Round out OnPair CI: widen-on-decode + public-api locks + lints

* `vortex-onpair`: the cascading compressor narrows the integer slot
  children to their tightest ptype (e.g. `codes` from u16 down to u8),
  so the decoder's `as_slice::<u16>()` was tripping a panic. Widen all
  three primitive children back to their canonical types
  (`Buffer<u16>` for codes, `Buffer<u32>` for both offsets) at
  materialisation time. Adds three round-trip tests in
  `vortex-btrblocks/tests/onpair_roundtrip.rs` that exercise the full
  compressor + decompressor on string arrays (non-nullable, nullable,
  and an empty-string-heavy edge case) — all three are green.

* Fix the two `unresolved link` rustdoc warnings on `OnPair::compress`
  by pointing at the actual entry point (`crate::onpair_compress`).

* `Cargo.toml`: re-sort `vortex-onpair` / `vortex-onpair-sys` into
  alphabetical order in `[workspace.dependencies]` so `taplo fmt
  --check` (= the `lint-toml` CI job) stops complaining.

* SPDX headers on the three CMake files
  (`encodings/onpair-sys/cmake/{CMakeLists.txt,onpair_pin.cmake,strip_boost.cmake}`)
  so the `reuse-check` job passes.

* Regenerate `public-api.lock` for `vortex-btrblocks` and add the two
  missing locks (`encodings/onpair{,-sys}/public-api.lock`).

Test results
  vortex-onpair                 7 unit + 1 100k smoke   all green
  vortex-btrblocks            36 unit + 3 doctests +
                              3 new onpair_roundtrip      all green

Signed-off-by: Claude <noreply@anthropic.com>

Author: claude

Cargo.toml

@@ -286,13 +286,13 @@ vortex-fastlanes = { version = "0.1.0", path = "./encodings/fastlanes", default-
 vortex-file = { version = "0.1.0", path = "./vortex-file", default-features = false }
 vortex-flatbuffers = { version = "0.1.0", path = "./vortex-flatbuffers", default-features = false }
 vortex-fsst = { version = "0.1.0", path = "./encodings/fsst", default-features = false }
-vortex-onpair = { version = "0.1.0", path = "./encodings/onpair", default-features = false }
-vortex-onpair-sys = { version = "0.1.0", path = "./encodings/onpair-sys", default-features = false }
 vortex-io = { version = "0.1.0", path = "./vortex-io", default-features = false }
 vortex-ipc = { version = "0.1.0", path = "./vortex-ipc", default-features = false }
 vortex-layout = { version = "0.1.0", path = "./vortex-layout", default-features = false }
 vortex-mask = { version = "0.1.0", path = "./vortex-mask", default-features = false }
 vortex-metrics = { version = "0.1.0", path = "./vortex-metrics", default-features = false }
+vortex-onpair = { version = "0.1.0", path = "./encodings/onpair", default-features = false }
+vortex-onpair-sys = { version = "0.1.0", path = "./encodings/onpair-sys", default-features = false }
 vortex-pco = { version = "0.1.0", path = "./encodings/pco", default-features = false }
 vortex-proto = { version = "0.1.0", path = "./vortex-proto", default-features = false }
 vortex-runend = { version = "0.1.0", path = "./encodings/runend", default-features = false }

encodings/onpair-sys/cmake/CMakeLists.txt

@@ -1,3 +1,6 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright the Vortex contributors
+
 cmake_minimum_required(VERSION 3.21)
 project(onpair_shim CXX)
 

encodings/onpair-sys/cmake/onpair_pin.cmake

@@ -1,3 +1,6 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright the Vortex contributors
+#
 # Pin of gargiulofrancesco/onpair_cpp consumed by FetchContent.
 # Bump `ONPAIR_CPP_TAG` to a full commit SHA when updating — never use a
 # branch name in CI, otherwise builds become non-reproducible.

encodings/onpair-sys/cmake/strip_boost.cmake

@@ -1,3 +1,6 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright the Vortex contributors
+#
 # Replaces boost::unordered_flat_{map,set} with std::unordered_{map,set}
 # in the fetched onpair_cpp source tree. Idempotent.
 #

