minimap2 0.2.31.0 → 1.2.31.0

data/lib/minimap2.rb

@@ -1,137 +1,66 @@
 # frozen_string_literal: true
 
-# dependencies
-require "ffi"
-
-# modules
-require_relative "minimap2/aligner"
-require_relative "minimap2/alignment"
-require_relative "minimap2/version"
-
-# Minimap2 mapper for long read sequences
-# https://github.com/lh3/minimap2
-# Li, H. (2018). Minimap2: pairwise alignment for nucleotide sequences. Bioinformatics, 34:3094-3100.
-# doi:10.1093/bioinformatics/bty191
 module Minimap2
   class Error < StandardError; end
+end
 
-  class << self
-    attr_accessor :ffi_lib
-  end
+begin
+  require "minimap2/minimap2_ext"
+rescue LoadError => e
+  # Allow running directly from a source checkout after `rake compile`.
+  extension = File.expand_path("../ext/ruby_minimap2/minimap2_ext", __dir__)
+  raise e unless Dir["#{extension}.{so,bundle,dylib,dll}"].any?
 
-  lib_name = ::FFI.map_library_name("minimap2")
-  self.ffi_lib = if ENV["MINIMAPDIR"]
-                   File.expand_path(lib_name, ENV["MINIMAPDIR"])
-                 else
-                   File.expand_path("../vendor/#{lib_name}", __dir__)
-                 end
+  require extension
+end
 
-  # friendlier error message
-  autoload :FFI, "minimap2/ffi"
+require_relative "minimap2/alignment"
+require_relative "minimap2/aligner"
+require_relative "minimap2/version"
 
-  # methods from mappy
+# Ruby bindings for the minimap2 sequence aligner.
+module Minimap2
   class << self
-    # Execute minimap2 comannd with given options.
-    # @overload  execute(arg0,arg1,...)
-    # @param [String] arg minimap2 command option.
-    # @example Get minimap2 version
-    #   Minimap2.execute('--version')
-
-    def execute(*rb_argv)
-      str_ptrs = []
-      # First argument is the program name.
-      str_ptrs << ::FFI::MemoryPointer.from_string("minimap2")
-      rb_argv.each do |arg|
-        arg.to_s.split(/\s+/).each do |s|
-          str_ptrs << ::FFI::MemoryPointer.from_string(s)
-        end
-      end
-      str_ptrs << nil
-
-      # Load all the pointers into a native memory block
-      argv = ::FFI::MemoryPointer.new(:pointer, str_ptrs.length)
-      str_ptrs.each_with_index do |p, i|
-        argv[i].put_pointer(0,  p)
-      end
-
-      FFI.main(str_ptrs.length - 1, argv)
+    # Run minimap2's command entry point. Each Ruby argument is one argv item.
+    def execute(*arguments)
+      Native.execute(*arguments)
     end
 
-    # Get verbosity level.
-    # @return [Integer] verbosity level.
-
     def verbose
-      FFI.mm_verbose_level(-1)
+      Native.verbose
     end
 
-    # Set verbosity level.
-    # @param [Integer] verbosity level
-    # @return [Integer] verbosity level.
-
     def verbose=(level)
-      FFI.mm_verbose_level(level)
+      Native.verbose = level
     end
 
-    # Read fasta/fastq file.
-    # @param [String] file_path
-    # @param [Boolean] comment If True, the comment will be read.
-    # @yield [name, seq, qual, comment]
-    # @return [Enumerator] enum Return Enumerator if not block given.
-    # Note: You can BioRuby instead of this method.
-
-    def fastx_read(file_path, comment: false, &block)
-      path = File.expand_path(file_path)
-
-      # raise error in Ruby because ks.null? is false even if file not exist.
-      raise ArgumentError, "File not found: #{path}" unless File.exist?(path)
-
-      ks = FFI.mm_fastx_open(path)
+    # Read records from a FASTA or FASTQ file.
+    def fastx_read(file_path, comment: false)
+      path = file_path.to_s == "-" ? "-" : File.expand_path(file_path)
+      raise Error, "Cannot open FASTA/FASTQ file: #{path}" unless path == "-" || File.file?(path)
 
