Rduckhts 1.1.4-0.0.1

• expand rduckhts_score() test coverage: add FORMAT/AS integer dosage fixture, missing DS value fixture, and seven GWAS summary preset fixtures (REGENIE, SAIGE, BOLT, METAL, PGS, SSF/GWAS-SSF); add R tinytest cases for TSV counts without threshold, GWAS-VCF with q_score_thr, and mutual-exclusion error assertions (include+exclude, regions+regions_file, targets+targets_file)

• Bundle duckhts 1.1.4 extension.

• add no_left_align parameter to rduckhts_liftover() (default FALSE): when TRUE, skips post-liftover left-alignment of indels, mirroring bcftools +liftover --no-left-align

• harden rduckhts_score() filter-expression failures for installed-package CI runs on Linux/Windows: invalid include/exclude expressions now surface as normal package errors from the bundled score engine instead of relying on a cross-frame bcftools shim jump path

• extend rduckhts_score() with bcftools-style filtering arguments (regions, regions_file, regions_overlap, targets, targets_file, targets_overlap, apply_filters, include, exclude) and package-level tests for region/target/FILTER behavior; include/exclude expressions are now evaluated against core VCF fields (POS, QUAL, CHROM, ID, REF, ALT, FILTER)

• improve rduckhts_liftover() multiallelic semantic parity with upstream bcftools +liftover: preserve all ALT alleles through reference-introduction/swaps, apply dynamic-allele indel normalization/left-alignment, treat ALT='.' as no alternate alleles, and add tinytest coverage for forward/reverse multiallelic liftover outputs

• add GWAS-VCF multi-PRS scoring: rduckhts_score() now supports GWAS-VCF summary files where each sample in the VCF defines a separate PRS with FORMAT/ES and FORMAT/LP fields

• add INFO/END liftover: rduckhts_liftover() gains an end_pos_col parameter to lift INFO/END positions alongside the primary coordinate, producing a dest_end output column

• add mitochondria passthrough: rduckhts_liftover() gains a lift_mt parameter (default FALSE) that renames MT/chrM contigs with matching source/destination sizes without chain lifting, matching upstream bcftools +liftover

• harden rduckhts_liftover() wrapper validation: negative max_snp_gap/max_indel_inc now fail in R before SQL execution to keep tinytest/check-r-package behavior stable across CI platforms

• add fai-only mode: rduckhts_munge() fasta_ref parameter is now optional (default NULL); when omitted, alleles pass through as-is without reference matching, matching upstream bcftools +munge --fai-only behavior

• port full indel liftover pipeline from upstream bcftools +liftover (Tier 3 — Algorithmic Completeness): allele extension, Needleman-Wunsch re-alignment with BWA-MEM affine gap scoring, 3-phase find_reference with interior matching and SNP rescue, left-alignment post-liftover via VT normalization, and SNP retry-as-indel dispatch — rduckhts_liftover() now handles indels near chain gaps, difficult SNPs at chain boundaries, and padded variants with correct coordinate adjustment

• breaking: some previously-rejected edge-of-chain SNPs now map successfully with note = 'Padded' due to the upstream-compatible retry-as-indel dispatch; R test expectations updated accordingly

• harden rduckhts_score() preset column parity: add 45 missing column entries across all 8 presets so non-scoring columns (A2, FRQ, SE, N, INFO, etc.) are recognized and silently consumed during header matching, matching upstream score.h definitions

• add gzipped summary file support to rduckhts_score(): .gz/.bgz compressed PGS Catalog and GWAS summary files can now be passed directly without decompression

• fix rduckhts_score() PRS name derivation to strip multiple known extensions ({gz,txt,tsv,vcf,bcf}) iteratively, matching upstream behavior

• add validation for explicitly requested use tag in rduckhts_score(): error if the VCF header lacks the specified FORMAT field instead of silently scoring zero

