Skip to content
Merged
Show file tree
Hide file tree
Changes from all 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
608 changes: 396 additions & 212 deletions Cargo.lock

Large diffs are not rendered by default.

1 change: 1 addition & 0 deletions pyproject.toml
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@ requires-python = ">=3.12"

dependencies = [
"pyarrow~=22.0",
"matplotlib~=3.10.8",
"numpy~=2.3",
"ruff~=0.14.6",
"polars~=1.35",
Expand Down
9 changes: 7 additions & 2 deletions serde_arrow/src/internal/deserialization/bool_deserializer.rs
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
use marrow::view::BooleanView;
use serde::de::Visitor;

use crate::internal::error::{fail, set_default, try_, Context, ContextSupport, Error, Result};
use crate::internal::error::{
fail, set_default, try_, Context, ContextSupport, Error, ErrorKind, Result,
};

use super::{random_access_deserializer::RandomAccessDeserializer, utils::bitset_is_set};

Expand Down Expand Up @@ -32,7 +34,10 @@ impl<'a> BoolDeserializer<'a> {
if let Some(value) = self.get(idx)? {
Ok(value)
} else {
fail!("Required value was not defined");
Err(Error::new(
ErrorKind::NullabilityViolation { field: None },
"Required value was not defined".into(),
))
}
}
}
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@ use marrow::view::{BytesView, PrimitiveView};
use serde::de::Visitor;

use crate::internal::{
error::{fail, set_default, try_, Context, ContextSupport, Result},
error::{set_default, try_, Context, ContextSupport, Error, ErrorKind, Result},
utils::{array_view_ext::ViewAccess, Offset},
};

Expand All @@ -21,7 +21,11 @@ impl<'a, K: Integer, V: Offset> DictionaryDeserializer<'a, K, V> {
pub fn new(path: String, keys: PrimitiveView<'a, K>, values: BytesView<'a, V>) -> Result<Self> {
if let Some(validity) = &values.validity {
if !bitset_all_set(validity, values.offsets.len() - 1)? {
fail!("Null for non-nullable type: dictionaries do not support nullable values");
return Err(Error::new(
ErrorKind::NullabilityViolation { field: None },
"Null for non-nullable type: dictionaries do not support nullable values"
.into(),
));
}
}
Ok(Self {
Expand Down
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
use marrow::view::FixedSizeBinaryView;
use serde::de::Visitor;

use crate::internal::error::{fail, set_default, try_, Context, ContextSupport, Result};
use crate::internal::error::{
fail, set_default, try_, Context, ContextSupport, Error, ErrorKind, Result,
};

use super::{
random_access_deserializer::RandomAccessDeserializer,
Expand Down Expand Up @@ -53,7 +55,10 @@ impl<'a> FixedSizeBinaryDeserializer<'a> {

pub fn get_required(&self, idx: usize) -> Result<&'a [u8]> {
let Some(s) = self.get(idx)? else {
fail!("Required value is not defined");
return Err(Error::new(
ErrorKind::NullabilityViolation { field: None },
"Required value is not defined".into(),
));
};
Ok(s)
}
Expand Down
7 changes: 5 additions & 2 deletions serde_arrow/src/internal/utils/array_view_ext.rs
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@ use marrow::view::{BytesView, BytesViewView, PrimitiveView, View};

use crate::internal::{
deserialization::utils::bitset_is_set,
error::{fail, Result},
error::{fail, Error, ErrorKind, Result},
};

use super::Offset;
Expand Down Expand Up @@ -109,7 +109,10 @@ pub trait ViewAccess<'a, Item: ?Sized + 'a> {
if let Some(val) = self.get(idx)? {
Ok(val)
} else {
fail!("Required item was not present");
Err(Error::new(
ErrorKind::NullabilityViolation { field: None },
"Required item was not present".into(),
))
}
}

Expand Down
593 changes: 473 additions & 120 deletions uv.lock

Large diffs are not rendered by default.

20 changes: 8 additions & 12 deletions x.py
Original file line number Diff line number Diff line change
@@ -1,7 +1,3 @@
# /// script
# requires-python = ">=3.11"
# dependencies = []
# ///
self_path = __import__("pathlib").Path(__file__).parent.resolve()
python = __import__("shlex").quote(__import__("sys").executable)

Expand Down Expand Up @@ -135,7 +131,8 @@ def update_workflows():


def _update_workflow(path, template):
import copy, json
import copy
import json

workflow = copy.deepcopy(template)

Expand Down Expand Up @@ -165,8 +162,8 @@ def _generate_workflow_check_steps():
}

yield {
"name": f"Check support packages",
"run": f"cargo check --all-features --package bench --package example --package integration_tests",
"name": "Check support packages",
"run": "cargo check --all-features --package bench --package example --package integration_tests",
}

yield {
Expand Down Expand Up @@ -363,10 +360,8 @@ def check_cargo_toml():
@cmd(help="Run the benchmarks")
@arg("--quick", action="store_true", default=False)
def bench(quick=False):
import os

_sh(
f"cargo bench -p bench",
"cargo bench -p bench",
env=({"SERDE_ARROW_BENCH_QUICK": "1"} if quick else {}),
)
summarize_bench()
Expand Down Expand Up @@ -394,7 +389,8 @@ def summarize_bench(update=False):


def load_times():
import json, statistics
import json
import statistics

root = self_path / "target" / "criterion/"

Expand Down Expand Up @@ -472,7 +468,7 @@ def plot_times(mean_times, ignore_groups=()):
(
pl.col("time")
/ pl.col("time")
.where(
.filter(
pl.col("impl")
== benchmark_renames.get("arrow2_convert", "arrow2_convert")
)
Expand Down