Dev 0.9.4

.github/workflows/ci.yml

@@ -1,6 +1,9 @@
 name: CI
 
-on: [push, pull_request]
+on:
+  push:
+    branches: [main]
+  pull_request:
 
 env:
   CARGO_TERM_COLOR: always
@@ -9,7 +12,7 @@ jobs:
   build:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v7
       - name: Build
         run: cargo build --verbose
       - name: Run tests
@@ -20,19 +23,19 @@ jobs:
   fmt_lint:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v7
       - name: Formatting
         run: cargo fmt --check
       - name: Linting
-        run: cargo clippy --verbose
+        run: cargo clippy --all-features --all-targets -- -D warnings
 
   example_grep:
     runs-on: ubuntu-latest
     strategy:
       matrix:
         ext: [bq, vbq, cbq]
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v7
       - name: run example ${{ matrix.ext }}
         run: cargo run --release --example grep -- ./data/subset.${{ matrix.ext }} "ACGTACGT"
 
@@ -42,7 +45,7 @@ jobs:
       matrix:
         ext: [bq, vbq, cbq]
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v7
       - name: run example ${{ matrix.ext }}
         run: cargo run --release --example parallel_range -- ./data/subset.${{ matrix.ext }} 4 30 200
 
@@ -52,7 +55,7 @@ jobs:
       matrix:
         ext: [bq, vbq, cbq]
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v7
       - name: run example (single) ${{ matrix.ext }}
         run: cargo run --release --example write -- ./data/subset_R1.fastq.gz -o ./output.${{ matrix.ext }}
       - name: run example (paired) ${{ matrix.ext }}
@@ -64,6 +67,6 @@ jobs:
       matrix:
         ext: [bq, vbq, cbq]
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v7
       - name: run example ${{ matrix.ext }}
         run: cargo run --release --example read -- ./data/subset.${{ matrix.ext }}

