Skip to content

Commit

Permalink
feat: Support all decompression types (#20)
Browse files Browse the repository at this point in the history
* Support all decompression types

* Move default compression block size to const
  • Loading branch information
Jefffrey committed Nov 4, 2023
1 parent abd9960 commit cbc6371
Show file tree
Hide file tree
Showing 18 changed files with 332 additions and 55 deletions.
64 changes: 64 additions & 0 deletions Cargo.lock

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

3 changes: 3 additions & 0 deletions Cargo.toml
Original file line number Diff line number Diff line change
Expand Up @@ -25,9 +25,12 @@ flate2 = "1"
futures = { version = "0.3", default-features = false, features = ["std"] }
futures-util = "0.3"
lazy_static = "1.4"
lz4_flex = "0.11"
lzokay-native = "0.1"
paste = "1.0"
prost = { version = "0.11" }
snafu = "0.7"
snap = "1.1"
tokio = { version = "1.28", features = [
"io-util",
"sync",
Expand Down
6 changes: 5 additions & 1 deletion src/arrow_reader.rs
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@ use crate::arrow_reader::column::timestamp::new_timestamp_iter;
use crate::arrow_reader::column::NullableIterator;
use crate::error::{self, Result};
use crate::proto::{StripeFooter, StripeInformation};
use crate::reader::decompress::Compression;
use crate::reader::schema::{create_field, TypeDescription};
use crate::reader::Reader;

Expand Down Expand Up @@ -427,7 +428,10 @@ impl Stripe {
) -> Result<Self> {
let footer = Arc::new(r.stripe_footer(stripe).clone());

let compression = r.metadata().postscript.compression();
let compression = Compression::from_proto(
r.metadata().postscript.compression(),
r.metadata().postscript.compression_block_size,
);
//TODO(weny): add tz
let columns = columns
.iter()
Expand Down
10 changes: 5 additions & 5 deletions src/arrow_reader/column.rs
Original file line number Diff line number Diff line change
Expand Up @@ -7,8 +7,8 @@ use tokio::io::{AsyncRead, AsyncReadExt, AsyncSeek, AsyncSeekExt};

use crate::error::{self, Result};
use crate::proto::stream::Kind;
use crate::proto::{ColumnEncoding, CompressionKind, StripeFooter, StripeInformation};
use crate::reader::decompress::Decompressor;
use crate::proto::{ColumnEncoding, StripeFooter, StripeInformation};
use crate::reader::decompress::{Compression, Decompressor};
use crate::reader::schema::TypeDescription;
use crate::reader::Reader;

Expand All @@ -25,7 +25,7 @@ pub mod timestamp;
pub struct Column {
data: Bytes,
number_of_rows: u64,
compression: CompressionKind,
compression: Option<Compression>,
footer: Arc<StripeFooter>,
name: String,
column: Arc<TypeDescription>,
Expand Down Expand Up @@ -99,7 +99,7 @@ impl Column {

pub fn new<R: Read + Seek>(
reader: &mut Reader<R>,
compression: CompressionKind,
compression: Option<Compression>,
name: &str,
column: &Arc<TypeDescription>,
footer: &Arc<StripeFooter>,
Expand All @@ -120,7 +120,7 @@ impl Column {

pub async fn new_async<R: AsyncRead + AsyncSeek + Unpin + Send>(
reader: &mut Reader<R>,
compression: CompressionKind,
compression: Option<Compression>,
name: &str,
column: &Arc<TypeDescription>,
footer: &Arc<StripeFooter>,
Expand Down
15 changes: 9 additions & 6 deletions src/arrow_reader/column/date.rs
Original file line number Diff line number Diff line change
Expand Up @@ -14,12 +14,15 @@ pub struct DateIterator {
}

pub fn convert_date(data: i64) -> Result<NaiveDate> {
let date = NaiveDate::from_ymd_opt(1970, 1, 1)
.context(error::InvalidDateSnafu)?
.checked_add_days(Days::new(data as u64))
.context(error::AddDaysSnafu)?;

Ok(date)
let days = Days::new(data.unsigned_abs());
// safe unwrap as is valid date
let epoch = NaiveDate::from_ymd_opt(1970, 1, 1).unwrap();
let date = if data.is_negative() {
epoch.checked_sub_days(days)
} else {
epoch.checked_add_days(days)
};
date.context(error::AddDaysSnafu)
}

impl Iterator for DateIterator {
Expand Down
6 changes: 5 additions & 1 deletion src/async_arrow_reader.rs
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@ use crate::arrow_reader::{
};
use crate::error::Result;
use crate::proto::StripeInformation;
use crate::reader::decompress::Compression;
use crate::reader::schema::TypeDescription;
use crate::reader::Reader;

Expand Down Expand Up @@ -188,7 +189,10 @@ impl Stripe {
) -> Result<Self> {
let footer = Arc::new(r.stripe_footer(stripe).clone());

let compression = r.metadata().postscript.compression();
let compression = Compression::from_proto(
r.metadata().postscript.compression(),
r.metadata().postscript.compression_block_size,
);
//TODO(weny): add tz
let mut columns = Vec::with_capacity(column_defs.len());
for (name, typ) in column_defs.iter() {
Expand Down
18 changes: 18 additions & 0 deletions src/error.rs
Original file line number Diff line number Diff line change
Expand Up @@ -87,6 +87,24 @@ pub enum Error {
location: Location,
source: io::Error,
},

#[snafu(display("Failed to build snappy decoder: {}", source))]
BuildSnappyDecoder {
location: Location,
source: snap::Error,
},

#[snafu(display("Failed to build lzo decoder: {}", source))]
BuildLzoDecoder {
location: Location,
source: lzokay_native::Error,
},

#[snafu(display("Failed to build lz4 decoder: {}", source))]
BuildLz4Decoder {
location: Location,
source: lz4_flex::block::DecompressError,
},
}

pub type Result<T> = std::result::Result<T, Error>;
Loading

0 comments on commit cbc6371

Please sign in to comment.