+      reader = Native::FastxReader.new(path)
       if block_given?
-        fastx_each(ks, comment, &block)
+        begin
+          while (record = reader.next_record(comment))
+            yield record
+          end
+        ensure
+          reader.close
+        end
       else
-        Enumerator.new do |y|
-          # rewind not work
-          fastx_each(ks, comment) { |r| y << r }
+        Enumerator.new do |yielder|
+          while (record = reader.next_record(comment))
+            yielder << record
+          end
+        ensure
+          reader.close
         end
       end
     end
 
-    # Reverse complement sequence.
-    # @param [String] seq
-    # @return [string] seq
-
-    def revcomp(seq)
-      l = seq.size
-      bseq = ::FFI::MemoryPointer.new(:char, l)
-      bseq.put_bytes(0, seq)
-      p = FFI.mappy_revcomp(l, bseq)
-      return "" if p.nil? || p.null?
-
-      begin
-        p.read_string(l)
-      ensure
-        FFI.mappy_free(p) unless p.nil? || p.null?
-      end
-    end
-
-    private
-
-    def fastx_each(ks, comment)
-      yield fastx_next(ks, comment) while FFI.kseq_read(ks) >= 0
-      FFI.mm_fastx_close(ks)
-    end
-
-    def fastx_next(ks, read_comment)
-      qual = ks[:qual][:s] if ks[:qual][:l] > 0
-      name = ks[:name][:s]
-      seq  = ks[:seq][:s]
-      if read_comment
-        comment = ks[:comment][:s] if ks[:comment][:l] > 0
-        [name, seq, qual, comment]
-      else
-        [name, seq, qual]
-      end
+    def revcomp(sequence)
+      Native.revcomp(sequence)
     end
   end
 end

