biology/vsearch: Versatile open-source tool for metagenomics

VSEARCH supports de novo and reference based chimera detection, clustering,
full-length and prefix dereplication, rereplication, reverse complementation,
masking, all-vs-all pairwise global alignment, exact and global alignment
searching, shuffling, subsampling and sorting. It also supports FASTQ file
analysis, filtering, conversion and merging of paired-end reads.

The aim of this project is to create an alternative to the USEARCH tool
developed by Robert C. Edgar (2010).

7 files changed, 162 insertions, 0 deletions

diff --git a/biology/vsearch/Makefile b/biology/vsearch/Makefile
new file mode 100644
index 000000000000..eed8a9d773c9
--- /dev/null
+++ b/biology/vsearch/Makefile
@@ -0,0 +1,32 @@
+# $FreeBSD$
+
+PORTNAME=		vsearch
+DISTVERSIONPREFIX=	v
+DISTVERSION=		2.13.3
+CATEGORIES=		biology
+
+MAINTAINER=	jwb@FreeBSD.org
+COMMENT=	Versatile open-source tool for metagenomics
+
+LICENSE=	GPLv3
+LICENSE_FILE=	${WRKSRC}/LICENSE.txt
+
+# Note: For PPC need power8 or later, little endian
+ONLY_FOR_ARCHS=		aarch64 amd64 powerpc64
+ONLY_FOR_ARCHS_REASON=	Explicitly enumerated by upstream developers
+
+LIB_DEPENDS=	libsysinfo.so:devel/libsysinfo
+
+USES=		autoreconf gmake localbase
+USE_GITHUB=	yes
+
+GH_ACCOUNT=	torognes
+GNU_CONFIGURE=	yes
+
+PLIST_FILES=	bin/vsearch man/man1/vsearch.1.gz
+PORTDOCS=	*
+
+OPTIONS_DEFINE=	DOCS
+OPTIONS_SUB=	yes
+
+.include <bsd.port.mk>
diff --git a/biology/vsearch/distinfo b/biology/vsearch/distinfo
new file mode 100644
index 000000000000..584b0d4a06ae
--- /dev/null
+++ b/biology/vsearch/distinfo
@@ -0,0 +1,3 @@
+TIMESTAMP = 1557339663
+SHA256 (torognes-vsearch-v2.13.3_GH0.tar.gz) = e5f34ece28b76403d3ba4a673eca41178fe399c35a1023dbc87d0c0da5efaa52
+SIZE (torognes-vsearch-v2.13.3_GH0.tar.gz) = 233846
diff --git a/biology/vsearch/files/patch-configure.ac b/biology/vsearch/files/patch-configure.ac
new file mode 100644
index 000000000000..94578406e2fd
--- /dev/null
+++ b/biology/vsearch/files/patch-configure.ac
@@ -0,0 +1,12 @@
+--- configure.ac.orig	2019-05-09 20:58:52 UTC
++++ configure.ac
+@@ -14,9 +14,6 @@ MACOSX_DEPLOYMENT_TARGET="10.9"
+ 
+ # Set default gcc and g++ options
+ 
+-CFLAGS='-g'
+-CXXFLAGS='-g'
+-
+ # Checks for programs.
+ AC_PROG_CXX
+ AC_PROG_RANLIB
diff --git a/biology/vsearch/files/patch-src_Makefile.am b/biology/vsearch/files/patch-src_Makefile.am
new file mode 100644
index 000000000000..e54c9a606035
--- /dev/null
+++ b/biology/vsearch/files/patch-src_Makefile.am
@@ -0,0 +1,18 @@
+--- src/Makefile.am.orig	2019-04-30 11:57:32 UTC
++++ src/Makefile.am
+@@ -1,12 +1,12 @@
+ bin_PROGRAMS = $(top_builddir)/bin/vsearch
+ 
+ if TARGET_PPC
+-AM_CXXFLAGS=-Wall -Wsign-compare -O3 -g -mcpu=power8
++AM_CXXFLAGS=-Wall -Wsign-compare
+ else
+ if TARGET_AARCH64
+-AM_CXXFLAGS=-Wall -Wsign-compare -O3 -g -march=armv8-a+simd -mtune=generic
++AM_CXXFLAGS=-Wall -Wsign-compare
+ else
+-AM_CXXFLAGS=-Wall -Wsign-compare -O3 -g -march=x86-64 -mtune=generic
++AM_CXXFLAGS=-Wall -Wsign-compare
+ endif
+ endif
+ 
