[med-svn] [Git][med-team/libvcflib][master] 8 commits: routine-update: New upstream version
Andreas Tille (@tille)
gitlab at salsa.debian.org
Sun Feb 5 17:21:43 GMT 2023
Andreas Tille pushed to branch master at Debian Med / libvcflib
Commits:
677dbcbb by Andreas Tille at 2023-02-05T18:07:41+01:00
routine-update: New upstream version
- - - - -
c4cb620d by Andreas Tille at 2023-02-05T18:07:42+01:00
New upstream version 1.0..6.zip+dfsg
- - - - -
ccbe42de by Andreas Tille at 2023-02-05T18:08:11+01:00
Update upstream source from tag 'upstream/1.0.#.6.zip+dfsg'
Update to upstream version '1.0..6.zip+dfsg'
with Debian dir 906c083c48c690a25eceff12d3b3e61115731922
- - - - -
e30dfcb1 by Andreas Tille at 2023-02-05T18:12:05+01:00
Refresh patches
- - - - -
03ab2426 by Andreas Tille at 2023-02-05T18:14:33+01:00
Do not permit zip files as download (after there was a broken tag which prefered zip over tar.gz)
- - - - -
b09b362c by Andreas Tille at 2023-02-05T18:17:32+01:00
Do not exclude Files that are not in upstream tarball any more
- - - - -
a206b088 by Andreas Tille at 2023-02-05T18:18:28+01:00
New upstream version 1.0.6+dfsg
- - - - -
7b2b4502 by Andreas Tille at 2023-02-05T18:18:57+01:00
Update upstream source from tag 'upstream/1.0.6+dfsg'
Update to upstream version '1.0.6+dfsg'
with Debian dir 861c6e166ca5a680ac613bcda81ee34c6dcb1092
- - - - -
19 changed files:
- .gitmodules
- CMakeLists.txt
- RELEASE_NOTES.md
- VERSION
- debian/changelog
- debian/control
- debian/copyright
- debian/patches/no_fsom
- debian/patches/shared_lib.patch
- debian/patches/tabix_linking
- debian/patches/use_debian_packaged_fastahack.patch
- debian/patches/use_debian_packaged_smithwaterman.patch
- debian/watch
- guix.scm
- src/vcfcreatemulti.cpp
- src/vcfwave.cpp
- src/zig/build.zig
- src/zig/vcf.zig
- test/doc/performance.md
Changes:
=====================================
.gitmodules
=====================================
@@ -28,7 +28,7 @@
[submodule "contrib/tabixpp"]
path = contrib/tabixpp
url = https://github.com/vcflib/tabixpp.git
- branch = pjotrp
+ branch = master
[submodule "contrib/WFA2-lib"]
path = contrib/WFA2-lib
url = https://github.com/smarco/WFA2-lib
=====================================
CMakeLists.txt
=====================================
@@ -1,3 +1,11 @@
+# For Debian currently with
+#
+# cd build
+# cmake -DCMAKE_BUILD_TYPE=RelWithDebInfo -DZIG=OFF -DWFA_GITMODULE=OFF ..
