Skip to content

Update to arrow 25.0.0 #3844

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.

Sign up for GitHub

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

Merged
merged 1 commit into from
Oct 18, 2022
Merged

Update to arrow 25.0.0 #3844

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
4 changes: 2 additions & 2 deletions benchmarks/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,14 +32,14 @@ simd = ["datafusion/simd"]
snmalloc = ["snmalloc-rs"]

[dependencies]
arrow = "24.0.0"
arrow = "25.0.0"
datafusion = { path = "../datafusion/core" }
env_logger = "0.9"
futures = "0.3"
mimalloc = { version = "0.1", optional = true, default-features = false }
num_cpus = "1.13.0"
object_store = "0.5.0"
parquet = "24.0.0"
parquet = "25.0.0"
rand = "0.8.4"
serde = { version = "1.0.136", features = ["derive"] }
serde_json = "1.0.78"
Expand Down
25 changes: 13 additions & 12 deletions datafusion-cli/Cargo.lock
View file
Open in desktop

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

2 changes: 1 addition & 1 deletion datafusion-cli/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,7 @@ rust-version = "1.62"
readme = "README.md"

[dependencies]
arrow = "24.0.0"
arrow = "25.0.0"
clap = { version = "3", features = ["derive", "cargo"] }
datafusion = { path = "../datafusion/core", version = "13.0.0" }
dirs = "4.0.0"
Expand Down
4 changes: 2 additions & 2 deletions datafusion-examples/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,8 +34,8 @@ path = "examples/avro_sql.rs"
required-features = ["datafusion/avro"]

[dev-dependencies]
arrow = "24.0.0"
arrow-flight = "24.0.0"
arrow = "25.0.0"
arrow-flight = "25.0.0"
async-trait = "0.1.41"
datafusion = { path = "../datafusion/core" }
datafusion-common = { path = "../datafusion/common" }
Expand Down
4 changes: 2 additions & 2 deletions datafusion/common/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,10 +40,10 @@ pyarrow = ["pyo3", "arrow/pyarrow"]

[dependencies]
apache-avro = { version = "0.14", default-features = false, features = ["snappy"], optional = true }
arrow = { version = "24.0.0", default-features = false }
arrow = { version = "25.0.0", default-features = false }
cranelift-module = { version = "0.88.0", optional = true }
object_store = { version = "0.5.0", default-features = false, optional = true }
ordered-float = "3.0"
parquet = { version = "24.0.0", default-features = false, optional = true }
parquet = { version = "25.0.0", default-features = false, optional = true }
pyo3 = { version = "0.17.1", optional = true }
sqlparser = "0.25"
15 changes: 5 additions & 10 deletions datafusion/common/src/scalar.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,7 +34,6 @@ use arrow::{
TimestampSecondType, UInt16Type, UInt32Type, UInt64Type, UInt8Type,
DECIMAL128_MAX_PRECISION,
},
util::decimal::Decimal128,
};
use ordered_float::OrderedFloat;

Expand Down Expand Up @@ -1696,7 +1695,7 @@ impl ScalarValue {
if array.is_null(index) {
ScalarValue::Decimal128(None, precision, scale)
} else {
ScalarValue::Decimal128(Some(array.value(index).as_i128()), precision, scale)
ScalarValue::Decimal128(Some(array.value(index)), precision, scale)
}
}

Expand Down Expand Up @@ -1881,11 +1880,7 @@ impl ScalarValue {
}
match value {
None => array.is_null(index),
Some(v) => {
!array.is_null(index)
&& array.value(index)
== Decimal128::new(precision, scale, &v.to_le_bytes())
}
Some(v) => !array.is_null(index) && array.value(index) == *v,
}
}

Expand Down Expand Up @@ -2501,15 +2496,15 @@ mod tests {
let array = array.as_any().downcast_ref::<Decimal128Array>().unwrap();
assert_eq!(1, array.len());
assert_eq!(DataType::Decimal128(10, 1), array.data_type().clone());
assert_eq!(123i128, array.value(0).as_i128());
assert_eq!(123i128, array.value(0));

// decimal scalar to array with size
let array = decimal_value.to_array_of_size(10);
let array_decimal = array.as_any().downcast_ref::<Decimal128Array>().unwrap();
assert_eq!(10, array.len());
assert_eq!(DataType::Decimal128(10, 1), array.data_type().clone());
assert_eq!(123i128, array_decimal.value(0).as_i128());
assert_eq!(123i128, array_decimal.value(9).as_i128());
assert_eq!(123i128, array_decimal.value(0));
assert_eq!(123i128, array_decimal.value(9));
// test eq array
assert!(decimal_value.eq_array(&array, 1));
assert!(decimal_value.eq_array(&array, 5));
Expand Down
6 changes: 3 additions & 3 deletions datafusion/core/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -56,7 +56,7 @@ unicode_expressions = ["datafusion-physical-expr/regex_expressions", "datafusion
[dependencies]
ahash = { version = "0.8", default-features = false, features = ["runtime-rng"] }
apache-avro = { version = "0.14", optional = true }
arrow = { version = "24.0.0", features = ["prettyprint"] }
arrow = { version = "25.0.0", features = ["prettyprint"] }
async-compression = { version = "0.3.14", features = ["bzip2", "gzip", "futures-io", "tokio"] }
async-trait = "0.1.41"
bytes = "1.1"
Expand All @@ -81,7 +81,7 @@ num_cpus = "1.13.0"
object_store = "0.5.0"
ordered-float = "3.0"
parking_lot = "0.12"
parquet = { version = "24.0.0", features = ["arrow", "async"] }
parquet = { version = "25.0.0", features = ["arrow", "async"] }
paste = "^1.0"
percent-encoding = "2.2.0"
pin-project-lite = "^0.2.7"
Expand All @@ -98,7 +98,7 @@ url = "2.2"
uuid = { version = "1.0", features = ["v4"] }

