minimap2 0.2.24.6 → 0.2.25.1

data/ext/minimap2/options.c

@@ -8,7 +8,7 @@ void mm_idxopt_init(mm_idxopt_t *opt)
 	opt->k = 15, opt->w = 10, opt->flag = 0;
 	opt->bucket_bits = 14;
 	opt->mini_batch_size = 50000000;
-	opt->batch_size = 4000000000ULL;
+	opt->batch_size = 8000000000ULL;
 }
 
 void mm_mapopt_init(mm_mapopt_t *opt)

data/ext/minimap2/pyproject.toml

@@ -0,0 +1,2 @@
+[build-system]
+requires = ["setuptools", "wheel", "Cython"]

data/ext/minimap2/python/mappy.pyx

@@ -3,7 +3,7 @@ from libc.stdlib cimport free
 cimport cmappy
 import sys
 
-__version__ = '2.24'
+__version__ = '2.25'
 
 cmappy.mm_reset_timer()
 
@@ -172,6 +172,7 @@ cdef class Aligner:
 		cdef cmappy.mm_mapopt_t map_opt
 
 		if self._idx == NULL: return
+		if ((self.map_opt.flag & 4) and (self._idx.flag & 2)): return
 		map_opt = self.map_opt
 		if max_frag_len is not None: map_opt.max_frag_len = max_frag_len
 		if extra_flags is not None: map_opt.flag |= extra_flags
@@ -217,6 +218,7 @@ cdef class Aligner:
 		cdef int l
 		cdef char *s
 		if self._idx == NULL: return
+		if ((self.map_opt.flag & 4) and (self._idx.flag & 2)): return
 		s = cmappy.mappy_fetch_seq(self._idx, name.encode(), start, end, &l)
 		if l == 0: return None
 		r = s[:l] if isinstance(s, str) else s[:l].decode()

data/ext/minimap2/seed.c

@@ -7,7 +7,7 @@ void mm_seed_mz_flt(void *km, mm128_v *mv, int32_t q_occ_max, float q_occ_frac)
 	mm128_t *a;
 	size_t i, j, st;
 	if (mv->n <= q_occ_max || q_occ_frac <= 0.0f || q_occ_max <= 0) return;
-	KMALLOC(km, a, mv->n);
+	a = Kmalloc(km, mm128_t, mv->n);
 	for (i = 0; i < mv->n; ++i)
 		a[i].x = mv->a[i].x, a[i].y = i;
 	radix_sort_128x(a, a + mv->n);

data/ext/minimap2/setup.py

@@ -1,29 +1,40 @@
 try:
 	from setuptools import setup, Extension
+	from setuptools.command.build_ext import build_ext
 except ImportError:
 	from distutils.core import setup
 	from distutils.extension import Extension
+	from distutils.command.build_ext import build_ext
 
-import sys, platform
+import sys, platform, subprocess
 
-sys.path.append('python')
-
-extra_compile_args = ['-DHAVE_KALLOC']
-include_dirs = ["."]
-
-if platform.machine() in ["aarch64", "arm64"]:
-	include_dirs.append("sse2neon/")
-	extra_compile_args.extend(['-ftree-vectorize', '-DKSW_SSE2_ONLY', '-D__SSE2__'])
-else:
-	extra_compile_args.append('-msse4.1') # WARNING: ancient x86_64 CPUs don't have SSE4
 
 def readme():
 	with open('python/README.rst') as f:
 		return f.read()
 
