Disabled packages add unnecessary complexity and introduce no big gain.
Furthermore they might and do hide if package is fixed by upstream.

There are two reason why package can be disabled. Either id does not
compile because it is broken and in that case we should build it in hope
that it will be fixed. Or it might be because it just does not apply on
given board or architecture and in that case that package should be
marked as incompatible with that architecture in package itself not
somewhere outside.

Upstream renamed it to config.buildinfo.

They also introduced new file feeds.buildinfo. We could use that instead
of our git-hash generated file.

Configuration override options was misleading. In reality those are
options you most probably don't want to mess too much with so let's call
them rather advanced options.

This also moves PUBLISH_BRANCH in example configuration file to
section: don't mess with. This is because now we provide it from
defaults and user just should not mess with it unless he know what he is
doing.

OpenWRT commit to be used is specified in feeds.conf file with rest of
feeds now so it makes not sense to overwrite it. This is just confusing
without any real gain in here. You can always modify feeds.conf file to
achieve the same.

Thanks to midnightmagic, who pointed out it on IRC channel #turris on Freenode

Karel Koci authored 5 years ago and Josef Schlehofer committed 5 years ago

Parameter -e is now obsolete and by default, we want to build everything.
Removed it from turris-build.conf.example

Because of that we or users don't need to select packages in make
menuconfig which we or they want to build. Now each package will be selected
but not compiled which depends on user choice.

Also, compile_pkgs shouldn't be run under root or with sudo. Do it as
normal user.

This is more appropriate naming for Turris 1.x routers because otherwise
just plain turris is confusing.

This should make much more clearer what are usable and what are not
configuration options.

When branch is provided to feed with `;` we should also replace that
with exact hash.

This fixes problem with ignored packages that were introduced by patch.
What this does is that it installs packages after patching is done. It
also first updates indexes again before installation.

Because this feature was originally split between two function but
patch_feeds had to be inserted to the middle of get_feeds I decided to
drop both and used upper repatch_feeds instead. This means that we are
no longer able to get and patch feeds separately but I think that this
separation was almost useless.

This should consistently expand path to compile_pkgs script no matter if
symbolic link for used or not.

Master id deployed to dragons and there will be no point of deploying it
to sloths. It is question how to make this change more general across
forks and merges but let's ignore that for now.

Previous implementation was just wrong lets face it.

This reverts commit 4ac86b03.

On server is currently used invalid configuration. This commit can be
reverted after merge to master and configuration file update.

Until now it was called somewhat not systematically before every build
and after OpenWRT patching.

This is code that was originally only in Jenkins but it makes sense to
have this prepare stage externally documented and git tracked so I moved
it here.

Empty make argument causes make to build default target. This commit
removes any empty arguments from make calls.

Build is not split to multiple steps and having one central statistics
does not make sense. Also this stat is tracked by external build system
tool (Jenkins).