as fedora uses pkgconf to help locate the thrift files during
compilation instead of a Thrift.cmake file. I have added this logic in
such a way that the existing logic should not be affected. Therefore
enabling the use of system thrift even without pkgconf.
* chore(build): refactor main CMakeLists.txt
* chore(build): fix boost version for packages, set clang-20 only for amd and arm
* chore(build): boost 4 sm
* chore(build): boost dep for rowgroup
* chore(build): toolset for boost
* chore(build): suppress clang warnings for boost
* chore(ci, build): use ASAN for unittest on ubuntu 24.04 only, added custom cmake flag option for bootstrap, custom params for new and existing pipelines
* chore(build): sort bootstrap flags
* chore(CI): remove publish pkg step, adding clickable link instead to publish steps, fix customenv
The commit introduces a new CMake configuration option, WITH_THRIFT,
which accepts "auto", "system", or "bundled". In "auto" mode (the
default), the build system attempts to use the system-installed Thrift
library and falls back to the bundled version if the system library is
not available. Setting WITH_THRIFT to "system" enforces the use of the
system Thrift, causing the configuration to fail if it isn't found,
while "bundled" forces the use of the bundled version.
The change mainly useful for downstream maintainers as it gives them
flexibility over dependency management.
Downstream-issue: https://bugs.gentoo.org/949680
* feat(BRM,tools): couple utilities to watch/operate shared memory locks and extent map
* feat(BRM,tools): merged two utilities and added some extra dbbuilder output in case of upgrade
* fix(dbbuilder): extra output to log upgrade detection.
[add] dynamic year to cmapi CmakeLists
[add] package description to cmapi CmakeLists
[add] use MDB ES repo cmake dir
[fix] cmapi package filenaming
[fix] .drone.jsonnet to do cmapi build step depending on clonning mdb step
[add] cmapi_misc.cmake file
[add] lsb-release package to build deps in drone.jsonnet file
[fix] wrong debian and rpm requires curl placement
[fix] SET(CPACK_PACKAGE_FILE_NAME
[add] redhat-lsb-core package for RHEL systems
[add] SERVER_DIR argument for cmapi CmakeLists.txt
[add] get_linux_lsb_release_information() for both deb and rpm builds
It is currently not possible to do a full MariaDB build without touching the
source tree, i.e. such that the the source tree could exist on a read-only
filesystem. This columnstore build file currently creates temporary files in
the source tree to evaluate how the operating system handles them, so it is
one blocker to these isolated builds.
This patch adjusts these tests to create the temporary files in the output
directory, rather than in the source tree.
Signed-off-by: Trevor Gross <tmgross@umich.edu>
according to https://bugzilla.redhat.com/show_bug.cgi?id=1806545
GCC's PCH mechanism is fundamentally incompatible with the security
hardening requirements for RHEL (PIE in particular). We are currently
evaluating the best way to present a more friendly diagnostic for this
issue.
and also
we don't turn on hardening the compiler and its binaries in DTS 9.
[fix] columnstore_version.cmake setting extra variables for CMAPI versioning
[fix] CMAPI CmakeLists.txt using columnstore_version.cmake file
[fix] CMAPI VERSION file to template file due to using it inside CMAPI applicatio
n
* more build dependencies
* fix for cmake < 3.11
It cannot do ADD_LIBRARY(... ALIAS ...) on IMPORTED targets
* another fix for cmake 3.10.2
It doesn't know about CMAKE_CXX_STANDARD=20,
let's add the correct flag manually
* gcc 8 on aarch64
utils/common/simd_arm.h:241:16: error: need ‘typename’ before ‘simd::TypeToVecWrapperType<T>::WrapperType’ because ‘simd::TypeToVecWrapperType<T>’ is a dependent scope
* build boost during build phase, not during configure
* add dependency for generated header files errorids.h messageids.h
see 7e868bc588
* set explicit dependencies on external_boost for #include's
* clang-14 compatibility fix
On AMD64 machines, the fpu is 80 bits. The unused bits must be masked for memcmp to work properly. For other archetectures, we don't want to mask those bits.
* Adds CompressInterfaceLZ4 which uses LZ4 API for compress/uncompress.
* Adds CMake machinery to search LZ4 on running host.
* All methods which use static data and do not modify any internal data - become `static`,
so we can use them without creation of the specific object. This is possible, because
the header specification has not been modified. We still use 2 sections in header, first
one with file meta data, the second one with pointers for compressed chunks.
* Methods `compress`, `uncompress`, `maxCompressedSize`, `getUncompressedSize` - become
pure virtual, so we can override them for the other compression algos.
* Adds method `getChunkMagicNumber`, so we can verify chunk magic number
for each compression algo.
* Renames "s/IDBCompressInterface/CompressInterface/g" according to requirement.
This commit introduces DataConvert UTs.
DataConvert::decimalToString now can negative values.
Next version for Row::toString(), applyMapping UT checks.
Row:equals() is now wide-DECIMAL aware.
Binary NULL magic now consists of a series of BINARYEMPTYROW-s + BINARYNULL
in the end.
ByteStream now has hexbyte alias.
Added ColumnCommand::getEmptyRowValue to support 16 byte EMPTY values.
