[gentoo-commits] repo/proj/guru:dev commit in: sys-fs/dwarfs/files/, sys-fs/dwarfs/

[Denis Reva]

commit:     253b9fd4d798d3f817b4d9f735cff0580202375a
Author:     Denis Reva <denis7774 <AT> gmail <DOT> com>
AuthorDate: Tue Feb 28 12:53:52 2023 +0000
Commit:     Denis Reva <denis7774 <AT> gmail <DOT> com>
CommitDate: Tue Feb 28 12:57:56 2023 +0000
URL:        https://gitweb.gentoo.org/repo/proj/guru.git/commit/?id=253b9fd4

dwarfs: unorphaned, work in progress

Signed-off-by: Denis Reva <denis7774 <AT> gmail.com>

 sys-fs/dwarfs/Manifest                             |   1 +
 sys-fs/dwarfs/dwarfs-0.7.0_rc4.ebuild              | 103 +++++++++++++
 .../dwarfs/files/dwarfs-0.7.0_rc4-unbundle.patch   | 161 +++++++++++++++++++++
 sys-fs/dwarfs/metadata.xml                         |  24 +++
 4 files changed, 289 insertions(+)

diff --git a/sys-fs/dwarfs/Manifest b/sys-fs/dwarfs/Manifest
new file mode 100644
index 000000000..f6a1d88fe
--- /dev/null
+++ b/sys-fs/dwarfs/Manifest
@@ -0,0 +1 @@
+DIST dwarfs-0.7.0-RC4.tar.xz 14425176 BLAKE2B fdb2852da81f8d567c19dc83e3f7dd74e0de35e56579d50b6ce9ee6adda2d540bbbc7d5655a499f19ee340a2f904bfe9b2b977b9539e8e99dc9e8667b9b7435e SHA512 1e0d603507f93b6bcb79050877de0ac64eed931b192b43466d3b5e1ee55a759261ad937d33f24d145a61e8b3d431405a35bbd8ff95ead10ccdabe14ba4d840a5