encodings/onpair-sys/public-api.lock

@@ -0,0 +1,351 @@
+pub mod vortex_onpair_sys
+
+pub mod vortex_onpair_sys::ffi
+
+#[repr(u32)] pub enum vortex_onpair_sys::ffi::OnPairStatus
+
+pub vortex_onpair_sys::ffi::OnPairStatus::BadFormat = 2
+
+pub vortex_onpair_sys::ffi::OnPairStatus::Internal = 99
+
+pub vortex_onpair_sys::ffi::OnPairStatus::InvalidArg = 1
+
+pub vortex_onpair_sys::ffi::OnPairStatus::Ok = 0
+
+pub vortex_onpair_sys::ffi::OnPairStatus::Oom = 4
+
+pub vortex_onpair_sys::ffi::OnPairStatus::OutOfRange = 3
+
+impl vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::from_raw(u32) -> Self
+
+impl core::clone::Clone for vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::clone(&self) -> vortex_onpair_sys::OnPairStatus
+
+impl core::cmp::Eq for vortex_onpair_sys::OnPairStatus
+
+impl core::cmp::PartialEq for vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::eq(&self, &vortex_onpair_sys::OnPairStatus) -> bool
+
+impl core::fmt::Debug for vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::OnPairStatus
+
+impl core::marker::StructuralPartialEq for vortex_onpair_sys::OnPairStatus
+
+#[repr(C)] pub struct vortex_onpair_sys::ffi::OnPairColumnHandle
+
+#[repr(C)] pub struct vortex_onpair_sys::ffi::OnPairColumnParts
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::bits: u32
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::codes_boundaries: *const u32
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::codes_boundaries_len: usize
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::codes_packed: *const u64
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::codes_packed_u64_len: usize
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::dict_bytes: *const u8
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::dict_bytes_len: usize
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::dict_offsets: *const u32
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::dict_offsets_len: usize
+
+pub vortex_onpair_sys::ffi::OnPairColumnParts::num_rows: usize
+
+impl core::clone::Clone for vortex_onpair_sys::OnPairColumnParts
+
+pub fn vortex_onpair_sys::OnPairColumnParts::clone(&self) -> vortex_onpair_sys::OnPairColumnParts
+
+impl core::fmt::Debug for vortex_onpair_sys::OnPairColumnParts
+
+pub fn vortex_onpair_sys::OnPairColumnParts::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::OnPairColumnParts
+
+#[repr(C)] pub struct vortex_onpair_sys::ffi::OnPairTrainingConfig
+
+pub vortex_onpair_sys::ffi::OnPairTrainingConfig::bits: u32
+
+pub vortex_onpair_sys::ffi::OnPairTrainingConfig::seed: u64
+
+pub vortex_onpair_sys::ffi::OnPairTrainingConfig::threshold: f64
+
+impl core::clone::Clone for vortex_onpair_sys::OnPairTrainingConfig
+
+pub fn vortex_onpair_sys::OnPairTrainingConfig::clone(&self) -> vortex_onpair_sys::OnPairTrainingConfig
+
+impl core::fmt::Debug for vortex_onpair_sys::OnPairTrainingConfig
+
+pub fn vortex_onpair_sys::OnPairTrainingConfig::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::OnPairTrainingConfig
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_buffer_free(*mut u8, usize)
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_bits(*const vortex_onpair_sys::OnPairColumnHandle) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_compress(*const u8, *const u64, usize, vortex_onpair_sys::OnPairTrainingConfig, *mut *mut vortex_onpair_sys::OnPairColumnHandle) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_contains_into(*const vortex_onpair_sys::OnPairColumnHandle, *const u8, usize, *mut u8) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_decompress(*const vortex_onpair_sys::OnPairColumnHandle, usize, *mut u8, usize, *mut usize) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_decompress_capacity(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_deserialize(*const u8, usize, *mut *mut vortex_onpair_sys::OnPairColumnHandle) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_dict_bytes(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_dict_copy(*const vortex_onpair_sys::OnPairColumnHandle, *mut u8, usize, *mut u64) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_dict_size(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_equals_into(*const vortex_onpair_sys::OnPairColumnHandle, *const u8, usize, *mut u8) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_free(*mut vortex_onpair_sys::OnPairColumnHandle)