metadata

@@ -1,68 +1,29 @@
 --- !ruby/object:Gem::Specification
 name: minimap2
 version: !ruby/object:Gem::Version
-  version: 0.2.31.0
+  version: 1.2.31.0
 platform: ruby
 authors:
 - kojix2
 bindir: bin
 cert_chain: []
 date: 1980-01-02 00:00:00.000000000 Z
-dependencies:
-- !ruby/object:Gem::Dependency
-  name: ffi
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-- !ruby/object:Gem::Dependency
-  name: ffi-bitfield
-  requirement: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
-  type: :runtime
-  prerelease: false
-  version_requirements: !ruby/object:Gem::Requirement
-    requirements:
-    - - ">="
-      - !ruby/object:Gem::Version
-        version: '0'
+dependencies: []
 description: Ruby bindings to the Minimap2 aligner.
 email:
 - 2xijok@gmail.com
 executables: []
 extensions:
-- ext/Rakefile
+- ext/ruby_minimap2/extconf.rb
 extra_rdoc_files: []
 files:
 - LICENSE.txt
 - README.md
-- ext/Rakefile
-- ext/cmappy/cmappy.c
-- ext/cmappy/cmappy.h
-- ext/minimap2.patch
-- ext/minimap2/FAQ.md
 - ext/minimap2/LICENSE.txt
-- ext/minimap2/MANIFEST.in
 - ext/minimap2/Makefile
-- ext/minimap2/Makefile.simde
-- ext/minimap2/NEWS.md
-- ext/minimap2/README.md
 - ext/minimap2/align.c
 - ext/minimap2/bseq.c
 - ext/minimap2/bseq.h
-- ext/minimap2/code_of_conduct.md
-- ext/minimap2/cookbook.md
 - ext/minimap2/esterr.c
 - ext/minimap2/example.c
 - ext/minimap2/format.c
@@ -90,64 +51,22 @@ files:
 - ext/minimap2/main.c
 - ext/minimap2/map.c
 - ext/minimap2/minimap.h
-- ext/minimap2/minimap2.1
 - ext/minimap2/misc.c
-- ext/minimap2/misc/README.md
-- ext/minimap2/misc/pafcluster.js
-- ext/minimap2/misc/paftools.js
 - ext/minimap2/mmpriv.h
 - ext/minimap2/options.c
 - ext/minimap2/pe.c
-- ext/minimap2/pyproject.toml
-- ext/minimap2/python/README.rst
-- ext/minimap2/python/cmappy.h
-- ext/minimap2/python/cmappy.pxd
-- ext/minimap2/python/mappy.pyx
-- ext/minimap2/python/minimap2.py
 - ext/minimap2/sdust.c
 - ext/minimap2/sdust.h
 - ext/minimap2/seed.c
-- ext/minimap2/setup.py
 - ext/minimap2/sketch.c
 - ext/minimap2/splitidx.c
 - ext/minimap2/sse2neon/emmintrin.h
-- ext/minimap2/test/MT-human.fa
-- ext/minimap2/test/MT-orang.fa
-- ext/minimap2/test/q-inv.fa
-- ext/minimap2/test/q2.fa
-- ext/minimap2/test/t-inv.fa
-- ext/minimap2/test/t2.fa
-- ext/minimap2/test/x3s-aln.txt
-- ext/minimap2/test/x3s-qry.fa
-- ext/minimap2/test/x3s-ref.fa
-- ext/minimap2/tex/Makefile
-- ext/minimap2/tex/bioinfo.cls
-- ext/minimap2/tex/blasr-mc.eval
-- ext/minimap2/tex/bowtie2-s3.sam.eval
-- ext/minimap2/tex/bwa-s3.sam.eval
-- ext/minimap2/tex/bwa.eval
-- ext/minimap2/tex/eval2roc.pl
-- ext/minimap2/tex/graphmap.eval
-- ext/minimap2/tex/hs38-simu.sh
-- ext/minimap2/tex/minialign.eval
-- ext/minimap2/tex/minimap2.bib
-- ext/minimap2/tex/minimap2.tex
-- ext/minimap2/tex/mm2-s3.sam.eval
-- ext/minimap2/tex/mm2-update.tex
-- ext/minimap2/tex/mm2.approx.eval
-- ext/minimap2/tex/mm2.eval
-- ext/minimap2/tex/natbib.bst
-- ext/minimap2/tex/natbib.sty
-- ext/minimap2/tex/ngmlr.eval
-- ext/minimap2/tex/roc.gp
-- ext/minimap2/tex/snap-s3.sam.eval
+- ext/ruby_minimap2/extconf.rb
+- ext/ruby_minimap2/native_main.c
+- ext/ruby_minimap2/ruby_minimap2.c
 - lib/minimap2.rb
 - lib/minimap2/aligner.rb
 - lib/minimap2/alignment.rb
-- lib/minimap2/ffi.rb
-- lib/minimap2/ffi/constants.rb
-- lib/minimap2/ffi/functions.rb
-- lib/minimap2/ffi/mappy.rb
 - lib/minimap2/version.rb
 homepage: https://github.com/kojix2/ruby-minimap2
 licenses:
@@ -160,14 +79,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: '2.5'
+      version: '3.3'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 4.0.10
+rubygems_version: 4.0.16
 specification_version: 4
 summary: minimap2
 test_files: []