+
+class LibMM2Build(build_ext):
+	# Uses Makefile to build library, avoids duplicating logic
+	# determining which objects to compile but does require
+	# end users to have Make (since precompiled wheels are not
+	# distributed on PyPI).
+	def run(self):
+		def compile_libminimap2(*args, **kwargs):
+			cmd = ['make', 'libminimap2.a'] + list(args)
+			subprocess.check_call(cmd)
+		options = []
+		if platform.machine() in ["aarch64", "arm64"]:
+			options = ["arm_neon=1", "aarch64=1"]
+		self.execute(
+			compile_libminimap2, options,
+			'Compiling libminimap2 using Makefile')
+		build_ext.run(self)
+
+
 setup(
 	name = 'mappy',
-	version = '2.24',
+	version = '2.25',
 	url = 'https://github.com/lh3/minimap2',
 	description = 'Minimap2 python binding',
 	long_description = readme(),
@@ -32,16 +43,15 @@ setup(
 	license = 'MIT',
 	keywords = 'sequence-alignment',
 	scripts = ['python/minimap2.py'],
-	ext_modules = [Extension('mappy',
-		sources = ['python/mappy.pyx', 'align.c', 'bseq.c', 'lchain.c', 'seed.c', 'format.c', 'hit.c', 'index.c', 'pe.c', 'options.c',
-				   'ksw2_extd2_sse.c', 'ksw2_exts2_sse.c', 'ksw2_extz2_sse.c', 'ksw2_ll_sse.c',
-				   'kalloc.c', 'kthread.c', 'map.c', 'misc.c', 'sdust.c', 'sketch.c', 'esterr.c', 'splitidx.c'],
-		depends = ['minimap.h', 'bseq.h', 'kalloc.h', 'kdq.h', 'khash.h', 'kseq.h', 'ksort.h',
-				   'ksw2.h', 'kthread.h', 'kvec.h', 'mmpriv.h', 'sdust.h',
-				   'python/cmappy.h', 'python/cmappy.pxd'],
-		extra_compile_args = extra_compile_args,
-		include_dirs = include_dirs,
-		libraries = ['z', 'm', 'pthread'])],
+	cmdclass = {'build_ext': LibMM2Build},
+	ext_modules = [
+		Extension(
+			'mappy',
+			sources = ['python/mappy.pyx'],
+			depends = ['python/cmappy.h', 'python/cmappy.pxd'],
+			include_dirs = ['.'],
+			extra_objects = ['libminimap2.a'],
+			libraries = ['z', 'm', 'pthread'])],
 	classifiers = [
 		'Development Status :: 5 - Production/Stable',
 		'License :: OSI Approved :: MIT License',

data/ext/minimap2.patch

@@ -1,5 +1,5 @@
---- Makefile.org	2021-05-27 15:45:11.993128205 +0900
-+++ Makefile	2021-05-27 15:46:02.320569154 +0900
+--- a/Makefile
++++ b/Makefile
 @@ -1,9 +1,9 @@
 -CFLAGS=		-g -Wall -O2 -Wc++-compat #-Wextra
 +CFLAGS=		-g -Wall -O2 -Wc++-compat -fPIC #-Wextra
@@ -12,7 +12,7 @@
  PROG=		minimap2
  PROG_EXTRA=	sdust minimap2-lite
  LIBS=		-lm -lz -lpthread
-@@ -130,3 +130,4 @@ sdust.o: kalloc.h kdq.h kvec.h sdust.h
+@@ -134,3 +134,4 @@ sdust.o: kalloc.h kdq.h kvec.h sdust.h
  seed.o: mmpriv.h minimap.h bseq.h kseq.h kalloc.h ksort.h
  sketch.o: kvec.h kalloc.h mmpriv.h minimap.h bseq.h kseq.h
  splitidx.o: mmpriv.h minimap.h bseq.h kseq.h

data/lib/minimap2/aligner.rb

@@ -139,6 +139,7 @@ module Minimap2
     )
 
       return if index.null?
+      return if (map_opt[:flag] & 4).zero? && (index[:flag] & 2).zero?
 
       map_opt[:max_frag_len] = max_frag_len if max_frag_len
       map_opt[:flag] |= extra_flags if extra_flags
@@ -201,6 +202,9 @@ module Minimap2
     # @param stop
 
     def seq(name, start = 0, stop = 0x7fffffff)
+      return if index.null?
+      return if (map_opt[:flag] & 4).zero? && (index[:flag] & 2).zero?
+
       lp = ::FFI::MemoryPointer.new(:int)
       s = FFI.mappy_fetch_seq(index, name, start, stop, lp)
       l = lp.read_int

data/lib/minimap2/ffi/constants.rb

@@ -34,10 +34,12 @@ module Minimap2
     NO_END_FLT       = 0x10000000
     HARD_MLEVEL      = 0x20000000
     SAM_HIT_ONLY     = 0x40000000
-    RMQ              = 0x80000000  # LL
-    QSTRAND          = 0x100000000 # LL
-    NO_INV           = 0x200000000 # LL
-    NO_HASH_NAME     = 0x400000000 # LL
+    RMQ              = 0x80000000
+    QSTRAND          = 0x100000000
+    NO_INV           = 0x200000000
+    NO_HASH_NAME     = 0x400000000
+    SPLICE_OLD       = 0x800000000
+    SECONDARY_SEQ    = 0x1000000000 # output SEQ field for seqondary alignments using hard clipping
 