+# make
+# make test
+# make install
+#
cmake_minimum_required(VERSION 3.16)
project(vcflib)
@@ -5,10 +13,12 @@ set(CMAKE_CXX_STANDARD 17)
include(ExternalProject)
include(FeatureSummary)
+include(GNUInstallDirs)
find_package(PkgConfig REQUIRED)
find_package(pybind11 CONFIG)
+include(GNUInstallDirs)
include(FindBZip2)
include(FindLibLZMA)
include(FindZLIB)
@@ -27,7 +37,8 @@ option(OPENMP "Enable OpenMP" ON) # disabling does not work because of vcfwave
option(PROFILING "Enable profiling" OFF)
option(GPROF "Enable gprof profiling" OFF)
option(ASAN "Use address sanitiser" OFF)
-option(ZIG "Disable using the zig compiler" ON)
+option(ZIG "Set to OFF to disable the zig code" ON)
+option(WFA_GITMODULE "Force local git submodule for WFA2LIB" ON) # disable in distros, you may need to add path to WFA_INCLUDE_DIRS
include(CheckIPOSupported) # adds lto
check_ipo_supported(RESULT ipo_supported OUTPUT output)
@@ -42,8 +53,6 @@ endif(OPENMP)
find_package(ZLIB)
set_package_properties(ZLIB PROPERTIES TYPE REQUIRED)
-#find_package(LIBLZMA)
-#set_package_properties(LIBLZMA PROPERTIES TYPE REQUIRED)
find_package(Threads)
set_package_properties(Threads PROPERTIES TYPE REQUIRED)
@@ -91,7 +100,7 @@ endif(GPROF)
if (ZIG)
find_program(ZIG_EXE NAMES "zig")
if (NOT ZIG_EXE)
- MESSAGE(FATAL_ERROR "zig binary not found in PATH: either use cmake -DNO_ZIG option or add zig to the PATH")
+ MESSAGE(FATAL_ERROR "zig binary not found in PATH. zig is used for vcfcreatemulti. Either use cmake -DZIG=OFF option or add zig to the PATH")
endif (NOT ZIG_EXE)
endif(ZIG)
@@ -118,7 +127,9 @@ if (NOT TABIX_FOUND)
)
endif()
-set(WFA_LOCAL contrib/WFA2-lib)
+if(WFA_GITMODULE)
+ set(WFA_LOCAL contrib/WFA2-lib)
+endif()
file(GLOB INCLUDES
src/*.h*
@@ -248,23 +259,12 @@ set(BINS
vcfinfo2qual
vcfglbound
vcfinfosummarize
+ vcfcreatemulti
)
-if (ZIG)
- list(APPEND BINS vcfcreatemulti)
-endif (ZIG)
-
set(SCRIPTS
bed2region
bgziptabix
- # plotBfst.R
- # plotHaplotypes.R
- # plotHapLrt.R
- # plotPfst.R
- # plot_roc.r
- # plotSmoothed.R
- # plotWCfst.R
- # plotXPEHH.R
vcf2bed.py
vcf2sqlite.py
vcfbiallelic
@@ -307,7 +307,6 @@ file (STRINGS "VERSION" BUILD_NUMBER)
add_definitions(-DVCFLIB_VERSION="${BUILD_NUMBER}")
add_definitions(-DVERSION="${BUILD_NUMBER}")
-
# ---- Build htslib
#
# Note by default we use the distributed htslib! These are
@@ -356,42 +355,17 @@ if (HTSLIB_LOCAL)
endif(HTSLIB_LOCAL)
-# FIXME: hard-coded compile switches need to be fixed upstream
-if (ASAN)
- set(wfa_MAKE_ARGS BUILD_WFA_PARALLEL=1 BUILD_TOOLS=0 BUILD_EXAMPLES=0 CC=gcc CC_FLAGS=-fPIC setup asan lib_wfa)
-else()
- set(wfa_MAKE_ARGS BUILD_WFA_PARALLEL=1 BUILD_TOOLS=0 BUILD_EXAMPLES=0 CC=gcc CC_FLAGS=-fPIC setup lib_wfa)
-endif()
-
-# if ((${CMAKE_BUILD_TYPE} MATCHES Release) OR (${CMAKE_BUILD_TYPE} MATCHES RelWithDebInfo))
-# set(wfa_MAKE_ARGS ${wfa_MAKE_ARGS})
-# endif()
-
-ExternalProject_Add(wfa-EXT
- SOURCE_DIR "${CMAKE_SOURCE_DIR}/${WFA_LOCAL}"
- UPDATE_COMMAND ""
- CONFIGURE_COMMAND ""
- INSTALL_COMMAND ""
- BUILD_IN_SOURCE ON
- BUILD_ALWAYS ON
- BUILD_COMMAND $(MAKE) ${wfa_MAKE_ARGS}
-)
-ExternalProject_Get_property(wfa-EXT SOURCE_DIR)
-set(WFA_INCLUDE_DIRS ${SOURCE_DIR})
-set(WFA_LINK_LIBRARIES ${SOURCE_DIR}/lib/libwfa.a )
-# set(WFA_LINK_LIBRARIES ${SOURCE_DIR}/lib/libwfa.a ${SOURCE_DIR}/lib/libwfacpp.a )
-link_directories($SOURCE_DIR}/lib)
-MESSAGE(STATUS ${WFA_LINK_LIBRARIES})
-
-# add_library(WAFCPP_STATIC_LIB STATIC IMPORTED)
-# set_target_properties(WAFCPP_STATIC_LIB PROPERTIES IMPORTED_LOCATION ${WFA_LINK_LIBRARIES})