data/ext/Rakefile

@@ -1,63 +0,0 @@
-# frozen_string_literal: true
-
-require "rake"
-require "fileutils"
-require "ffi"
-
-minimap2_dir = File.expand_path("minimap2", __dir__)
-target_dir = "../../vendor"
-target_fname = FFI.map_library_name("minimap2")
-target_path = File.join(target_dir, target_fname)
-
-task default: ["minimap2:build", "minimap2:clean"]
-
-namespace :minimap2 do
-  desc "Compile Minimap2"
-  task :build do
-    Dir.chdir(minimap2_dir) do
-      # Add -fPIC option to Makefile
-      unless system("git", "--version", out: File::NULL, err: File::NULL)
-        abort "Git is required to build ruby-minimap2 from source. Install Git and ensure `git` is on PATH."
-      end
-      sh "git apply ../minimap2.patch"
-      sh "cp ../cmappy/cmappy.h ../cmappy/cmappy.c ."
-      case RbConfig::CONFIG["host_cpu"]
-      when /arm64/
-        sh "make arm_neon=1 aarch64=1"
-      when /arm/
-        sh "make arm_neon=1"
-      else
-        sh "make"
-      end
-      case RbConfig::CONFIG["host_os"]
-      when /mswin|msys|mingw|cygwin|bccwin|wince|emc/
-        sh "cc *.o -shared -o #{target_fname} -lm -lz -lpthread"
-      when /darwin|mac os/
-        sh "clang -dynamiclib -undefined dynamic_lookup -o #{target_fname} *.o -lm -lz -lpthread"
-        sh "otool -L #{target_fname}"
-      else
-        sh "cc *.o -shared -o #{target_fname} -lm -lz -lpthread"
-        sh "ldd -r #{target_fname}"
-      end
-      sh "rm cmappy.h cmappy.c"
-      sh "git apply -R ../minimap2.patch"
-      FileUtils.mkdir_p(target_dir)
-      warn "mkdir -p #{target_dir}"
-      sh "mv #{target_fname} #{target_path}"
-    end
-  end
-
-  desc "`make clean`"
-  task :clean do
-    Dir.chdir(minimap2_dir) do
-      sh "make clean"
-    end
-  end
-
-  desc "`make clean` and remove shared lib"
-  task cleanall: [:clean] do
-    Dir.chdir(minimap2_dir) do
-      sh "rm #{target_path}" if File.exist?(target_path)
-    end
-  end
-end