diff --git a/sys-fs/dwarfs/dwarfs-0.7.0_rc4.ebuild b/sys-fs/dwarfs/dwarfs-0.7.0_rc4.ebuild
new file mode 100644
index 000000000..d1a86ea98
--- /dev/null
+++ b/sys-fs/dwarfs/dwarfs-0.7.0_rc4.ebuild
@@ -0,0 +1,103 @@
+# Copyright 1999-2021 Gentoo Authors
+# Distributed under the terms of the GNU General Public License v2
+
+EAPI=7
+
+PYTHON_COMPAT=( python3_{10..11} )
+
+inherit check-reqs cmake flag-o-matic python-single-r1
+
+MY_P="${P/_rc/-RC}"
+
+DESCRIPTION="A fast very high compression read-only FUSE file system"
+HOMEPAGE="https://github.com/mhx/dwarfs"
+SRC_URI="https://github.com/mhx/dwarfs/releases/download/v0.7.0-RC4/dwarfs-0.7.0-RC4.tar.xz" #TODO: change to ${PV}
+
+LICENSE="GPL-3"
+SLOT="0"
+KEYWORDS=""
+IUSE="python +jemalloc test man"
+S="${WORKDIR}/${MY_P}"
+
+RDEPEND="
+	${PYTHON_DEPS}
+	app-arch/libarchive
+	app-arch/lz4
+	app-arch/snappy
+	app-arch/xz-utils
+	app-arch/zstd
+	dev-cpp/fbthrift:=
+	dev-cpp/folly:=
+	dev-cpp/gflags
+	dev-cpp/glog[gflags]
+	dev-cpp/parallel-hashmap:=
+	dev-cpp/sparsehash
+	dev-libs/boost[context,threads(+),python?]
+	dev-libs/double-conversion
+	dev-libs/fsst:=
+	dev-libs/libevent
+	dev-libs/libfmt
+	dev-libs/xxhash
+	sys-fs/fuse:3
+	sys-libs/binutils-libs
+	sys-libs/libunwind
+	sys-libs/zlib
+
+	jemalloc? ( >=dev-libs/jemalloc-5.3.0-r1 )
+"
+DEPEND="
+	${RDEPEND}
+	sys-devel/flex
+	!sys-fs/dwarfs-bin
+"
+BDEPEND="
+	man? ( || ( app-text/ronn app-text/ronn-ng ) )
+	sys-devel/bison
+	virtual/pkgconfig
+
+	test? ( dev-cpp/gtest )
+"
+
+DOCS=( "README.md" "CHANGES.md" "TODO" )
+RESTRICT="!test? ( test )"
+REQUIRED_USE="python? ( ${PYTHON_REQUIRED_USE} )"
+PATCHES=( "${FILESDIR}/${P}-unbundle.patch" )
+
+CHECKREQS_DISK_BUILD="1300M"
+CMAKE_IN_SOURCE_BUILD=1
+CMAKE_WARN_UNUSED_CLI=0
+
+src_prepare(){
+	rm -r fsst zstd fbthrift/* folly xxHash parallel-hashmap || die
+	cmake_src_prepare
+	sed "s/DESTINATION lib/DESTINATION $(get_libdir)/" -i CMakeLists.txt || die
+}
+
+src_configure(){
+	append-cxxflags "-I/usr/include"
+
+	mycmakeargs=(
+		-DUSE_JEMALLOC=$(usex jemalloc ON OFF)
+		-DWITH_PYTHON=$(usex python ON OFF)
+		-DWITH_TESTS=$(usex test ON OFF)
+		-WITH_MAN_PAGES=$(usex man ON OFF)
+		-DPREFER_SYSTEM_ZSTD=1
+		-DPREFER_SYSTEM_XXHASH=1
+		-DPREFER_SYSTEM_GTEST=1
+		-DWITH_LEGACY_FUSE=0
+	)
+	use python && mycmakeargs+=( "-DWITH_PYTHON_VERSION=${EPYTHON#python}" )
+	cmake_src_configure
+}
+
+src_install(){
+	cmake_src_install
+	dolib.so libdwarfs.so
+}
+
+pkg_postinst(){
+	elog "You may find more information in the"
+	elog "${HOMEPAGE}"
+	elog "About creating: ${HOMEPAGE}/blob/main/doc/mkdwarfs.md"
+	elog "About mounting: ${HOMEPAGE}/blob/main/doc/dwarfs.md"
+}