-# add_dependencies(WAFCPP_STATIC_LIB wfa-EXT)
-
-add_custom_target(wfa
- DEPENDS wfa-EXT
- VERBATIM)
+if(WFA_GITMODULE)
+ set(WFA_INCLUDE_DIRS ${WFA_LOCAL})
+ add_subdirectory(${WFA_LOCAL})
+ set(WFALIB wfa2) # pick up the wfa2 lib target from the included CMakeLists.txt
+else(WFA_GITMODULE)
+ set(WFA_INCLUDE_DIRS ${CMAKE_INSTALL_PREFIX}/${CMAKE_INSTALL_INCLUDEDIR}/wfa2lib)
+ find_library(WFALIB wfa2 wfa) # distro search for shared lib
+endif(WFA_GITMODULE)
include_directories(${WFA_INCLUDE_DIRS})
+MESSAGE(STATUS "WFA using include ${WFA_INCLUDE_DIRS}")
# ZIG VCF imported library is part of VCFLIB source tree. We designate it an
# external project so we can run 'zig build'
@@ -404,31 +378,25 @@ ExternalProject_Add(ZIG-EXT
INSTALL_COMMAND ""
BUILD_IN_SOURCE ON
BUILD_ALWAYS ON
- BUILD_COMMAND ${CMAKE_SOURCE_DIR}/src/zig/compile.sh -Drelease-fast=true -freference-trace
+ BUILD_COMMAND ${CMAKE_SOURCE_DIR}/src/zig/compile.sh # -Drelease-fast=true -freference-trace
)
ExternalProject_Get_property(ZIG-EXT SOURCE_DIR)
set(ZIG_INCLUDE_DIRS ${SOURCE_DIR})
set(ZIG_LINK_LIBRARIES ${SOURCE_DIR}/zig-out/lib/libzig.a)
-# link_directories($SOURCE_DIR}/lib)
-# MESSAGE(STATUS ${ZIG_LINK_LIBRARIES})
add_library(ZIGCPP_STATIC_LIB STATIC IMPORTED)
set_target_properties(ZIGCPP_STATIC_LIB PROPERTIES IMPORTED_LOCATION ${ZIG_LINK_LIBRARIES})
add_dependencies(ZIGCPP_STATIC_LIB ZIG-EXT vcf.zig)
-# add_custom_target(ZIGLIB
-# DEPENDS ZIG-EXT
-# VERBATIM)
-
include_directories(${ZIG_INCLUDE_DIRS})
else (ZIG)
set (ZIG_LINK_LIBRARIES )
+ add_definitions(-DNO_ZIG=1)
endif (ZIG)
set(vcflib_DEPS
CURL::libcurl
- wfa
)
if (ZIG)
@@ -455,7 +423,6 @@ if (ZIG)
list(APPEND vcflib_LIBS ${ZIG_LINK_LIBRARIES})
endif()
-# target_link_libraries(vcflib PUBLIC ${vcflib_LIBS})
add_dependencies(vcflib ${vcflib_DEPS})
# ---- Build all
@@ -464,8 +431,9 @@ if (NOT BUILD_ONLY_LIB)
foreach(BIN ${BINS})
add_executable(${BIN} src/${BIN}.cpp)
add_dependencies(${BIN} vcflib)
- target_link_libraries(${BIN} PUBLIC ${vcflib_LIBS} vcflib)
+ target_link_libraries(${BIN} PUBLIC ${vcflib_LIBS} vcflib ${WFALIB})
endforeach(BIN ${BINS})
+ # target_link_libraries(vcfwave PUBLIC ${WFALIB})
install(TARGETS ${BINS} RUNTIME DESTINATION bin)
# ---- Copy scripts
@@ -478,7 +446,7 @@ endif()
# ---- Python bindings - mostly for testing at this stage
pybind11_add_module(pyvcflib "${CMAKE_SOURCE_DIR}/src/pythonffi.cpp")
add_dependencies(pyvcflib ${vcflib_DEPS})
-target_link_libraries(pyvcflib PUBLIC vcflib "${vcflib_LIBS}")
+target_link_libraries(pyvcflib PUBLIC vcflib ${vcflib_LIBS} ${WFALIB})
install(TARGETS pyvcflib LIBRARY DESTINATION lib)
# ---- Test
@@ -548,7 +516,6 @@ endif (ZIG)
add_pydoctest(vcfnulldotslashdot)
add_doctest(doc/vcfintersect)
add_doctest(doc/vcffilter)
-# add_pydoctest_fullname(../README.md)
# ---- Build docs
#
=====================================
RELEASE_NOTES.md
=====================================
@@ -5,32 +5,35 @@ and
## TODO
-- [X] vcfcreatemulti merge multiple rows
- + [X] rewrite vcfcreatemulti using zig
- + [X] merge genotypes correctly, with tests
- + [X] adjust info and genotypes for variants that have multiple alts already (now errors)
- + [X] handle phase
- + [X] document using with `vcfwave` and `bcftools norm -m-`
- + [X] document building with zig
- + [X] added progress bar to vcfwave and vcfcreatemulti with update to tabixpp
- + [X] default vcfwave and vcfcreatemulti to nextgen mode
- + [X] why is vcfwave on a single thread?
