Skip to content

perf(onpair): adopt onpair 0.2 storage-backed dictionary and safety-only validation #9117

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

Open
gargiulofrancesco wants to merge 10 commits into
base: develop
Choose a base branch
from
Open

perf(onpair): adopt onpair 0.2 storage-backed dictionary and safety-only validation #9117

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
10 commits
Select commit Hold shift + click to select a range
f7bfd60
feat(onpair): preserve data when rebuilding arrays
gargiulofrancesco Jul 21, 2026
fb91f0f
feat(onpair): cache storage-backed dictionary
gargiulofrancesco Jul 21, 2026
2a4be65
feat(onpair): use safety-only dictionary validation
gargiulofrancesco Jul 22, 2026
767906f
fix(onpair): use upstream default configuration
gargiulofrancesco Jul 22, 2026
ea6ba2c
build: use onpair develop
gargiulofrancesco Jul 23, 2026
141dd76
build: use released onpair
gargiulofrancesco Jul 30, 2026
e5976d2
Merge branch 'develop' into fg/onpair-improvements
gargiulofrancesco Jul 31, 2026
ad7ef3c
refactor(onpair): validate the compress-time dictionary
gargiulofrancesco Aug 3, 2026
e4a1edf
fix(onpair): error instead of panic on inconsistent uncompressed_lengths
gargiulofrancesco Aug 3, 2026
27c0f56
Merge branch 'develop' into fg/onpair-improvements
gargiulofrancesco Aug 3, 2026
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
6 changes: 3 additions & 3 deletions 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 Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -196,7 +196,7 @@ object_store = { version = "0.13.2", default-features = false }
object_store_opendal = "0.57.0"
once_cell = "1.21"
oneshot = { version = "0.2.0", features = ["async"] }
onpair = "0.1.1"
onpair = "0.2.0"
opendal = { version = "0.57.0", default-features = false }
opentelemetry = "0.32.0"
opentelemetry-otlp = "0.32.0"
Expand Down
4 changes: 2 additions & 2 deletions benchmarks/string-bench/src/codec.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,7 +23,7 @@ use vortex_bench::measurements::CustomUnitMeasurement;
use vortex_fsst::fsst_compress;
use vortex_fsst::fsst_train_compressor;
use vortex_onpair::Config;
use vortex_onpair::DEFAULT_DICT12_CONFIG;
use vortex_onpair::DEFAULT_CONFIG;
use vortex_onpair::MaxDictBits;
use vortex_onpair::onpair_compress;

Expand Down Expand Up @@ -59,7 +59,7 @@ impl DirectCandidate {
})?;
Ok(Self::OnPair(Config {
max_dict_bits,
..DEFAULT_DICT12_CONFIG
..DEFAULT_CONFIG
}))
}

Expand Down
4 changes: 2 additions & 2 deletions benchmarks/string-bench/src/serialized.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,7 @@ use vortex_btrblocks::schemes::string::FSSTScheme;
use vortex_btrblocks::schemes::string::NullDominatedSparseScheme;
use vortex_btrblocks::schemes::string::OnPairScheme;
use vortex_btrblocks::schemes::string::StringDictScheme;
use vortex_onpair::DEFAULT_DICT12_CONFIG;
use vortex_onpair::DEFAULT_CONFIG;

use crate::StringColumn;
use crate::StringEncoder;
Expand Down Expand Up @@ -87,7 +87,7 @@ fn serialized_encoder_label(encoder: StringEncoder) -> String {
match encoder {
// The config `OnPairScheme` compresses with. When btrblocks gains a
// configurable budget, pass the benchmark's own config here.
StringEncoder::OnPair => onpair_label(&DEFAULT_DICT12_CONFIG),
StringEncoder::OnPair => onpair_label(&DEFAULT_CONFIG),
StringEncoder::Fsst => encoder.label().to_string(),
}
}
Expand Down
11 changes: 7 additions & 4 deletions encodings/onpair/benches/decode.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,7 +45,7 @@ use vortex_array::dtype::NativePType;
use vortex_array::dtype::Nullability;
use vortex_buffer::Buffer;
use vortex_mask::Mask;
use vortex_onpair::DEFAULT_DICT12_CONFIG;
use vortex_onpair::DEFAULT_CONFIG;
use vortex_onpair::OnPair;
use vortex_onpair::OnPairArray;
use vortex_onpair::OnPairArraySlotsExt;
Expand Down Expand Up @@ -162,7 +162,7 @@ fn compress(n: usize, shape: Shape, ctx: &mut ExecutionCtx) -> OnPairArray {
strings.iter().map(|s| Some(s.as_bytes())),
DType::Utf8(Nullability::NonNullable),
);
onpair_compress(varbin.as_array(), DEFAULT_DICT12_CONFIG, ctx)
onpair_compress(varbin.as_array(), DEFAULT_CONFIG, ctx)
.unwrap_or_else(|e| panic!("onpair_compress failed: {e}"))
.try_downcast::<OnPair>()
.unwrap_or_else(|array| panic!("expected OnPair array, got {}", array.encoding_id()))
Expand All @@ -181,8 +181,11 @@ fn materialise(arr: &OnPairArray, ctx: &mut ExecutionCtx) -> (DecodeInputs, usiz
let view = arr.as_view();
let dict_offsets = widen::<u32>(view.dict_offsets(), ctx);
let dict_bytes = view.dict_bytes_handle().clone();
CompactDictionaryView::validate(dict_bytes.as_host().as_slice(), dict_offsets.as_slice())
.expect("valid OnPair dictionary");
CompactDictionaryView::validate_safety(
dict_bytes.as_host().as_slice(),
dict_offsets.as_slice(),
)
.expect("valid OnPair dictionary");
let inputs = DecodeInputs {
dict_bytes,
dict_offsets,
Expand Down
Loading
Loading