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_len(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_parts(*const vortex_onpair_sys::OnPairColumnHandle, *mut vortex_onpair_sys::OnPairColumnParts) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_serialize(*const vortex_onpair_sys::OnPairColumnHandle, *mut *mut u8, *mut usize) -> u32
+
+pub unsafe c fn vortex_onpair_sys::ffi::onpair_column_starts_with_into(*const vortex_onpair_sys::OnPairColumnHandle, *const u8, usize, *mut u8) -> u32
+
+pub enum vortex_onpair_sys::Error
+
+pub vortex_onpair_sys::Error::BadFormat
+
+pub vortex_onpair_sys::Error::Internal
+
+pub vortex_onpair_sys::Error::InvalidArg
+
+pub vortex_onpair_sys::Error::Oom
+
+pub vortex_onpair_sys::Error::OutOfRange
+
+impl core::clone::Clone for vortex_onpair_sys::Error
+
+pub fn vortex_onpair_sys::Error::clone(&self) -> vortex_onpair_sys::Error
+
+impl core::cmp::Eq for vortex_onpair_sys::Error
+
+impl core::cmp::PartialEq for vortex_onpair_sys::Error
+
+pub fn vortex_onpair_sys::Error::eq(&self, &vortex_onpair_sys::Error) -> bool
+
+impl core::error::Error for vortex_onpair_sys::Error
+
+impl core::fmt::Debug for vortex_onpair_sys::Error
+
+pub fn vortex_onpair_sys::Error::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::fmt::Display for vortex_onpair_sys::Error
+
+pub fn vortex_onpair_sys::Error::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::Error
+
+impl core::marker::StructuralPartialEq for vortex_onpair_sys::Error
+
+#[repr(u32)] pub enum vortex_onpair_sys::OnPairStatus
+
+pub vortex_onpair_sys::OnPairStatus::BadFormat = 2
+
+pub vortex_onpair_sys::OnPairStatus::Internal = 99
+
+pub vortex_onpair_sys::OnPairStatus::InvalidArg = 1
+
+pub vortex_onpair_sys::OnPairStatus::Ok = 0
+
+pub vortex_onpair_sys::OnPairStatus::Oom = 4
+
+pub vortex_onpair_sys::OnPairStatus::OutOfRange = 3
+
+impl vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::from_raw(u32) -> Self
+
+impl core::clone::Clone for vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::clone(&self) -> vortex_onpair_sys::OnPairStatus
+
+impl core::cmp::Eq for vortex_onpair_sys::OnPairStatus
+
+impl core::cmp::PartialEq for vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::eq(&self, &vortex_onpair_sys::OnPairStatus) -> bool
+
+impl core::fmt::Debug for vortex_onpair_sys::OnPairStatus
+
+pub fn vortex_onpair_sys::OnPairStatus::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::OnPairStatus
+
+impl core::marker::StructuralPartialEq for vortex_onpair_sys::OnPairStatus
+
+pub struct vortex_onpair_sys::Column
+
+impl vortex_onpair_sys::Column
+
+pub fn vortex_onpair_sys::Column::bits(&self) -> u32
+
+pub fn vortex_onpair_sys::Column::compress(&[u8], &[u64], vortex_onpair_sys::OnPairTrainingConfig) -> core::result::Result<Self, vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::contains_bitmap(&self, &[u8]) -> core::result::Result<alloc::vec::Vec<u8>, vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::decompress_row(&self, usize, &mut alloc::vec::Vec<u8>) -> core::result::Result<(), vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::dict(&self) -> core::result::Result<(alloc::vec::Vec<u8>, alloc::vec::Vec<u64>), vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::dict_bytes(&self) -> usize
+
+pub fn vortex_onpair_sys::Column::dict_size(&self) -> usize
+
+pub fn vortex_onpair_sys::Column::equals_bitmap(&self, &[u8]) -> core::result::Result<alloc::vec::Vec<u8>, vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::from_bytes(&[u8]) -> core::result::Result<Self, vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::is_empty(&self) -> bool
+
+pub fn vortex_onpair_sys::Column::len(&self) -> usize
+
+pub fn vortex_onpair_sys::Column::max_decompress_capacity(&self) -> usize
+
+pub unsafe fn vortex_onpair_sys::Column::raw(&self) -> *const core::ffi::c_void
+
+pub fn vortex_onpair_sys::Column::starts_with_bitmap(&self, &[u8]) -> core::result::Result<alloc::vec::Vec<u8>, vortex_onpair_sys::Error>
+
+pub fn vortex_onpair_sys::Column::to_bytes(&self) -> core::result::Result<alloc::vec::Vec<u8>, vortex_onpair_sys::Error>
+
+impl vortex_onpair_sys::Column
+
+pub fn vortex_onpair_sys::Column::parts(&self) -> core::result::Result<vortex_onpair_sys::Parts<'_>, vortex_onpair_sys::Error>
+
+impl core::marker::Send for vortex_onpair_sys::Column
+
+impl core::marker::Sync for vortex_onpair_sys::Column
+
+impl core::ops::drop::Drop for vortex_onpair_sys::Column
+
+pub fn vortex_onpair_sys::Column::drop(&mut self)
+
+#[repr(C)] pub struct vortex_onpair_sys::OnPairColumnHandle
+
