Hi Peter, What version of make do you use?
Regards, Matthias Am 30.12.22 um 20:44 schrieb Peter Kovacs: > Hello all, > > I managed to Build AOO trunk on Ubuntu 22.04. The build has been rough > and with a lot of Issues. > > I would like to write the issues here for documentation. > > 1, 2 and 3 are Issues on the old dependencies we build. All other > Issues was on our Code as far as I understand. > > 1) CoinMP > > 1.1) dmake builds with C17 standard > > I have seen a lot of warnings that register is not allowed to use with > C++17 Standard. > > (I worked around from building the Library directly and then use dmake > in order to continue.) > > 1.2) Macro Variable not set. > > In the file CoinFinite.cpp the following code has not worked through: > > [...] > > #ifdef HAVE_CFLOAT > # include <cfloat> > > [...] > > I was not able to figure out how to fix this. I did just add the > include line again without the ifdef condition. > > 2)SQLite > > I had to do 3 Adjustments. > > 2.1) in one line is follort by an assert statement, which triggers a > error in g++. > > workaround: Added a linebreak > > 2.2) A warning that a Pointer is returned. > > See following discussion: > > https://sqlite.org/forum/info/eadcdb7058803d20 > > In fact I added following Change: > > https://www.sqlite.org/src/vdiff?branch=cgo-warning-workaround&dc=20 > > Just be carefull with line 143. Just add pAllocated = > > 3) Boost > > I had to change the file > > main/solver/450/unxlngx6/inc/boost/mpl/assert.hpp > > I ended up changing and commenting out: > > //template< typename Pred >(Pred::assert_arg( void (*)(Pred), typename > assert_arg_pred<Pred>::type )); > //template< typename Pred >(boost::mpl::not_<Pred>::assert_not_arg( > void (*)(Pred), typename assert_arg_pred_not<Pred>::type )); > > I am not sure if this is a local issue, or if this is in general. The > lines look odd. (it is line 186 if you want to have a look.) > > > 4) osl mutex > > main/sal/osl/unx/mutex.c > > I commented out the following code: > #if defined LINUX /* bad hack / > int pthread_mutexattr_setkind_np(pthread_mutexattr_t *, int); > #define pthread_mutexattr_settype pthread_mutexattr_setkind_np > #define PTHREAD_MUTEX_RECURSIVE PTHREAD_MUTEX_RECURSIVE_NP > #endif > > > Not sure what the background is, but it compiled afterwards. > > 5) CUI > > main/cui/Library_cui.mk > > I had this issue before. And I am not able to figure out what the > issue is exactly. Maybe my version number is wrong. > > $(eval $(call gb_Library_add_defs,cui, > -DBUILD_VER_STRING="$(BUILD_VER_STRING)" )) > > So I skipped the line for now: > > My line: > > --with-build-version="aoo4ever-$(date +"%Y-%m-%d %H:%M") - $(uname > -sm)\n ${GitHash}" > > I will have a look if I can change something in order to improve this. > > > I had no other issues with the build. Especially the Issue Matthias > has reported did not appear. > > I will keep looking on the Issues Matthias has reported. I assume this > is more a Windows Issue. > > I hope this information is useful to some. > > Have a good new years end party. :) > > All the best > > Peter > > >
smime.p7s
Description: S/MIME Cryptographic Signature