diff --git a/sys-fs/dwarfs/files/dwarfs-0.7.0_rc4-unbundle.patch b/sys-fs/dwarfs/files/dwarfs-0.7.0_rc4-unbundle.patch
new file mode 100644
index 000000000..101af0dc6
--- /dev/null
+++ b/sys-fs/dwarfs/files/dwarfs-0.7.0_rc4-unbundle.patch
@@ -0,0 +1,161 @@
+--- a/CMakeLists.txt
++++ b/CMakeLists.txt
+@@ -151,8 +151,6 @@
+     ON
+     CACHE BOOL "only build thrift compiler")
+ 
+-add_subdirectory(folly EXCLUDE_FROM_ALL)
+-add_subdirectory(fbthrift EXCLUDE_FROM_ALL)
+ if(NOT (ZSTD_FOUND AND PREFER_SYSTEM_ZSTD))
+   add_subdirectory(zstd/build/cmake EXCLUDE_FROM_ALL)
+ endif()
+@@ -383,23 +381,6 @@
+ 
+ list(
+   APPEND
+-  FROZEN_THRIFT_SRC
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_data.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_data.cpp
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_types.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_types.tcc
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_types.cpp
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_types_custom_protocol.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_constants.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_constants.cpp
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_metadata.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_metadata.cpp
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_visitation.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_for_each_field.h
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_visit_union.h)
+-
+-list(
+-  APPEND
+   METADATA_THRIFT_SRC
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs/gen-cpp2/metadata_constants.cpp
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs/gen-cpp2/metadata_constants.h
+@@ -418,35 +399,20 @@
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs/gen-cpp2/metadata_visitation.h)
+ 
+ add_custom_command(
+-  OUTPUT ${FROZEN_THRIFT_SRC}
+-  COMMAND mkdir -p ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift
+-  COMMAND
+-    cp ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/thrift/frozen.thrift
+-    ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/
+-  COMMAND cd ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift &&
+-          ${CMAKE_CURRENT_BINARY_DIR}/bin/thrift1 --gen mstch_cpp2 frozen.thrift
+-  DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/bin/thrift1
+-          ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/thrift/frozen.thrift)
+-
+-add_custom_command(
+   OUTPUT ${METADATA_THRIFT_SRC}
+   COMMAND mkdir -p ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs
+   COMMAND cp ${CMAKE_CURRENT_SOURCE_DIR}/thrift/metadata.thrift
+           thrift/dwarfs/metadata.thrift
+   COMMAND
+     cd ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs &&
+-    ${CMAKE_CURRENT_BINARY_DIR}/bin/thrift1 --gen mstch_cpp2:frozen2
++    thrift1 --gen mstch_cpp2:frozen2
+     metadata.thrift
+-  DEPENDS ${CMAKE_CURRENT_BINARY_DIR}/bin/thrift1
+-          ${CMAKE_CURRENT_SOURCE_DIR}/thrift/metadata.thrift)
++  DEPENDS ${CMAKE_CURRENT_SOURCE_DIR}/thrift/metadata.thrift)
+ 
+ list(
+   APPEND
+   INCLUDE_DIRS
+-  ${CMAKE_CURRENT_BINARY_DIR}/folly
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift
+-  ${CMAKE_CURRENT_SOURCE_DIR}/folly
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift
+   ${CMAKE_CURRENT_BINARY_DIR})
+ 
+ if(NOT (ZSTD_FOUND AND PREFER_SYSTEM_ZSTD))
+@@ -472,43 +438,24 @@
+ endif()
+ 
+ add_library(
+-  thrift_light
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/FieldRef.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/protocol/CompactProtocol.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/protocol/BinaryProtocol.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/protocol/DebugProtocol.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/protocol/JSONProtocolCommon.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/protocol/JSONProtocol.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp/protocol/TProtocolException.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp/util/VarintUtils.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/gen/module_types_cpp.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/frozen/Frozen.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/frozen/FrozenUtil.cpp
+-  ${CMAKE_CURRENT_SOURCE_DIR}/fbthrift/thrift/lib/cpp2/frozen/schema/MemorySchema.cpp
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_data.cpp
+-  ${CMAKE_CURRENT_BINARY_DIR}/thrift/lib/thrift/gen-cpp2/frozen_types.cpp)
+-
+-set_property(TARGET thrift_light PROPERTY CXX_STANDARD 17)
+-
+-target_include_directories(thrift_light PRIVATE ${INCLUDE_DIRS})
+-
+-add_library(
+   metadata_thrift
++  STATIC
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs/gen-cpp2/metadata_layouts.cpp
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs/gen-cpp2/metadata_types.cpp
+   ${CMAKE_CURRENT_BINARY_DIR}/thrift/dwarfs/gen-cpp2/metadata_data.cpp)
+ 
+ set_property(TARGET metadata_thrift PROPERTY CXX_STANDARD 17)
+-
++set_property(TARGET metadata_thrift PROPERTY POSITION_INDEPENDENT_CODE ON)
++set_property(TARGET metadata_thrift PROPERTY CXX_VISIBILITY_PRESET hidden)
++set_property(TARGET metadata_thrift PROPERTY VISIBILITY_INLINES_HIDDEN 1)
+ target_include_directories(metadata_thrift PRIVATE ${INCLUDE_DIRS})
+-
+-add_dependencies(metadata_thrift thrift_light)
++target_link_libraries(metadata_thrift fmt glog folly)
+ 
+ foreach(tgt dwarfs ${BINARY_TARGETS})
+   target_include_directories(
+     ${tgt} SYSTEM
+     PRIVATE ${Boost_INCLUDE_DIRS} ${Python3_INCLUDE_DIRS} ${INCLUDE_DIRS}
+-            ${CMAKE_CURRENT_SOURCE_DIR}/parallel-hashmap)
++            )
+ 
+   target_include_directories(${tgt} PUBLIC ${CMAKE_CURRENT_SOURCE_DIR}/include)
+ 
+@@ -565,23 +512,11 @@
+   endif()
+ endforeach()
+ 
+-# not sure why exactly, copied from fsst/CMakeLists.txt
+-if(CMAKE_BUILD_TYPE STREQUAL Release)
+-  set_source_files_properties(fsst/fsst_avx512.cpp PROPERTIES COMPILE_FLAGS -O1)
+-endif()
+-
+-add_library(
+-  fsst
+-  fsst/libfsst.cpp fsst/fsst_avx512.cpp fsst/fsst_avx512_unroll1.inc
+-  fsst/fsst_avx512_unroll2.inc fsst/fsst_avx512_unroll3.inc
+-  fsst/fsst_avx512_unroll4.inc)
+-
+-target_include_directories(dwarfs PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}/fsst)
+-
+ target_link_libraries(
+   dwarfs
++  thriftprotocol
++  thriftfrozen2
+   metadata_thrift
+-  thrift_light
+   folly
+   fsst
+   ${Boost_LIBRARIES}
+@@ -612,11 +547,6 @@
+   if(USE_JEMALLOC)
+     target_link_libraries(${tgt} ${Jemalloc_LIBRARIES})
+   endif()
+-  if(DWARFS_USE_EXCEPTION_TRACER)
+-    target_link_libraries(
+-      ${tgt} -Wl,--whole-archive folly_exception_tracer_base
+-      folly_exception_tracer -Wl,--no-whole-archive)
+-  endif()
+ endforeach()
+ 
+ if(STATIC_BUILD_DO_NOT_USE)