+#[repr(C)] pub struct vortex_onpair_sys::OnPairColumnParts
+
+pub vortex_onpair_sys::OnPairColumnParts::bits: u32
+
+pub vortex_onpair_sys::OnPairColumnParts::codes_boundaries: *const u32
+
+pub vortex_onpair_sys::OnPairColumnParts::codes_boundaries_len: usize
+
+pub vortex_onpair_sys::OnPairColumnParts::codes_packed: *const u64
+
+pub vortex_onpair_sys::OnPairColumnParts::codes_packed_u64_len: usize
+
+pub vortex_onpair_sys::OnPairColumnParts::dict_bytes: *const u8
+
+pub vortex_onpair_sys::OnPairColumnParts::dict_bytes_len: usize
+
+pub vortex_onpair_sys::OnPairColumnParts::dict_offsets: *const u32
+
+pub vortex_onpair_sys::OnPairColumnParts::dict_offsets_len: usize
+
+pub vortex_onpair_sys::OnPairColumnParts::num_rows: usize
+
+impl core::clone::Clone for vortex_onpair_sys::OnPairColumnParts
+
+pub fn vortex_onpair_sys::OnPairColumnParts::clone(&self) -> vortex_onpair_sys::OnPairColumnParts
+
+impl core::fmt::Debug for vortex_onpair_sys::OnPairColumnParts
+
+pub fn vortex_onpair_sys::OnPairColumnParts::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::OnPairColumnParts
+
+#[repr(C)] pub struct vortex_onpair_sys::OnPairTrainingConfig
+
+pub vortex_onpair_sys::OnPairTrainingConfig::bits: u32
+
+pub vortex_onpair_sys::OnPairTrainingConfig::seed: u64
+
+pub vortex_onpair_sys::OnPairTrainingConfig::threshold: f64
+
+impl core::clone::Clone for vortex_onpair_sys::OnPairTrainingConfig
+
+pub fn vortex_onpair_sys::OnPairTrainingConfig::clone(&self) -> vortex_onpair_sys::OnPairTrainingConfig
+
+impl core::fmt::Debug for vortex_onpair_sys::OnPairTrainingConfig
+
+pub fn vortex_onpair_sys::OnPairTrainingConfig::fmt(&self, &mut core::fmt::Formatter<'_>) -> core::fmt::Result
+
+impl core::marker::Copy for vortex_onpair_sys::OnPairTrainingConfig
+
+pub struct vortex_onpair_sys::Parts<'a>
+
+pub vortex_onpair_sys::Parts::bits: u32
+
+pub vortex_onpair_sys::Parts::codes_boundaries: &'a [u32]
+
+pub vortex_onpair_sys::Parts::codes_packed: &'a [u64]
+
+pub vortex_onpair_sys::Parts::dict_bytes: &'a [u8]
+
+pub vortex_onpair_sys::Parts::dict_offsets: &'a [u32]
+
+pub vortex_onpair_sys::Parts::num_rows: usize
+
+impl<'a> core::clone::Clone for vortex_onpair_sys::Parts<'a>
+
+pub fn vortex_onpair_sys::Parts<'a>::clone(&self) -> vortex_onpair_sys::Parts<'a>
+
+impl<'a> core::marker::Copy for vortex_onpair_sys::Parts<'a>
+
+pub const vortex_onpair_sys::DEFAULT_DICT12_CONFIG: vortex_onpair_sys::OnPairTrainingConfig
+
+pub unsafe c fn vortex_onpair_sys::onpair_buffer_free(*mut u8, usize)
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_bits(*const vortex_onpair_sys::OnPairColumnHandle) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_compress(*const u8, *const u64, usize, vortex_onpair_sys::OnPairTrainingConfig, *mut *mut vortex_onpair_sys::OnPairColumnHandle) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_contains_into(*const vortex_onpair_sys::OnPairColumnHandle, *const u8, usize, *mut u8) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_decompress(*const vortex_onpair_sys::OnPairColumnHandle, usize, *mut u8, usize, *mut usize) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_decompress_capacity(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_deserialize(*const u8, usize, *mut *mut vortex_onpair_sys::OnPairColumnHandle) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_dict_bytes(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_dict_copy(*const vortex_onpair_sys::OnPairColumnHandle, *mut u8, usize, *mut u64) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_dict_size(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_equals_into(*const vortex_onpair_sys::OnPairColumnHandle, *const u8, usize, *mut u8) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_free(*mut vortex_onpair_sys::OnPairColumnHandle)
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_len(*const vortex_onpair_sys::OnPairColumnHandle) -> usize
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_parts(*const vortex_onpair_sys::OnPairColumnHandle, *mut vortex_onpair_sys::OnPairColumnParts) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_serialize(*const vortex_onpair_sys::OnPairColumnHandle, *mut *mut u8, *mut usize) -> u32
+
+pub unsafe c fn vortex_onpair_sys::onpair_column_starts_with_into(*const vortex_onpair_sys::OnPairColumnHandle, *const u8, usize, *mut u8) -> u32
+
+pub fn vortex_onpair_sys::read_bits_lsb(&[u64], usize, u32) -> u16
+
+pub fn vortex_onpair_sys::unpack_codes_to_u16(&[u64], usize, u32) -> alloc::vec::Vec<u16>