• fix rduckhts_score() CNT column naming to match upstream: when q_score_thr and counts are both active, count columns are now named <prs>_CNT_p<thr> (upstream pattern) instead of <prs>_p<thr>_CNT (breaking column name change for q_score_thr + counts queries)

• fix rduckhts_score() threshold boundary precision to match upstream: parse q_score_thr values with strtof→double promotion before -log10, reproducing the exact float→double comparison asymmetry in upstream bcftools +score; markers at exact P-value boundaries may now be excluded where they were previously included

• fix rduckhts_score() wrong-result bugs: remove incorrect METAL Zscore → P and SSF standard_error → P column mappings, add haploid GT support for chrX male samples, and fix memory leak on skipped markers

• fix rduckhts_liftover() wrong-result bugs: skip reverse complement for symbolic alleles and detect insertions as indels in the liftover path

• fix rduckhts_munge() wrong-result bugs: emit filter = 'MissingContig' instead of aborting when FASTA ref fetch fails for unknown contigs, and propagate NAN correctly on AC swap with missing NS

• harden rduckhts_score() upstream parity: port flexible chromosome name resolution (chr prefix strip/prepend, 23→X, 24→Y, 26/MT→chrM aliases), numerically stable -log10(p) parsing for very small p-values, and NA/. missing value handling in summary stats fields (BETA, OR, P, LP)

• add comprehensive rduckhts_score() test coverage: DS/HDS/AP/GP dosage modes with real FORMAT values, OR-to-beta conversion, PLINK2 preset with LOG10_P, custom columns_file mapping, allele mismatch (zero-match), missing genotype (./.) handling, NA in summary stats, chr-prefix flexible matching, auto-detection priority, and small p-value threshold precision

• strengthen rduckhts_score() parity coverage for variant matching by adding rsID-based fixture/tests (use_variant_id = TRUE), explicit CHR/BP mismatch behavior checks, and an error assertion when marker IDs are required but missing

• add METAL meta-analysis output support: rduckhts_munge() auto-detects METAL/heterogeneity keys (INFO, HET_I2, HET_P, HET_LP, DIRE) in the resolved column map and dispatches to duckdb_munge_metal(...), returning si, i2, cq, and ed columns; non-METAL presets continue to use the base 16-column schema without these fields

• align bundled munge outputs with the extension-level schema by renaming swapped to alleles_swapped with explicit allele-orientation semantics

• include munge APIs (duckdb_munge and bcftools_munge_row) in the generated bundled function catalog metadata and add a dedicated benchmark_munge.Rmd workflow (make bench-munge) for DuckHTS vs bcftools +munge benchmarking

• harden bundled liftover diagnostics: rduckhts_liftover() now surfaces explicit SQL errors for invalid chrom/pos rows through duckdb_liftover(...), direct bcftools_liftover(...) calls also error on invalid required inputs, and the package tests cover both wrapper-level and scalar invalid-row paths

• harden bundled liftover upstream parity: full IUPAC complement table, IUPAC→N reference sequence sanitization, 25→X/26→MT contig aliases, Ensembl chain ID dedup, chain coverage validation, and uint64_t regidx payload — all exact ports from upstream bcftools +liftover

• align bundled liftover outputs with bcftools +liftover semantics by replacing the old warning field with reject_reason for rejected rows and note for emitted rows that carry extra annotations

• add README liftover examples and broaden rduckhts_liftover() tinytest coverage for unmapped rows plus chain/FASTA and parameter validation failures

• expose the bundled liftover(...) table macro for score-style variant rows via a new rduckhts_liftover() helper that runs the macro against an input SQL query/table expression and returns lifted coordinates, alleles, reject reasons, and note annotations

• Keep the generated community extension metadata in sync with the bundled extension version by sourcing the emitted top-level version field from the repo-level description.yml.

• Bundle the duckhts 0.1.3.9001 extension update.

• Add quality_representation to rduckhts_bam() and rduckhts_fastq() so qualities can be returned as raw UTINYINT[] Phred values.