diff --git a/sys-fs/dwarfs/metadata.xml b/sys-fs/dwarfs/metadata.xml
new file mode 100644
index 000000000..7d948c451
--- /dev/null
+++ b/sys-fs/dwarfs/metadata.xml
@@ -0,0 +1,24 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE pkgmetadata SYSTEM 'http://www.gentoo.org/dtd/metadata.dtd'>
+<pkgmetadata>
+	<maintainer type="person">
+		<email>denis7774@×××××.com</email>
+		<name>Denis Reva</name>
+		<description>rarogcmex</description>
+	</maintainer>
+	<longdescription lang="en">
+		DwarFS is a read-only file system with a focus on achieving very high compression ratios in particular for very redundant data.
+		This probably doesn't sound very exciting, because if it's redundant, it should compress well. However, I found that other read-only, compressed file systems don't do a very good job at making use of this redundancy. See here for a comparison with other compressed file systems.
+		DwarFS also doesn't compromise on speed and for my use cases I've found it to be on par with or perform better than SquashFS. For my primary use case, DwarFS compression is an order of magnitude better than SquashFS compression, it's 4 times faster to build the file system, it's typically faster to access files on DwarFS and it uses less CPU resources.
+		Distinct features of DwarFS are:
+		* Clustering of files by similarity using a similarity hash function. This makes it easier to exploit the redundancy across file boundaries.
+		* Segmentation analysis across file system blocks in order to reduce the size of the uncompressed file system. This saves memory when using the compressed file system and thus potentially allows for higher cache hit rates as more data can be kept in the cache.
+		* Highly multi-threaded implementation. Both the file system creation tool as well as the FUSE driver are able to make good use of the many cores of your system.
+		* Optional experimental Lua support to provide custom filtering and ordering functionality.
+	</longdescription>
+	<upstream>
+		<bugs-to>https://github.com/mhx/dwarfs/issues</bugs-to>
+		<remote-id type="github">mhx/dwarfs</remote-id>
+	</upstream>
+
+</pkgmetadata>