- + [X] check file is sorted for vcfcreatemulti and improve suggestions
- + [X] update WFA-LIB to main
- + [X] add tests (zig memory handler)
- + [X] check for memory leaks
- + [ ] make CI pass with recent zig in path
-- [ ] RELEASE 1.0.5
+- [ ] vcfcreatemulti: fix problem with slow and wrong complex regions
- [ ] complete vcfcreatemulti merge multiple rows
+ [ ] check for indels which are really the same
+ [ ] combine vcfwave duplicated functionality
-- [ ] vcfuniq combine genotypes from non-unique records: use vcfcreatemulti and bcftools norm as above
-- [ ] vcfwave check polyploids?
-- [ ] RELEASE 1.0.6
+- bgzip
+- tabix -p vcf my_file.vcf.gz
+- pangenie, vg deconstruct, vcfbub
-## ChangeLog v1.0.5 (2023xxxx)
+## ChangeLog v1.0.6 (20230129)
-Release with some major changes.
+Vcflib maintenance release - mostly for including in Debian
+
++ Fixed zig complaining about leaking memory
++ Added CMake Debian support with -DWFA_GITMODULE=OFF
++ Introduced CMake include(GNUInstallDirs)
++ Successfully built wfa2 using embedded CMakeLists.txt
++ Cleaned up CMakeLists.txt removing comments etc.
++ Reintroduced vcfcreatemulti in legacy mode when ZIG=OFF (for Debian)
+
+## ChangeLog v1.0.5 (20230116)
+
+Vcflib's first *Humpty Dumpty* release: [vcfcreatemulti](./doc/vcfcreatemulti.md) is the natural companion to [vcfwave](./doc/vcfwave.md).
+
+Often variant callers are not perfect.
+**vcfwave** with its companion tool **vcfcreatemulti** can take an existing VCF file that contains multiple complex overlapping and even nested alleles and, unlike Humpty Dumpty, take them apart and put them together again.
+Thereby, hopefully, creating sane VCF output that is useful for analysis and getting rid of false positives.
+
+We created these tools by including the state-of-the-art [biWFA](https://github.com/smarco/WFA2-lib) wavefront aligner.
+The tools are particularly useful for the output from structural variation callers and pangenome genotypers, such as used by the Human Pangenome Reference Consortium (HPRC) because of overlapping ALT segments.
Important changes:
@@ -58,6 +61,15 @@ Introduction of O(n) wavefront algorithm WF to replace O(n^2) Smith-Waterman SW.