     HPC              = 0x1
     NO_SEQ           = 0x2
@@ -74,79 +76,6 @@ module Minimap2
         :a, MM128.ptr
     end
 
-    # indexing option
-    class IdxOpt < ::FFI::Struct
-      layout \
-        :k,               :short,
-        :w,               :short,
-        :flag,            :short,
-        :bucket_bits,     :short,
-        :mini_batch_size, :int64_t,
-        :batch_size,      :uint64_t
-    end
-
-    # mapping option
-    class MapOpt < ::FFI::Struct
-      layout \
-        :flag,                 :int64_t, # see MM_F_* macros
-        :seed,                 :int,
-        :sdust_thres,          :int,     # score threshold for SDUST; 0 to disable
-        :max_qlen,             :int,     # max query length
-        :bw,                   :int,     # bandwidth
-        :bw_long,              :int,
-        :max_gap,              :int,     # break a chain if there are no minimizers in a max_gap window
-        :max_gap_ref,          :int,
-        :max_frag_len,         :int,
-        :max_chain_skip,       :int,
-        :max_chain_iter,       :int,
-        :min_cnt,              :int,     # min number of minimizers on each chain
-        :min_chain_score,      :int,     # min chaining score
-        :chain_gap_scale,      :float,
-        :chain_skip_scale,     :float,
-        :rmq_size_cap,         :int,
-        :rmq_inner_dist,       :int,
-        :rmq_rescue_size,      :int,
-        :rmq_rescue_ratio,     :float,
-        :mask_level,           :float,
-        :mask_len,             :int,
-        :pri_ratio,            :float,
-        :best_n,               :int,     # top best_n chains are subjected to DP alignment
-        :alt_drop,             :float,
-        :a,                    :int,     # matching score
-        :b,                    :int,     # mismatch
-        :q,                    :int,     # gap-open
-        :e,                    :int,     # gap-ext
-        :q2,                   :int,     # gap-open
-        :e2,                   :int,     # gap-ext
-        :sc_ambi,              :int,     # score when one or both bases are "N"
-        :noncan,               :int,     # cost of non-canonical splicing sites
-        :junc_bonus,           :int,
-        :zdrop,                :int,     # break alignment if alignment score drops too fast along the diagonal
-        :zdrop_inv,            :int,
-        :end_bonus,            :int,
-        :min_dp_max,           :int,     # drop an alignment if the score of the max scoring segment is below this threshold
-        :min_ksw_len,          :int,
-        :anchor_ext_len,       :int,
-        :anchor_ext_shift,     :int,
-        :max_clip_ratio,       :float,   # drop an alignment if BOTH ends are clipped above this ratio
-        :rank_min_len,         :int,
-        :rank_frac,            :float,
-        :pe_ori,               :int,
-        :pe_bonus,             :int,
-        :mid_occ_frac,         :float,   # only used by mm_mapopt_update(); see below
-        :q_occ_frac,           :float,
-        :min_mid_occ,          :int32,
-        :max_mid_occ,          :int32,
-        :mid_occ,              :int32, # ignore seeds with occurrences above this threshold
-        :max_occ,              :int32,
-        :max_max_occ,          :int32,
-        :occ_dist,             :int32,
-        :mini_batch_size,      :int64_t, # size of a batch of query bases to process in parallel
-        :max_sw_mat,           :int64_t,
-        :cap_kalloc,           :int64_t,
-        :split_prefix,         :string
-    end
-
     # minimap2 index
     class IdxSeq < ::FFI::Struct
       layout \
@@ -173,17 +102,6 @@ module Minimap2
         :h,     :pointer
     end
 
-    # index reader
-    class IdxReader < ::FFI::Struct
-      layout \
-        :is_idx,      :int,
-        :n_parts,     :int,
-        :idx_size,    :int64_t,
-        :opt,         IdxOpt,
-        :fp_out,      :pointer, # FILE
-        :seq_or_idx,  :pointer  # FIXME: Union mm_bseq_files or FILE
-    end
-
     # minimap2 alignment
     class Extra < ::FFI::BitStruct
       layout \
@@ -242,6 +160,90 @@ module Minimap2
                 :dummy,           5
     end
 