diff --git a/biology/vsearch/files/patch-src_city.cc b/biology/vsearch/files/patch-src_city.cc
new file mode 100644
index 000000000000..a379a802bc90
--- /dev/null
+++ b/biology/vsearch/files/patch-src_city.cc
@@ -0,0 +1,20 @@
+$NetBSD$
+
+# Portability
+
+--- src/city.cc.orig	2019-04-30 11:57:32 UTC
++++ src/city.cc
+@@ -60,6 +60,13 @@ static uint32 UNALIGNED_LOAD32(const char *p) {
+ #define bswap_32(x) OSSwapInt32(x)
+ #define bswap_64(x) OSSwapInt64(x)
+ 
++#elif defined(__FreeBSD__)
++
++#include <sys/endian.h>
++
++#define bswap_32(x) bswap32(x)
++#define bswap_64(x) bswap64(x)
++
+ #elif defined(__NetBSD__)
+ 
+ #include <sys/types.h>
diff --git a/biology/vsearch/files/patch-src_vsearch.h b/biology/vsearch/files/patch-src_vsearch.h
new file mode 100644
index 000000000000..2e74ffff8b56
--- /dev/null
+++ b/biology/vsearch/files/patch-src_vsearch.h
@@ -0,0 +1,67 @@
+$NetBSD$
+
+# Add FreeBSD and NetBSD support
+
+--- src/vsearch.h.orig	2019-04-30 11:57:32 UTC
++++ src/vsearch.h
+@@ -138,30 +138,50 @@
+ #define bswap_32(x) _byteswap_ulong(x)
+ #define bswap_64(x) _byteswap_uint64(x)
+ 
+-#else
++#elif defined(__APPLE__)
+ 
+-#ifdef __APPLE__
+-
+ #define PROG_OS "macos"
+ #include <sys/sysctl.h>
+ #include <libkern/OSByteOrder.h>
++#include <sys/resource.h>
+ #define bswap_16(x) OSSwapInt16(x)
+ #define bswap_32(x) OSSwapInt32(x)
+ #define bswap_64(x) OSSwapInt64(x)
+ 
+-#else
++#elif defined(__linux__)
+ 
+-#ifdef __linux__
+ #define PROG_OS "linux"
+-#else
+-#define PROG_OS "unknown"
+-#endif
+-
+ #include <sys/sysinfo.h>
+ #include <byteswap.h>
++#include <sys/resource.h>
+ 
+-#endif
++#elif defined(__FreeBSD__)
+ 
++#define PROG_OS "freebsd"
++#include <sys/sysinfo.h>
++#include <sys/resource.h>
++#include <sys/endian.h>
++#define bswap_16(x) bswap16(x)
++#define bswap_32(x) bswap32(x)
++#define bswap_64(x) bswap64(x)
++
++#elif defined(__NetBSD__)
++
++#define PROG_OS "netbsd"
++#include <sys/resource.h>
++#include <sys/types.h>
++#include <sys/bswap.h>
++#define bswap_16(x) bswap16(x)
++#define bswap_32(x) bswap32(x)
++#define bswap_64(x) bswap64(x)
++// Alters behavior, but NetBSD 7 does not have getopt_long_only()
++#define getopt_long_only getopt_long
++
++#else	// Other Unix
++
++#define PROG_OS "unknown"
++#include <sys/sysinfo.h>
++#include <byteswap.h>
+ #include <sys/resource.h>
+ 
+ #endif
diff --git a/biology/vsearch/pkg-descr b/biology/vsearch/pkg-descr
new file mode 100644
index 000000000000..86302327f8bf
--- /dev/null
+++ b/biology/vsearch/pkg-descr
@@ -0,0 +1,10 @@
+VSEARCH supports de novo and reference based chimera detection, clustering,
+full-length and prefix dereplication, rereplication, reverse complementation,
+masking, all-vs-all pairwise global alignment, exact and global alignment
+searching, shuffling, subsampling and sorting. It also supports FASTQ file
+analysis, filtering, conversion and merging of paired-end reads.
+
+The aim of this project is to create an alternative to the USEARCH tool
+developed by Robert C. Edgar (2010).
+
+WWW: https://github.com/torognes/vsearch