+ Added python testing framework
+ Added tabixpp back in as a submodule, fixes https://github.com/vcflib/vcflib/issues/305
+ Optimizations and bug fixes. (thanks @mphschmitt)
++ vcfcreatemulti merge multiple rows
++ rewrite of vcfcreatemulti using zig
++ vcfcreatemulti merge genotypes correctly, with tests
++ vcfcreatemulti adjust info and genotypes for variants that have multiple alts already (now errors)
++ vcfcreatemulti handle phase
++ vcfcreatemulti document building with zig
++ vcfcreatemulti added progress bar to vcfwave and vcfcreatemulti with update to tabixpp
++ vcfcreatemulti default vcfwave and vcfcreatemulti to nextgen mode
++ vcfcreatemulti check file is sorted for vcfcreatemulti and improve suggestions
## ChangeLog v1.0.4
=====================================
VERSION
=====================================
@@ -1 +1 @@
-1.0.4
+1.0.6
=====================================
debian/changelog
=====================================
@@ -1,4 +1,4 @@
-libvcflib (1.0.5+dfsg-1) UNRELEASED; urgency=medium
+libvcflib (1.0.6+dfsg-1) UNRELEASED; urgency=medium
* New upstream version
* Standards-Version: 4.6.2 (routine-update)
@@ -6,8 +6,9 @@ libvcflib (1.0.5+dfsg-1) UNRELEASED; urgency=medium
* Build-Depends: python3-pybind11, libtabixpp-dev
* Fix name of upstream download tarball in watch file
* Versioned Build-Depends: libwfa2-dev (>= 2.3.2)
+ * Do not exclude Files that are not in upstream tarball any more
- -- Andreas Tille <tille at debian.org> Mon, 16 Jan 2023 11:31:16 +0100
+ -- Andreas Tille <tille at debian.org> Sun, 05 Feb 2023 18:07:41 +0100
libvcflib (1.0.3+dfsg-2) unstable; urgency=medium
=====================================
debian/control
=====================================
@@ -12,8 +12,7 @@ Build-Depends: debhelper-compat (= 13),
libsmithwaterman-dev,
libssw-dev,
libfastahack-dev,
- libtabixpp-dev,
- libwfa2-dev (>= 2.3.2),
+ libwfa2-dev,
pkg-config,
libgtest-dev <!nocheck>,
python3-pybind11
=====================================
debian/copyright
=====================================
@@ -5,11 +5,6 @@ Source: https://github.com/vcflib/vcflib
Files-Excluded: */paper
src/ssw.*
src/ssw_cpp.*
- fastahack
- fsom
- googletest
- libVCFH
- smithwaterman
src/simde
Files: *
=====================================
debian/patches/no_fsom
=====================================
@@ -3,7 +3,7 @@ Subject: fsom is not used
Forwarded: not-needed
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
-@@ -150,7 +150,6 @@ set(vcflib_SOURCE
+@@ -161,7 +161,6 @@ set(vcflib_SOURCE
contrib/smithwaterman/IndelAllele.cpp
contrib/smithwaterman/disorder.cpp
contrib/smithwaterman/LeftAlign.cpp
=====================================
debian/patches/shared_lib.patch
=====================================
@@ -4,7 +4,7 @@ Description: Create shared lib instead of static
Forwarded: not-needed
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
-@@ -147,19 +147,19 @@ set(vcflib_SOURCE
+@@ -158,19 +158,19 @@ set(vcflib_SOURCE
contrib/c-progress-bar/progress.c
)
@@ -39,5 +39,5 @@ Forwarded: not-needed
+ SOVERSION ${MAJOR_BUILD_NUMBER}
+ )
-
# ---- Build htslib
+ #
=====================================
debian/patches/tabix_linking
=====================================
@@ -3,7 +3,7 @@ Description: link against tabixpp
Forwarded: not-needed
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
-@@ -158,6 +158,7 @@ add_library(vcflib SHARED
+@@ -169,6 +169,7 @@ add_library(vcflib SHARED
target_link_libraries(vcflib
PkgConfig::SMITHWATERMAN
PkgConfig::FASTAHACK
=====================================
debian/patches/use_debian_packaged_fastahack.patch
=====================================
@@ -4,7 +4,7 @@ Description: Use Debian packaged libfastahack (and libdisorder)
Forwarded: not-needed
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
-@@ -50,6 +50,7 @@ set_package_properties(Threads PROPERTIE