+    # indexing option
+    class IdxOpt < ::FFI::Struct
+      layout \
+        :k,               :short,
+        :w,               :short,
+        :flag,            :short,
+        :bucket_bits,     :short,
+        :mini_batch_size, :int64_t,
+        :batch_size,      :uint64_t
+    end
+
+    # mapping option
+    class MapOpt < ::FFI::Struct
+      layout \
+        :flag,                 :int64_t, # see MM_F_* macros
+        :seed,                 :int,
+        :sdust_thres,          :int,     # score threshold for SDUST; 0 to disable
+        :max_qlen,             :int,     # max query length
+        :bw,                   :int,     # bandwidth
+        :bw_long,              :int,
+        :max_gap,              :int,     # break a chain if there are no minimizers in a max_gap window
+        :max_gap_ref,          :int,
+        :max_frag_len,         :int,
+        :max_chain_skip,       :int,
+        :max_chain_iter,       :int,
+        :min_cnt,              :int,     # min number of minimizers on each chain
+        :min_chain_score,      :int,     # min chaining score
+        :chain_gap_scale,      :float,
+        :chain_skip_scale,     :float,
+        :rmq_size_cap,         :int,
+        :rmq_inner_dist,       :int,
+        :rmq_rescue_size,      :int,
+        :rmq_rescue_ratio,     :float,
+        :mask_level,           :float,
+        :mask_len,             :int,
+        :pri_ratio,            :float,
+        :best_n,               :int,     # top best_n chains are subjected to DP alignment
+        :alt_drop,             :float,
+        :a,                    :int,     # matching score
+        :b,                    :int,     # mismatch
+        :q,                    :int,     # gap-open
+        :e,                    :int,     # gap-ext
+        :q2,                   :int,     # gap-open
+        :e2,                   :int,     # gap-ext
+        :sc_ambi,              :int,     # score when one or both bases are "N"
+        :noncan,               :int,     # cost of non-canonical splicing sites
+        :junc_bonus,           :int,
+        :zdrop,                :int,     # break alignment if alignment score drops too fast along the diagonal
+        :zdrop_inv,            :int,
+        :end_bonus,            :int,
+        :min_dp_max,           :int,     # drop an alignment if the score of the max scoring segment is below this threshold
+        :min_ksw_len,          :int,
+        :anchor_ext_len,       :int,
+        :anchor_ext_shift,     :int,
+        :max_clip_ratio,       :float,   # drop an alignment if BOTH ends are clipped above this ratio
+        :rank_min_len,         :int,
+        :rank_frac,            :float,
+        :pe_ori,               :int,
+        :pe_bonus,             :int,
+        :mid_occ_frac,         :float,   # only used by mm_mapopt_update(); see below
+        :q_occ_frac,           :float,
+        :min_mid_occ,          :int32,
+        :max_mid_occ,          :int32,
+        :mid_occ,              :int32, # ignore seeds with occurrences above this threshold
+        :max_occ,              :int32,
+        :max_max_occ,          :int32,
+        :occ_dist,             :int32,
+        :mini_batch_size,      :int64_t, # size of a batch of query bases to process in parallel
+        :max_sw_mat,           :int64_t,
+        :cap_kalloc,           :int64_t,
+        :split_prefix,         :string
+    end
+
+    # index reader
+    class IdxReader < ::FFI::Struct
+      layout \
+        :is_idx,      :int,
+        :n_parts,     :int,
+        :idx_size,    :int64_t,
+        :opt,         IdxOpt,
+        :fp_out,      :pointer, # FILE
+        :seq_or_idx,  :pointer  # FIXME: Union mm_bseq_files or FILE
+    end
+
     # memory buffer for thread-local storage during mapping
     class TBuf < ::FFI::Struct
       layout \

data/lib/minimap2/version.rb

@@ -1,6 +1,6 @@
 # frozen_string_literal: true
 
 module Minimap2
-  #  Minimap2-2.24 (r1122)
-  VERSION = "0.2.24.6"
+  # Minimap2-2.25 (r1173)
+  VERSION = "0.2.25.1"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: minimap2
 version: !ruby/object:Gem::Version
-  version: 0.2.24.6
+  version: 0.2.25.1
 platform: ruby
 authors:
 - kojix2
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2023-04-27 00:00:00.000000000 Z
+date: 2023-05-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: ffi
@@ -194,6 +194,7 @@ files:
 - 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
@@ -260,7 +261,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.4.1
+rubygems_version: 3.4.6
 signing_key: 
 specification_version: 4
 summary: minimap2