There are multiple changes to updater's lists. In most cases it is about
moving packages around and grouping them together to simplify future
maintenance.

Significant change is usage of conditional requests. Updater now
installs Foris plugins not because some package list was selected but
because some package was installed. This makes it so users running
minimal setup (without web interface) can use package lists and still
have full system with it.

One important change to note here is that SSH client was moved from
minimal base list to terminal-apps. The reason is that it is not
essentially required for router's functionality or access. It is more
about secondary usage from CLI so it should have been part of
terminal-apps all time along.

This has to be string otherwise it won't do replan. In the end it is not
that of a big issue. Subsequent execution later on removes this fix
package.

This has to be string otherwise it won't do replan. In the end it is not
that of a big issue. Subsequent execution later on removes this fix
package.

Co-authored-by: Jan Pavlinec <jan.pavlinec@nic.cz>
Co-authored-by: Vojtech Myslivec <vojtech.myslivec@nic.cz>

The hash character denotes hash over branch. Hash is prefixed by it. The
code was trying to remove if as suffic instead of prefix so it ended up
checking out invalid reference with hash char included.

We should verify that arguments are valid numbers to prevent other
nothing-saying errors.
It also improves little bit usage to print only binary name over full
path.

This simple typo caused forice-make to run only in fallback mode. That
is always building in single thread with verbose output.

In OpenWrt master, there were dropped SSL variants of transmission.
We backported change to have life easier and luckily for us,
Transmission list works even there is daemon-openssl.

This just ensures that transmission-daemon will be installed, if we will
remove some day fix package for Turris OS 5.x.

Fixes: 37e45862
("patches/backport/transmission: add patch to remove package variants")

I don't think it is a good idea to push all tags, which you are having.
Sometimes you can have local tags, which you don't want to push but with
suggested commands they will be pushed.

Let's advise to push tag, which was created.

In preparation for new reboot fix testing remove these hacks. They do
not work anyway.

This was a typo. This is not part of the package name.

We are removing cznic-cacert-bundle and this fix package is part of that
cleanup.

This is better approach not only because we can potentially upstream
this but also because it covers all packages and breaks potentially if
new variant is added.

Marek Behun authored 4 years ago and Karel Koci committed 4 years ago

This adds support for Turris SFP modules for kernel 4.14.

The idea is to have script that does the same thing as we do for
releases but for not yet released builds.
Usage is mostly for automated builds based on our repository.
Successful build on our servers somewhat ensures also successful build
anywhere else. Building latest successful version takes users to success
more likely than just building tip of git branch.

There is problem with repo_checkout as using it with newer version of
turris-build does not aways provide what you want. You should
effectivelly match turris-build version with target version when using
repo_checkout but that is not easily possible. It is easier to just drop
it and provide utility script that checkouts turris-build repository to
appropriate version and modified feeds.conf file. This script is going
to be added by subsequent commit.

This also optimizes corner case when we are cloning from mirror. We do
not have to fetch everything when we are cloning from mirror as git can
fetch just single commit. The problem is that this is not available when
commit is being fetched from servers. This change just tweaks condition
so that if mirror is used then hash fetch is used and not general fetch.

We in the end have only one place where we use those variables. It is
cleaner to have them specified directly by appropriate tool.

Running make multiple times for builds with last debug fallback run is
handy not even for calls available trough compile_pkgs. For example
building single selected package is not possible using compile_pkgs and
that script is not intended to even do it. The clean solution because of
that is to just simply extract logic for forced make run to separate
script that can be used without compile_pkgs,