+@@ -59,6 +59,7 @@ set_package_properties(Threads PROPERTIE
pkg_check_modules(HTSLIB htslib) # Optionally builds from contrib/
pkg_check_modules(TABIXPP tabixpp) # Optionally builds from contrib/
pkg_check_modules(SMITHWATERMAN REQUIRED IMPORTED_TARGET libsmithwaterman)
@@ -12,7 +12,7 @@ Forwarded: not-needed
# ---- Build switches
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
-@@ -99,7 +100,6 @@ endif(ZIG)
+@@ -108,7 +109,6 @@ endif(ZIG)
# ---- Include files
include_directories(include)
@@ -20,7 +20,7 @@ Forwarded: not-needed
include_directories(contrib/intervaltree)
include_directories(contrib/multichoose)
include_directories(contrib/filevercmp)
-@@ -124,7 +124,6 @@ file(GLOB INCLUDES
+@@ -135,7 +135,6 @@ file(GLOB INCLUDES
src/*.h*
contrib/multichoose/*.h*
contrib/intervaltree/*.h*
@@ -28,7 +28,7 @@ Forwarded: not-needed
contrib/filevercmp/*.h*
)
-@@ -150,6 +149,7 @@ set(vcflib_SOURCE
+@@ -161,6 +160,7 @@ set(vcflib_SOURCE
target_link_libraries(vcflib
PkgConfig::SMITHWATERMAN
=====================================
debian/patches/use_debian_packaged_smithwaterman.patch
=====================================
@@ -15,7 +15,7 @@ Forwarded: not-needed
#include "multichoose.h"
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
-@@ -49,6 +49,7 @@ set_package_properties(Threads PROPERTIE
+@@ -58,6 +58,7 @@ set_package_properties(Threads PROPERTIE
pkg_check_modules(HTSLIB htslib) # Optionally builds from contrib/
pkg_check_modules(TABIXPP tabixpp) # Optionally builds from contrib/
@@ -23,7 +23,7 @@ Forwarded: not-needed
# ---- Build switches
set(CMAKE_POSITION_INDEPENDENT_CODE ON)
-@@ -100,7 +101,6 @@ endif(ZIG)
+@@ -109,7 +110,6 @@ endif(ZIG)
include_directories(include)
include_directories(contrib/fastahack)
include_directories(contrib/intervaltree)
@@ -31,7 +31,7 @@ Forwarded: not-needed
include_directories(contrib/multichoose)
include_directories(contrib/filevercmp)
include_directories(contrib/c-progress-bar)
-@@ -124,7 +124,6 @@ file(GLOB INCLUDES
+@@ -135,7 +135,6 @@ file(GLOB INCLUDES
src/*.h*
contrib/multichoose/*.h*
contrib/intervaltree/*.h*
@@ -39,7 +39,7 @@ Forwarded: not-needed
contrib/fastahack/*.h*
contrib/filevercmp/*.h*
)
-@@ -145,15 +144,14 @@ set(vcflib_SOURCE
+@@ -156,15 +155,14 @@ set(vcflib_SOURCE
src/cigar.cpp
src/allele.cpp
contrib/fastahack/Fasta.cpp
=====================================
debian/watch
=====================================
@@ -1,4 +1,4 @@
version=4
opts="uversionmangle=s/-rc/~rc/,repacksuffix=+dfsg,dversionmangle=auto,filenamemangle=s%(?:.*?)?v?(\d[\d.]*)\.tar\.gz%@PACKAGE at -$1.tar.gz%" \
- https://github.com/vcflib/vcflib/tags .*/v(\d[-\d.rc]+)@ANY_VERSION@
+ https://github.com/vcflib/vcflib/tags .*/v(\d[-\d.rc]+)\.tar\.gz
=====================================
guix.scm
=====================================
@@ -91,7 +91,7 @@
("smithwaterman" ,smithwaterman)
("tabixpp" ,tabixpp)
("xz" ,xz)
- ; ("zig" ,zig) ;; note we use zig-0.9.1
+ ("zig" ,zig) ;; note we use zig-0.9.1
("zlib" ,zlib)))
(native-inputs
`(("pkg-config" ,pkg-config)))
=====================================
src/vcfcreatemulti.cpp
=====================================
@@ -19,12 +19,15 @@ extern "C" {
using namespace std;
using namespace vcflib;
+#ifdef NO_ZIG
+bool nextGen = false;
+#else
bool nextGen = true;
+#endif
+
bool quiet = false;
off_t file_size = -1;
-// extern "C" void *zig_create_multi_allelic(Variant *retvar, Variant *varlist[], long size);
-
void printSummary(char** argv) {
cerr << R"(
Usage: vcfcreatemulti [options] [file]
@@ -112,6 +115,7 @@ Variant createMultiallelic_legacy(vector<Variant>& vars) {
return mvar;
}
+#ifndef NO_ZIG
Variant createMultiallelic_zig(vector<Variant>& vars) {
if (vars.size() == 1) {
@@ -127,12 +131,15 @@ Variant createMultiallelic_zig(vector<Variant>& vars) {
return *mvar;
}
+#endif
// This function is called for every line/variant in the VCF file