• Add input_quality_encoding to rduckhts_fastq(), defaulting to modern phred33 while allowing explicit legacy FASTQ decoding; expose rduckhts_detect_quality_encoding() for heuristic FASTQ encoding inspection.

• Add sequence_encoding parameter to rduckhts_bam(), rduckhts_fasta(), and rduckhts_fastq() wrappers, forwarding sequence_encoding := 'nt16' to the extension readers for raw nt16-encoded sequence output as UTINYINT[].

• Breaking: seq_encode_4bit() no longer returns NULL for unknown characters — they now map to N (code 15), matching htslib seq_nt16_table behavior; UDF and reader nt16 paths use shared code.

• Remove the unsupported attributes_map argument from rduckhts_tabix() so the wrapper matches the generic read_tabix(...) surface; attribute maps remain available via rduckhts_gff() and rduckhts_gtf().

• Expose the bundled read_bcf(...) CSQ/ANN/BCSQ typing cleanup, including centralized builtin rules and the additional_csq_column_types := ... override parameter.

• Add read_bed(...) and fasta_nuc(...) to the bundled extension surface, plus rduckhts_bed() and rduckhts_fasta_nuc() wrappers.

• Add rduckhts_bgzip(), rduckhts_bgunzip(), rduckhts_bam_index(), rduckhts_bcf_index(), and rduckhts_tabix_index() wrappers for the new extension compression and indexing functions.

• Expose the newer bundled extension surface in the package catalog, including HTS metadata readers, additional sequence helpers, sam_flag_bits()/sam_flag_has(), the new CIGAR utils helpers, and the expanded SAM/tag and tabix reader capabilities.

• Rename the ambiguous SAM flag helper names in the bundled extension/catalog to the clearer forms is_paired(), is_proper_pair(), is_next_segment_unmapped(), and is_next_segment_reverse_complemented().

• Add is_forward_aligned() for mapped-strand checks in pure SQL workflows such as strand-split bin counting.

• Bootstrap the new extension sources into the package build and update configure/configure.win so the bundled extension compiles them on Unix and Windows.

• Regenerate the package-bundled function catalog and roxygen documentation for the new wrappers.

• Add installed-package tinytest coverage for BED reading, FASTA nucleotide composition, BGZF round-trips, tabix indexing, and BAM/BCF index creation.

Rduckhts 0.1.3-0.0.2.9000

Rduckhts 0.1.3-0.0.2

• Conditionaly enable plugins in windows

• Updates the configure script to avoid check faillure on CRAN MacOS

• Update the extension version to 0.1.3

Rduckhts 0.1.2-0.1.5

• Fixed inadvertant removal of libexec
• Updated the plugin to add header table functions

Rduckhts 0.1.2-0.1.4

• CRAN Submission

Rduckhts 0.1.2-0.0.9000

• Different fixes for CRAN submission
  • Updated DESCRIPTION Title/Description formatting and added HTSlib reference.
  • Removed default write paths in bootstrap/build helpers; now require explicit paths.
  • setup_hts_env now accepts an explicit plugins_dir parameter.
  • duckhts_build now accepts a make argument (GNU make required).
• modified configure to attemp to support wasm
• Update bootstrapped extension code to match duckhts 0.1.2.
• Add SAMtags + auxiliary tag support (standard_tags, auxiliary_tags).
• Add tabix header/typing options (header, header_names, auto_detect, column_types).

Rduckhts 0.1.1-0.0.3

• make the build single threaded

Rduckhts 0.1.1-0.0.3

• misspeling correction

Rduckhts 0.1.1-0.0.2

• CRAN resubmission: apply DuckDB C API header patch to avoid strict-prototypes warnings.

Rduckhts 0.1.1-0.0.1

• CRAN Submission

• Bump bundled duckhts extension version to 0.1.1.

• Initial development release.

• Bundles the DuckHTS DuckDB extension and htslib for HTS file readers.

• Adds table-creation helpers for VCF/BCF, BAM/CRAM, FASTA/FASTQ, GFF/GTF, and tabix.