data/ext/cmappy/cmappy.c

@@ -1,134 +0,0 @@
-#include "cmappy.h"
-
-void mm_reg2hitpy(const mm_idx_t *mi, mm_reg1_t *r, mm_hitpy_t *h)
-{
-	h->ctg = mi->seq[r->rid].name;
-	h->ctg_len = mi->seq[r->rid].len;
-	h->ctg_start = r->rs, h->ctg_end = r->re;
-	h->qry_start = r->qs, h->qry_end = r->qe;
-	h->strand = r->rev? -1 : 1;
-	h->mapq = r->mapq;
-	h->mlen = r->mlen;
-	h->blen = r->blen;
-	h->NM = r->blen - r->mlen + r->p->n_ambi;
-	h->trans_strand = r->p->trans_strand == 1? 1 : r->p->trans_strand == 2? -1 : 0;
-	h->is_primary = (r->id == r->parent);
-	h->seg_id = r->seg_id;
-	h->n_cigar32 = r->p->n_cigar;
-	h->cigar32 = r->p->cigar;
-}
-
-void mm_free_reg1(mm_reg1_t *r)
-{
-	free(r->p);
-}
-
-kseq_t *mm_fastx_open(const char *fn)
-{
-	gzFile fp;
-	fp = fn && strcmp(fn, "-") != 0? gzopen(fn, "r") : gzdopen(fileno(stdin), "r");
-	return kseq_init(fp);
-}
-
-void mm_fastx_close(kseq_t *ks)
-{
-	gzFile fp;
-	fp = ks->f->f;
-	kseq_destroy(ks);
-	gzclose(fp);
-}
-
-int mm_verbose_level(int v)
-{
-	if (v >= 0) mm_verbose = v;
-	return mm_verbose;
-}
-
-void mm_reset_timer(void)
-{
-	extern double realtime(void);
-	mm_realtime0 = realtime();
-}
-
-mm_reg1_t *mm_map_aux(const mm_idx_t *mi, const char* seqname, const char *seq1, const char *seq2, int *n_regs, mm_tbuf_t *b, const mm_mapopt_t *opt)
-{
-	mm_reg1_t *r;
-
-	// Py_BEGIN_ALLOW_THREADS
-	if (seq2 == 0) {
-		r = mm_map(mi, strlen(seq1), seq1, n_regs, b, opt, seqname);
-	} else {
-		int _n_regs[2];
-		mm_reg1_t *regs[2];
-		char *seq[2];
-		int i, len[2];
-
-		len[0] = strlen(seq1);
-		len[1] = strlen(seq2);
-		seq[0] = (char*)seq1;
-		seq[1] = strdup(seq2);
-		for (i = 0; i < len[1]>>1; ++i) {
-			int t = seq[1][len[1] - i - 1];
-			seq[1][len[1] - i - 1] = seq_comp_table[(uint8_t)seq[1][i]];
-			seq[1][i] = seq_comp_table[t];
-		}
-		if (len[1]&1) seq[1][len[1]>>1] = seq_comp_table[(uint8_t)seq[1][len[1]>>1]];
-		mm_map_frag(mi, 2, len, (const char**)seq, _n_regs, regs, b, opt, seqname);
-		for (i = 0; i < _n_regs[1]; ++i)
-			regs[1][i].rev = !regs[1][i].rev;
-		*n_regs = _n_regs[0] + _n_regs[1];
-		regs[0] = (mm_reg1_t*)realloc(regs[0], sizeof(mm_reg1_t) * (*n_regs));
-		memcpy(&regs[0][_n_regs[0]], regs[1], _n_regs[1] * sizeof(mm_reg1_t));
-		free(regs[1]);
-		r = regs[0];
-	}
-	// Py_END_ALLOW_THREADS
-
-	return r;
-}
-
-char *mappy_revcomp(int len, const uint8_t *seq)
-{
-	int i;
-	char *rev;
-	rev = (char*)malloc(len + 1);
-	for (i = 0; i < len; ++i)
-		rev[len - i - 1] = seq_comp_table[seq[i]];
-	rev[len] = 0;
-	return rev;
-}
-
-char *mappy_fetch_seq(const mm_idx_t *mi, const char *name, int st, int en, int *len)
-{
-	int i, rid;
-	char *s;
-	*len = 0;
-	rid = mm_idx_name2id(mi, name);
-	if (rid < 0) return 0;
-	if ((uint32_t)st >= mi->seq[rid].len || st >= en) return 0;
-	if (en < 0 || (uint32_t)en > mi->seq[rid].len)
-		en = mi->seq[rid].len;
-	s = (char*)malloc(en - st + 1);
-	*len = mm_idx_getseq(mi, rid, st, en, (uint8_t*)s);
-	for (i = 0; i < *len; ++i)
-		s[i] = "ACGTN"[(uint8_t)s[i]];
-	s[*len] = 0;
-	return s;
-}
-
-mm_idx_t *mappy_idx_seq(int w, int k, int is_hpc, int bucket_bits, const char *seq, int len)
-{
-	const char *fake_name = "N/A";
-	char *s;
-	mm_idx_t *mi;
-	s = (char*)calloc(len + 1, 1);
-	memcpy(s, seq, len);
-	mi = mm_idx_str(w, k, is_hpc, bucket_bits, 1, (const char**)&s, (const char**)&fake_name);
-	free(s);
-	return mi;
-}
-
-void mappy_free(void *p)
-{
-	free(p);
-}

