Re: [lmi] Getting rid of the rest of Boost

[Vadim Zeitlin]

On Mon, 4 Oct 2021 23:07:25 +0000 Greg Chicares <gchicares@sbcglobal.net> wrote:

GC> On 10/3/21 10:13 PM, Vadim Zeitlin wrote:
GC> [...]
GC> >  the step for building Boost.Regex has somehow
GC> > remained in the CI workflow file, but I've now removed it from there
GC> > myself, so this is not a problem neither.
GC> 
GC> Okay, thanks. After your commit e44b02881f7fc7d, only one hunk of your
GC> original PR remains in 'ci.yml':
GC> 
GC> -       if: matrix.autotools != true
GC> +       if: matrix.mingw != true && matrix.autotools != true
GC>         run: |
GC>           # Somehow these files are executable and this triggers complaints
GC>           # from check_script.sh, so avoid them by fixing permissions.
GC>           chmod -x .git/info/exclude .git/description
GC> -         make $coefficiency LMI_COMPILER=gcc 
LMI_TRIPLET=x86_64-pc-linux-gnu check_concinnity
GC> +         make $coefficiency check_concinnity
GC> 
GC> and that's only because I forgot to apply it. But maybe it's no longer
GC> wanted;

 No, it isn't, the version in the latest master, i.e. 3d8b0069f (Work
around wrong directory permissions in MSW cross-build job, 2021-10-03)
already contains these changes.

GC> > I've also fixed the problem with
GC> > the bad ownership of /opt/lmi/local directory, which somehow ended up 
being
GC> > owned by root, breaking the MSW cross-build using the official makefiles 
as
GC> > it couldn't create gcc-x86_64-pc-linux-gnu subdirectory there. To be
GC> > honest, I didn't spend time on finding out how exactly did it happen and
GC> > just corrected the problem in the workflow file, but this is somewhat
GC> > surprising because /opt/lmi is owned by the non-root user when it is
GC> > created, so I could look into this if you'd like.
GC> 
GC> It should be created in 'GNUmakefile' if it doesn't already exist
GC> (the /local.*dir/ variables all default to /opt/lmi/local/whatever):
GC> 
GC> .PHONY: $(build_dir)
GC> $(build_dir): $(gpl_files)
GC>     +@[ -d $@                 ] || $(MKDIR) --parents $@
GC>     +@[ -d $(localbindir)     ] || $(MKDIR) --parents $(localbindir)
GC>     +@[ -d $(locallibdir)     ] || $(MKDIR) --parents $(locallibdir)
GC>     +@[ -d $(localincludedir) ] || $(MKDIR) --parents $(localincludedir)
GC> 
GC> After rebuilding everything, I have:
GC> 
GC>   /opt/lmi/src/lmi[0]$ls -ld /opt/lmi/local 
GC>   drwxrwsr-x 7 greg lmi 4096 Oct  2 22:24 /opt/lmi/local
GC> 
GC> as desired. But I didn't remove /opt/lmi/* first; let me try doing
GC> that and then rebuilding again.

 FWIW I couldn't reproduce it locally neither, but without the commit
mentioned above, the CI build failed because the "Check concinnity and
physical closure" step failed to create the directory. And debugging this
(in the CI environment, i.e. by temporarily adding "ls -l" commands to this
step) showed that it failed because /opt/lmi/local belonged to root,
somehow. But even though I don't know why/how did it happen, I am still
almost sure that it's specific to the CI environment, which is rather
special, as I run the commands inside a Debian Docker container using
/usr/bin/setpriv to ensure that they run with the correct user -- because
by default everything would run as root, and I didn't want that. Maybe my
setpriv options are not completely correct, but I remember spending quite
some time on getting them at least approximately right, so I don't really
want to change them again, and the workaround is not too bad, so I'm sorely
tempted to just leave this mystery alone, for once.


GC> >  But for now I'd mostly like to know if you think it would be useful to 
get
GC> > rid of the rest of Boost
GC> 
GC> Absolutely.

 Excellent!

GC> >  Please let me know if you'd like me to make the changes suggested above,
GC> > test them and submit a patch with them and, also, removing everything
GC> > Boost-related from lmi entirely.
GC> 
GC> Let me do that, because I need something simple to work on this week.

 Thanks in advance and please let me know if I can help in any way!
VZ

pgp_YICRJXNpC.pgp
Description: PGP signature