[dev-dependencies]
arrow = { version = "24.0.0", features = ["prettyprint", "dyn_cmp_dict"] }
arrow = { version = "25.0.0", features = ["prettyprint", "dyn_cmp_dict"] }
async-trait = "0.1.53"
criterion = "0.4"
csv = "1.1.6"
Expand Down
2 changes: 1 addition & 1 deletion datafusion/core/fuzz-utils/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,6 @@ edition = "2021"
# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html

[dependencies]
arrow = { version = "24.0.0", features = ["prettyprint"] }
arrow = { version = "25.0.0", features = ["prettyprint"] }
env_logger = "0.9.0"
rand = "0.8"
14 changes: 4 additions & 10 deletions datafusion/core/src/physical_plan/hash_utils.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,29 +62,23 @@ fn hash_decimal128<'a>(
if array.null_count() == 0 {
if mul_col {
for (i, hash) in hashes_buffer.iter_mut().enumerate() {
Comment on lines 63 to 64
Copy link
Contributor

@Dandandan Dandandan Oct 15, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think we can we use hash_array here now instead?

Copy link
Contributor Author

@tustvold tustvold Oct 18, 2022

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Potentially, I think I would rather keep this as a simple upgrade PR, and address any simplifications that fall out of the new decimal abstractions as a follow up

*hash = combine_hashes(
random_state.hash_one(&array.value(i).as_i128()),
*hash,
);
*hash = combine_hashes(random_state.hash_one(&array.value(i)), *hash);
}
} else {
for (i, hash) in hashes_buffer.iter_mut().enumerate() {
*hash = random_state.hash_one(&array.value(i).as_i128());
*hash = random_state.hash_one(&array.value(i));
}
}
} else if mul_col {
for (i, hash) in hashes_buffer.iter_mut().enumerate() {
if !array.is_null(i) {
*hash = combine_hashes(
random_state.hash_one(&array.value(i).as_i128()),
*hash,
);
*hash = combine_hashes(random_state.hash_one(&array.value(i)), *hash);
}
}
} else {
for (i, hash) in hashes_buffer.iter_mut().enumerate() {
if !array.is_null(i) {
*hash = random_state.hash_one(&array.value(i).as_i128());
*hash = random_state.hash_one(&array.value(i));
}
}
}
Expand Down
11 changes: 7 additions & 4 deletions datafusion/core/src/test/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -242,14 +242,17 @@ pub fn table_with_decimal() -> Arc<dyn TableProvider> {
}

fn make_decimal() -> RecordBatch {
let mut decimal_builder = Decimal128Builder::with_capacity(20, 10, 3);
let mut decimal_builder = Decimal128Builder::with_capacity(20);
for i in 110000..110010 {
decimal_builder.append_value(i as i128).unwrap();
decimal_builder.append_value(i as i128);
}
for i in 100000..100010 {
decimal_builder.append_value(-i as i128).unwrap();
decimal_builder.append_value(-i as i128);
}
let array = decimal_builder.finish();
let array = decimal_builder
.finish()
.with_precision_and_scale(10, 3)
.unwrap();
let schema = Schema::new(vec![Field::new("c1", array.data_type().clone(), true)]);
RecordBatch::try_new(Arc::new(schema), vec![Arc::new(array)]).unwrap()
}
Expand Down
35 changes: 26 additions & 9 deletions datafusion/core/tests/sql/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -560,9 +560,9 @@ async fn register_tpch_csv_data(
DataType::Int64 => {
cols.push(Box::new(Int64Builder::with_capacity(records.len())))
}
DataType::Decimal128(p, s) => cols.push(Box::new(
Decimal128Builder::with_capacity(records.len(), *p, *s),
)),
DataType::Decimal128(_, _) => {
cols.push(Box::new(Decimal128Builder::with_capacity(records.len())))
}
_ => {
let msg = format!("Not implemented: {}", field.data_type());
Err(DataFusionError::Plan(msg))?
Expand Down Expand Up @@ -600,7 +600,7 @@ async fn register_tpch_csv_data(
.unwrap();
let val = val.trim().replace('.', "");
let value_i128 = val.parse::<i128>().unwrap();
sb.append_value(value_i128)?;
sb.append_value(value_i128);
}
_ => Err(DataFusionError::Plan(format!(
"Not implemented: {}",
Expand All @@ -609,7 +609,21 @@ async fn register_tpch_csv_data(
}
}
}
let cols: Vec<ArrayRef> = cols.iter_mut().map(|it| it.finish()).collect();
let cols: Vec<ArrayRef> = cols
.iter_mut()
.zip(schema.fields())
.map(|(it, field)| match field.data_type() {
DataType::Decimal128(p, s) => Arc::new(
it.as_any_mut()
.downcast_mut::<Decimal128Builder>()
.unwrap()
.finish()
.with_precision_and_scale(*p, *s)
.unwrap(),
),
_ => it.finish(),
})
.collect();

