On Mon, Jun 29, 2020 at 06:27:40PM +0200, Andrea Bolognani wrote:
On Mon, 2020-06-29 at 15:32 +0100, Daniel P. Berrangé wrote:
On Mon, Jun 29, 2020 at 03:58:43PM +0200, Andrea Bolognani wrote:
+for infile in templates/* +do + outfile="${infile##*/}.j2" + host="${outfile%%.*}" + + eval $("$LCITOOL" dockerfile "libvirt-$host" libvirt --variables) + + sed -e "s|[@]PKGS@|$PKGS|g" \ + -e "s|[@]CROSS_PKGS@|$CROSS_PKGS|g" \ + -e "s|[@]PYPI_PKGS@|$PYPI_PKGS|g" \ + -e "s|[@]CPAN_PKGS@|$CPAN_PKGS|g" \ + -e "s|[@]MAKE@|$MAKE|g" \ + -e "s|[@]NINJA@|$NINJA|g" \ + -e "s|[@]PYTHON@|$PYTHON|g" \ + <"$infile" >"$outfile" +done
I feel like this should really be reduced to just:
$LCITOOL cirrusci libvirt-$host libvirt > $outfile
+ build_script: + - mkdir build + - cd build + - ../autogen.sh --prefix=$(pwd)/install-root + - $MAKE -j3 + - $MAKE -j3 install + - $MAKE -j3 dist
This part can then be stored in ci/cirrus/build.yml since it is common to freebsd & macos.
So now in gitlab-ci.yml we can just concatenate the two into a temp file:
cat ci/cirrus/$NAME.yml ci/cirrus/build.yml > ci/cirrus/$NAME.yml.j2 cirrus-run ci/cirrus/$NAME.yml.j2
This avoids needing to store the same information in git twice.
Yeah, I don't like having to store the information twice in git, but I'm also not a fan of having an lcitool subcommand that is tightly coupled with the details of a specific CI service rather than dealing with the more general problem of managing build environments.
Hmm, I kind of feel that's exactly the job of lcitool. Generate whatever files/resources we need for integrating with whatever CI infra we're choosing to use. Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|