CHANGELOG.md

@@ -0,0 +1,167 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [0.9.4] - 2026-07-15
+
+### Fixed
+
+- Bug in the `bq` stream reader that used an incorrect record id for small buffers.
+
+### Changed
+
+- Added clippy checks and additional lint allowances to CI, plus general style fixes.
+- Improved test coverage throughout the library.
+
+## [0.9.3] - 2026-07-01
+
+### Changed
+
+- `Binseq` input is now accepted as a path rather than a pre-opened handle.
+
+### Fixed
+
+- Use `reserve` instead of `resize` to avoid unnecessary zeroing/copies (issue #94).
+- Ensure `len_nef` is cleared correctly.
+
+### Tests
+
+- Added coverage for the sequential reader and N-containing sequences.
+
+## [0.9.1] - 2026-06-25
+
+### Added
+
+- Support for writing only completed blocks, enabling on-thread compression and
+  reducing lock contention for parallel writers.
+- `on_thread_complete` callback support.
+- `ingest_completed` is now passed through, with behavior changes scoped to `cbq`.
+
+### Tests
+
+- Added coverage for `cbq` ingestion.
+
+## [0.9.0] - 2026-01-23
+
+### Added
+
+- Integrated `cbq` (compressed BINSEQ) into the core library as a first-class format.
+- Builder-based construction of a file from an existing header.
+- Auto-encode a FASTX file (or a pair of FASTX files) directly into a BINSEQ format via
+  a builder method.
+- A generic writer abstraction over BINSEQ files.
+- Default quality score handling across all readers.
+- Elias-Fano encoding for N-positions instead of raw `u64` indices, reducing index size.
+- An iterator over block headers, and convenience accessors for index metadata.
+
+### Changed
+
+- **Breaking:** renamed `BINSEQ`/`VBINSEQ` naming throughout to `BQ`/`VBQ`
+  (e.g. `BinseqHeader` → `bq::FileHeader`), and renamed several `vbq`/`bq`-specific
+  headers and writer structs for clarity.
+- Reworked the write API around a sequence record struct.
+- Reworked the internal index to track record count and sequence count separately.
+- Replaced `anyhow`-based errors throughout the crate with crate-native error types.
+- Removed the external `vbq` index file in favor of writing the index into the file itself.
+
+### Fixed
+
+- Several small hotfixes to the write path.
+- Incorrect binary file used in a test fixture.
+
+## [0.8.1] - 2025-12-10
+
+### Fixed
+
+- Incorrect decoding of the extended buffer on batch decoding in `bq` when a full
+  chunk of primary data was present (issue #75).
+
+## [0.8.0] - 2025-12-10
+
+### Added
+
+- `sseq`/`xseq` functions on the record trait for use during batch decoding, implemented
+  for both `bq` and `vbq`.
+- Additional usage examples.
+
+### Changed
+
+- **Breaking:** improved `bq` decoding via batch decoding, keeping data in SIMD form longer.
+- Improved header access logic for zero-copy reads.
+
+### Fixed
+
+- Correct index returned from decoding.
+- Correct slicing when per-record flags are set in `bq`.
+
+## [0.7.8] - 2025-12-09
+
+### Changed
+
+- `vbq` reader is now zero-copy or single-copy depending on compression status.
+- Simplified the encoding API.
+- Reusable decompression context for improved throughput.
+- Updated all dependencies.
+
+### Fixed
+
+- Validate the size of bytes provided to the non-compressed reader.
+
+## [0.7.7] - 2025-12-08
+
+### Changed
+
+- Split the sequencing context into multiple traits with distinct variants
+  (renamed `FullCtx` to `Ctx`).
+- Removed unsafe code from the context implementation in favor of a resize-copy pattern.
+
+### Fixed
+
+- Logical error in quality-score padding (`ensure_quality`).
+
+### Docs
+
+- Documented the different context variants.
+
+## [0.7.6] - 2025-12-08
+
+### Added
+
+- A basic sequencing context struct, with mutable references to its internals.
+- Ability to fill in missing quality scores when required.
+- A generic error-conversion trait, made public.
+
+## [0.7.5] - 2025-10-02
+
+### Added
+
+- BINSEQ v2 format.
+- Optional headers in `vbq` files, with the index moved into the file itself
+  (no more auxiliary index file).
+- Support for 4-bit encodings, with a `bitsize` re-export so downstream users don't
+  need to depend on `bitnuc` directly.
+- Header builder pattern for constructing `bq` and `vbq` files.
+
+### Changed
+
+- Per-record flags are now optional across `bq`, `vbq`, and the record API.
+- `vbq` index now uses `u64` for cumulative record counts.
+
+## [0.6.5] - 2025-07-01
+
+Baseline release for this changelog. Earlier history predates version tagging.
+
+[Unreleased]: https://github.com/arcinstitute/binseq/compare/v0.9.3...HEAD
+[0.9.3]: https://github.com/arcinstitute/binseq/compare/v0.9.1...v0.9.3
+[0.9.1]: https://github.com/arcinstitute/binseq/compare/v0.9.0...v0.9.1
+[0.9.0]: https://github.com/arcinstitute/binseq/compare/v0.8.1...v0.9.0
+[0.8.1]: https://github.com/arcinstitute/binseq/compare/v0.8.0...v0.8.1
+[0.8.0]: https://github.com/arcinstitute/binseq/compare/v0.7.8...v0.8.0
+[0.7.8]: https://github.com/arcinstitute/binseq/compare/v0.7.7...v0.7.8
+[0.7.7]: https://github.com/arcinstitute/binseq/compare/v0.7.6...v0.7.7
+[0.7.6]: https://github.com/arcinstitute/binseq/compare/v0.7.5...v0.7.6
+[0.7.5]: https://github.com/arcinstitute/binseq/compare/v0.6.5...v0.7.5
+[0.6.5]: https://github.com/arcinstitute/binseq/releases/tag/v0.6.5

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "binseq"
-version = "0.9.3"
+version = "0.9.4"
 edition = "2024"
 description = "A high efficiency binary format for sequencing data"
 license = "MIT"
@@ -11,27 +11,27 @@ categories = ["science::bioinformatics", "encoding", "data-structures"]
 keywords = ["binary", "nucleotide", "sequencing", "genomics", "fastq"]
 