let batch = RecordBatch::try_new(Arc::clone(&schema), cols)?;

Expand Down Expand Up @@ -879,14 +893,17 @@ pub fn table_with_decimal() -> Arc<dyn TableProvider> {
}

fn make_decimal() -> RecordBatch {
let mut decimal_builder = Decimal128Builder::with_capacity(20, 10, 3);
let mut decimal_builder = Decimal128Builder::with_capacity(20);
for i in 110000..110010 {
decimal_builder.append_value(i as i128).unwrap();
decimal_builder.append_value(i as i128);
}
for i in 100000..100010 {
decimal_builder.append_value(-i as i128).unwrap();
decimal_builder.append_value(-i as i128);
}
let array = decimal_builder.finish();
let array = decimal_builder
.finish()
.with_precision_and_scale(10, 3)
.unwrap();
let schema = Schema::new(vec![Field::new("c1", array.data_type().clone(), true)]);
RecordBatch::try_new(Arc::new(schema), vec![Arc::new(array)]).unwrap()
}
Expand Down
2 changes: 1 addition & 1 deletion datafusion/expr/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ path = "src/lib.rs"

[dependencies]
ahash = { version = "0.8", default-features = false, features = ["runtime-rng"] }
arrow = { version = "24.0.0", default-features = false }
arrow = { version = "25.0.0", default-features = false }
datafusion-common = { path = "../common", version = "13.0.0" }
log = "^0.4"
sqlparser = "0.25"
2 changes: 1 addition & 1 deletion datafusion/jit/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,7 +36,7 @@ path = "src/lib.rs"
jit = []

[dependencies]
arrow = { version = "24.0.0", default-features = false }
arrow = { version = "25.0.0", default-features = false }
cranelift = "0.88.0"
cranelift-jit = "0.88.0"
cranelift-module = "0.88.0"
Expand Down
2 changes: 1 addition & 1 deletion datafusion/optimizer/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,7 @@ default = ["unicode_expressions"]
unicode_expressions = []

[dependencies]
arrow = { version = "24.0.0", features = ["prettyprint"] }
arrow = { version = "25.0.0", features = ["prettyprint"] }
async-trait = "0.1.41"
chrono = { version = "0.4", default-features = false }
datafusion-common = { path = "../common", version = "13.0.0" }
Expand Down
2 changes: 1 addition & 1 deletion datafusion/physical-expr/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,7 @@ unicode_expressions = ["unicode-segmentation"]

[dependencies]
ahash = { version = "0.8", default-features = false, features = ["runtime-rng"] }
arrow = { version = "24.0.0", features = ["prettyprint"] }
arrow = { version = "25.0.0", features = ["prettyprint"] }
blake2 = { version = "^0.10.2", optional = true }
blake3 = { version = "1.0", optional = true }
chrono = { version = "0.4", default-features = false }
Expand Down
6 changes: 3 additions & 3 deletions datafusion/physical-expr/src/aggregate/min_max.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -177,17 +177,17 @@ macro_rules! typed_min_max_batch_decimal128 {
for i in 1..array.len() {
result = result.$OP(array.value(i));
}
ScalarValue::Decimal128(Some(result.as_i128()), *$PRECISION, *$SCALE)
ScalarValue::Decimal128(Some(result), *$PRECISION, *$SCALE)
} else {
let mut result = 0_i128;
let mut has_value = false;
for i in 0..array.len() {
if !has_value && array.is_valid(i) {
has_value = true;
result = array.value(i).as_i128();
result = array.value(i);
}
if array.is_valid(i) {
result = result.$OP(array.value(i).as_i128());
result = result.$OP(array.value(i));
}
}
ScalarValue::Decimal128(Some(result), *$PRECISION, *$SCALE)
Expand Down
2 changes: 1 addition & 1 deletion datafusion/physical-expr/src/aggregate/sum.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -169,7 +169,7 @@ fn sum_decimal_batch(values: &ArrayRef, precision: u8, scale: u8) -> Result<Scal
}

let result = array.into_iter().fold(0_i128, |s, element| match element {
Some(v) => s + v.as_i128(),
Some(v) => s + v,
None => s,
});

Expand Down
Loading