data/ext/cmappy/cmappy.h

@@ -1,46 +0,0 @@
-#ifndef CMAPPY_H
-#define CMAPPY_H
-
-#include <stdlib.h>
-#include <string.h>
-#include <zlib.h>
-#include "minimap.h"
-#include "kseq.h"
-KSEQ_DECLARE(gzFile)
-
-typedef struct {
-	const char *ctg;
-	int32_t ctg_start, ctg_end;
-	int32_t qry_start, qry_end;
-	int32_t blen, mlen, NM, ctg_len;
-	uint8_t mapq, is_primary;
-	int8_t strand, trans_strand;
-	int32_t seg_id;
-	int32_t n_cigar32;
-	uint32_t *cigar32;
-} mm_hitpy_t;
-
-void mm_reg2hitpy(const mm_idx_t *mi, mm_reg1_t *r, mm_hitpy_t *h);
-
-void mm_free_reg1(mm_reg1_t *r);
-
-kseq_t *mm_fastx_open(const char *fn);
-
-void mm_fastx_close(kseq_t *ks);
-
-int mm_verbose_level(int v);
-
-void mm_reset_timer(void);
-
-extern unsigned char seq_comp_table[256];
-mm_reg1_t *mm_map_aux(const mm_idx_t *mi, const char* seqname, const char *seq1, const char *seq2, int *n_regs, mm_tbuf_t *b, const mm_mapopt_t *opt);
-
-char *mappy_revcomp(int len, const uint8_t *seq);
-
-char *mappy_fetch_seq(const mm_idx_t *mi, const char *name, int st, int en, int *len);
-
-mm_idx_t *mappy_idx_seq(int w, int k, int is_hpc, int bucket_bits, const char *seq, int len);
-
-void mappy_free(void *p);
-
-#endif

data/ext/minimap2/FAQ.md

@@ -1,46 +0,0 @@
-#### 1. Alignment different with option `-a` or `-c`?
-
-Without `-a`, `-c` or `--cs`, minimap2 only finds *approximate* mapping
-locations without detailed base alignment. In particular, the start and end
-positions of the alignment are imprecise. With one of those options, minimap2
-will perform base alignment, which is generally more accurate but is much
-slower.
-
-#### 2. How to map Illumina short reads to noisy long reads?
-
-No good solutions. The better approach is to assemble short reads into contigs
-and then map noisy reads to contigs.
-
-#### 3. The output SAM doesn't have a header.
-
-By default, minimap2 indexes 4 billion reference bases (4Gb) in a batch and map
-all reads against each reference batch. Given a reference longer than 4Gb,
-minimap2 is unable to see all the sequences and thus can't produce a correct
-SAM header. In this case, minimap2 doesn't output any SAM header. There are two
-solutions to this issue. First, you may increase option `-I` to, for example,
-`-I8g` to index more reference bases in a batch. This is preferred if your
-machine has enough memory. Second, if your machines doesn't have enough memory
-to hold the reference index, you can use the `--split-prefix` option in a
-command line like:
-```sh
-minimap2 -ax map-ont --split-prefix=tmp ref.fa reads.fq
-```
-This second approach uses less memory, but it is slower and requires temporary
-disk space.
-
-#### 4. The output SAM is malformatted.
-
-This typically happens when you use nohup to wrap a minimap2 command line.
-Nohup is discouraged as it breaks piping. If you have to use nohup, please
-specify an output file with option `-o`.
-
-#### 5. How to output one alignment per read?
-
-You can use `--secondary=no` to suppress secondary alignments (aka multiple
-mappings), but you can't suppress supplementary alignment (aka split or
-chimeric alignment) this way. You can use samtools to filter out these
-alignments:
-```sh
-minimap2 -ax map-out ref.fa reads.fq | samtools view -F0x900
-```
-However, this is discouraged as supplementary alignment is informative.

data/ext/minimap2/MANIFEST.in

@@ -1,10 +0,0 @@
-include *.h
-include Makefile
-include ksw2_dispatch.c
-include main.c
-include README.md
-include sse2neon/emmintrin.h
-include python/cmappy.h
-include python/cmappy.pxd
-include python/mappy.pyx
-include python/README.rst