 [dependencies]
-anyhow = {version = "1.0.100", optional = true}
+anyhow = {version = "1.0.103", optional = true}
 auto_impl = "1.3.0"
-bitnuc = "0.4.0"
-bytemuck = { version = "1.24.0", features = ["derive", "extern_crate_alloc"] }
+bitnuc = "0.4.1"
+bytemuck = { version = "1.25.1", features = ["derive", "extern_crate_alloc"] }
 byteorder = "1.5.0"
-itoa = "1.0.17"
-memchr = "2.7.6"
-memmap2 = "0.9.9"
+itoa = "1.0.18"
+memchr = "2.8.3"
+memmap2 = "0.9.11"
 num_cpus = "1.17.0"
-paraseq = { version = "0.4.8", optional = true }
+paraseq = { version = "0.4.14", optional = true }
 parking_lot = {version = "0.12.5", optional = true }
-rand = { version = "0.9.2", features = ["small_rng"] }
+rand = { version = "0.9.5", features = ["small_rng"] }
 sucds = "0.8.3"
-thiserror = "2.0.17"
+thiserror = "2.0.18"
 zstd = { version = "0.13.3", features = ["zstdmt"] }
 
 [dev-dependencies]
-anyhow = "1.0.100"
+anyhow = "1.0.103"
 parking_lot = "0.12.5"
-clap = { version = "4.5.54", features = ["derive"] }
-paraseq = "0.4.8"
+clap = { version = "4.6.2", features = ["derive"] }
+paraseq = "0.4.14"
 
 [features]
 default = ["paraseq", "anyhow"]
@@ -41,4 +41,7 @@ paraseq = ["dep:paraseq", "dep:parking_lot"]
 [lints.clippy]
 pedantic = { level = "warn", priority = -1 }
 cast_possible_truncation = "allow"
+cast_precision_loss = "allow"
 missing_errors_doc = "allow"
+struct_excessive_bools = "allow"
+fn_params_excessive_bools = "allow"

examples/auto-write.rs

@@ -64,17 +64,15 @@ impl Args {
     fn format(&self) -> Format {
         if let Some(format) = self.format {
             format
-        } else {
-            if let Some(output) = &self.output {
-                match output.split(".").last() {
-                    Some("bq") => Format::Bq,
-                    Some("vbq") => Format::Vbq,
-                    Some("cbq") => Format::Cbq,
-                    _ => Format::default(),
-                }
-            } else {
-                Format::default()
+        } else if let Some(output) = &self.output {
+            match output.split('.').next_back() {
+                Some("bq") => Format::Bq,
+                Some("vbq") => Format::Vbq,
+                Some("cbq") => Format::Cbq,
+                _ => Format::default(),
             }
+        } else {
+            Format::default()
         }
     }
     fn bitsize(&self) -> BitSize {
@@ -87,7 +85,7 @@ impl Args {
     /// Creates an output file handle
     fn ohandle(&self) -> Result<BoxedWriter> {
         let path = if let Some(output) = &self.output {
-            output.to_string()
+            output.clone()
         } else {
             format!("{}{}", &self.prefix, self.format().extension())
         };

examples/grep.rs

@@ -37,7 +37,7 @@ impl GrepCounter {
 }
 impl ParallelProcessor for GrepCounter {
     fn process_record<R: binseq::BinseqRecord>(&mut self, record: R) -> binseq::Result<()> {
-        if self.match_sequence(&record.sseq()) || self.match_sequence(&record.xseq()) {
+        if self.match_sequence(record.sseq()) || self.match_sequence(record.xseq()) {
             self.local_count += 1;
         }
 

examples/parallel_range.rs

@@ -30,18 +30,18 @@ impl ParallelProcessor for RangeProcessor {
         let count = self.counter.fetch_add(1, Ordering::Relaxed);
 
         // Print progress every 10,000 records
-        if count % 10_000 == 0 {
-            if let Some(tid) = self.tid {
-                println!(
-                    "Thread {}: Processed {} records (Range: {}-{}, Index: {}, Len: {})",
-                    tid,
-                    count + 1,
-                    self.range_start,
-                    self.range_end,
-                    record.index(),
-                    record.sseq().len(),
-                );
-            }
+        if count.is_multiple_of(10_000)
+            && let Some(tid) = self.tid
+        {
+            println!(
+                "Thread {}: Processed {} records (Range: {}-{}, Index: {}, Len: {})",
+                tid,
+                count + 1,
+                self.range_start,
+                self.range_end,
+                record.index(),
+                record.sseq().len(),
+            );
         }
 
         Ok(())