Variant createMultiallelic(vector<Variant>& vars) {
+ #ifndef NO_ZIG
if (nextGen)
return createMultiallelic_zig(vars);
else
+ #endif
return createMultiallelic_legacy(vars);
}
@@ -211,7 +218,7 @@ int main(int argc, char** argv) {
size_t prev_pos = 0;
if (!quiet)
- cerr << "vcfcreatemulti processing..." << endl;
+ cerr << "vcfcreatemulti " << VCFLIB_VERSION << " processing..." << endl;
while (variantFile.getNextVariant(var)) {
@@ -276,10 +283,12 @@ int main(int argc, char** argv) {
cout << result << endl;
}
+ #ifndef NO_ZIG
if (nextGen) {
zig_display_warnings();
zig_cleanup();
}
+ #endif
if (!quiet) cerr << endl;
=====================================
src/vcfwave.cpp
=====================================
@@ -223,7 +223,7 @@ int main(int argc, char** argv) {
uint64_t start = get_timestamp();
if (!quiet)
- cerr << "vcfwave processing..." << endl;
+ cerr << "vcfwave " << VCFLIB_VERSION << " processing..." << endl;
while (variantFile.getNextVariant(var)) {
amount = (double)variantFile.file_pos()/(double)file_size;
=====================================
src/zig/build.zig
=====================================
@@ -19,7 +19,7 @@ pub fn build(b: *std.build.Builder) void {
const main_tests = b.addTest("vcf.zig");
main_tests.setBuildMode(mode);
- main_tests.addLibraryPath("../../build");
+ // main_tests.addLibraryPath("../../build");
// main_tests.addObjectFile("../../build/libvcflib.a");
const test_step = b.step("test", "Run library tests");
=====================================
src/zig/vcf.zig
=====================================
@@ -56,7 +56,6 @@ var gpa = std.heap.GeneralPurposeAllocator(.{}){};
const allocator = gpa.allocator();
var warnings = std.StringHashMap(bool).init(allocator); // Note: not thread safe
-// var warnings = ArrayList([] const u8).init(allocator); // NOTE: not thread safe
pub fn warning(str: [] const u8) !void {
try warnings.put(str,true);
@@ -308,29 +307,27 @@ export fn zig_create_multi_allelic(variant: ?*anyopaque, varlist: [*c]?* anyopaq
// Get genotypes and update mvar
var genotypes = samples.reduce_renumber_genotypes(Variant,vs) catch unreachable;
- // mvar.set_samples(genotypes.c_samples());
+ defer genotypes.s_samples.deinit();
+
mvar.set_samples(genotypes.s_samples);
var ninfo = ArrayList([] const u8).init(allocator);
- ninfo.deinit();
+ defer ninfo.deinit();
+
if (genotypes.g_err != samples.VcfSampleError.None) {
ninfo.append("ALTPROBLEM") catch unreachable;
mvar.set_info("MULTI",ninfo);
}
- genotypes.s_samples.deinit();
-
return mvar.v;
}
/// The C++ code should call this to cleanup
export fn zig_cleanup() void {
- // p("zig cleaning up!",.{});
- // std.debug.assert(!gpa.deinit());
warnings.deinit();
- const leaked = gpa.deinit();
- if (leaked) p("MEM FAIL",.{});
+ // ---- Not cleaning up the GPA unless we are debugging
+ // std.debug.assert(!gpa.deinit());
}
=====================================
test/doc/performance.md
=====================================
@@ -54,3 +54,7 @@ tux02:~/tmp/vcflib/build$ /usr/bin/time -v ./vcfwave ../samples/chr18.grch38.vcf
hmm. Not so exciting. Looks like we need to parallelize at a different level. Memory used was about 6Gb, and that is not bad.
The good news is that the output is identical.
+
+## vcfcreatemulti
+
+This program is actually slower than vcfwave.
View it on GitLab: https://salsa.debian.org/med-team/libvcflib/-/compare/2fee087a72e17bb749d866d9a5d75c03ddc59c47...7b2b4502f03df84844d9f4d94e7933215641c23c
--
View it on GitLab: https://salsa.debian.org/med-team/libvcflib/-/compare/2fee087a72e17bb749d866d9a5d75c03ddc59c47...7b2b4502f03df84844d9f4d94e7933215641c23c
You're receiving this email because of your account on salsa.debian.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-med-commit/attachments/20230205/2f26f3ba/attachment-0001.htm>
More information about the debian-med-commit
mailing list