct | bigint_neg_direct | bigint_other | utf8_increase | utf8_decrease | timestamp_simple | date_simple |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+
| 1.0 | true | a | a | ddd | aaaaa | 5 | -5 | 1 | 5 | 1 | -1 | 1 | -1 | 5 | a | eeeee | 2023-04-01T20:15:30.002 | 2023-04-01 |
| 2.0 | false | cccccc | bb | cc | bbbbb | 5 | -5 | 2 | 4 | 6 | -6 | 6 | -6 | -5 | bb | dddd | 2021-08-22T07:26:44.525777 | 2023-03-01 |
| | | | | | | | | | | | | | | 1 | ccc | ccc | 2023-01-01T00:00:00 | 2023-01-01 |
| 4.0 | true | ddd | ccc | bb | ccccc | 5 | -5 | 4 | 2 | 3 | -3 | 3 | -3 | 5 | dddd | bb | 2023-02-01T00:00:00 | 2023-02-01 |
| 5.0 | false | ee | ddd | a | ddddd | 5 | -5 | 5 | 1 | 2 | -2 | 2 | -2 | 5 | eeeee | a | 2023-03-01T00:00:00 | 2023-03-01 |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+"#;
let expected = r#"+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+----------------+
| a | b | str_direct | d | e | f | int_short_repeated | int_neg_short_repeated | int_delta | int_neg_delta | int_direct | int_neg_direct | bigint_direct | bigint_neg_direct | bigint_other | utf8_increase | utf8_decrease | timestamp_simple | date_simple | tinyint_simple |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+----------------+
| 1.0 | true | a | a | ddd | aaaaa | 5 | -5 | 1 | 5 | 1 | -1 | 1 | -1 | 5 | a | eeeee | 2023-04-01T20:15:30.002 | 2023-04-01 | -1 |
| 2.0 | false | cccccc | bb | cc | bbbbb | 5 | -5 | 2 | 4 | 6 | -6 | 6 | -6 | -5 | bb | dddd | 2021-08-22T07:26:44.525777 | 2023-03-01 | |
| | | | | | | | | | | | | | | 1 | ccc | ccc | 2023-01-01T00:00:00 | 2023-01-01 | 1 |
| 4.0 | true | ddd | ccc | bb | ccccc | 5 | -5 | 4 | 2 | 3 | -3 | 3 | -3 | 5 | dddd | bb | 2023-02-01T00:00:00 | 2023-02-01 | 127 |
| 5.0 | false | ee | ddd | a | ddddd | 5 | -5 | 5 | 1 | 2 | -2 | 2 | -2 | 5 | eeeee | a | 2023-03-01T00:00:00 | 2023-03-01 | -127 |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+----------------+"#;
assert_eq!(
expected,
pretty::pretty_format_batches(&batch).unwrap().to_string()
Expand All @@ -250,16 +250,15 @@ pub async fn async_basic_test_0() {
let reader = new_arrow_stream_reader_root(&path).await;
let batch = reader.try_collect::<Vec<_>>().await.unwrap();

let expected = r#"+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+
| a | b | str_direct | d | e | f | int_short_repeated | int_neg_short_repeated | int_delta | int_neg_delta | int_direct | int_neg_direct | bigint_direct | bigint_neg_direct | bigint_other | utf8_increase | utf8_decrease | timestamp_simple | date_simple |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+
| 1.0 | true | a | a | ddd | aaaaa | 5 | -5 | 1 | 5 | 1 | -1 | 1 | -1 | 5 | a | eeeee | 2023-04-01T20:15:30.002 | 2023-04-01 |
| 2.0 | false | cccccc | bb | cc | bbbbb | 5 | -5 | 2 | 4 | 6 | -6 | 6 | -6 | -5 | bb | dddd | 2021-08-22T07:26:44.525777 | 2023-03-01 |
| | | | | | | | | | | | | | | 1 | ccc | ccc | 2023-01-01T00:00:00 | 2023-01-01 |
| 4.0 | true | ddd | ccc | bb | ccccc | 5 | -5 | 4 | 2 | 3 | -3 | 3 | -3 | 5 | dddd | bb | 2023-02-01T00:00:00 | 2023-02-01 |
| 5.0 | false | ee | ddd | a | ddddd | 5 | -5 | 5 | 1 | 2 | -2 | 2 | -2 | 5 | eeeee | a | 2023-03-01T00:00:00 | 2023-03-01 |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+"#;

let expected = r#"+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+----------------+
| a | b | str_direct | d | e | f | int_short_repeated | int_neg_short_repeated | int_delta | int_neg_delta | int_direct | int_neg_direct | bigint_direct | bigint_neg_direct | bigint_other | utf8_increase | utf8_decrease | timestamp_simple | date_simple | tinyint_simple |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+----------------+
| 1.0 | true | a | a | ddd | aaaaa | 5 | -5 | 1 | 5 | 1 | -1 | 1 | -1 | 5 | a | eeeee | 2023-04-01T20:15:30.002 | 2023-04-01 | -1 |
| 2.0 | false | cccccc | bb | cc | bbbbb | 5 | -5 | 2 | 4 | 6 | -6 | 6 | -6 | -5 | bb | dddd | 2021-08-22T07:26:44.525777 | 2023-03-01 | |
| | | | | | | | | | | | | | | 1 | ccc | ccc | 2023-01-01T00:00:00 | 2023-01-01 | 1 |
| 4.0 | true | ddd | ccc | bb | ccccc | 5 | -5 | 4 | 2 | 3 | -3 | 3 | -3 | 5 | dddd | bb | 2023-02-01T00:00:00 | 2023-02-01 | 127 |
| 5.0 | false | ee | ddd | a | ddddd | 5 | -5 | 5 | 1 | 2 | -2 | 2 | -2 | 5 | eeeee | a | 2023-03-01T00:00:00 | 2023-03-01 | -127 |
+-----+-------+------------+-----+-----+-------+--------------------+------------------------+-----------+---------------+------------+----------------+---------------+-------------------+--------------+---------------+---------------+----------------------------+-------------+----------------+"#;
assert_eq!(
expected,
pretty::pretty_format_batches(&batch).unwrap().to_string()
Expand Down
Loading