We recently changed how we manage chroots, requiring the user (of the
chroot) to initialize it before doing stuff like installing packages.
There are however still certain edgecases in pmbootstrap where this
doesn't get done (for example qemu/run.py would attempt to install the
kernel package for the device, but we don't initialize the rootfs chroot
in QEMU since that doesn't make sense to do). Check for and catch this
situation explicitly so we don't ruin someones day, but still be loud
about it so we can hopefully catch the remaining instances of this.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Fix for:
$ pmbootstrap pull
[21:09:43] aports_upstream (branch: 3.20-stable): not on one of the official branches (master), skipping pull!
Replace the get_branches_official() function that according to git log
was never able to tell if an aports branch was official. The new
function branch_looks_official() can do that by just checking if the
branch follows the typical naming pattern.
The comment I had put into get_branches_official earlier was not true
anymore, by now this did not only get used by "pmbootstrap status", but
also by "pmbootstrap pull".
Let pmbootstrap display 3.0.0_alpha on crash, so it's clear when people
are using the new codebase.
Before you report this error, ensure that pmbootstrap is up to date.
Find the latest version here: https://gitlab.com/postmarketOS/pmbootstrap/-/tags
Your version: 3.0.0_alpha
This turns out to have been wrong, since we need to give apk the package
name AND the apk file, it installs the apk and then uninstalls it but
leaves the package installed by name.
Don't try and install random packages in the rootfs chroot when starting
QEMU. The chroot is not initialised and this just adds latency.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
One flaw of pathlib is that Path("/tmp/") / "/some/relative/path"
results in Path("/some/relative/path")....
Fix the places in the codebase where we get this wrong.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Turns out this is necessary, *sigh*. Since we could have one locally
built package depend on another locally built package.
So yeah, re-implement this, but in a slightly lazier / hopefully faster
way.
Bah
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
It was only introduced in 3.11. Instead just define __str__() manually
for the Enums where it's needed.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Sometimes the local binary repo will have no packages but still an
APKINDEX. It will be empty in this case. Handle this gracefully.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
This is a potential security risk. Tighten things up so we only allow
local checksums to mismatch but remote ones must be correct.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
This avoids copying local build artifacts and other things, if they're
in .gitignore then abuild shouldn't need them anyway.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Add a new flag --image which can be used to mount the rootfs generated
with "pmbootstrap install".
For now this is quite limited in scope. But it's enough to allow for
building a package, updating it in the QEMU image, and then booting it.
The major "gotcha" with this is that the QEMU uses the kernel and
initramfs from the device chroot unless you run it with --efi.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
We can fairly easily patch abuild by appending anything we want to the
APKBUILD file after copying it to the chroot.
Leverage this to override the verify() function so that it doesn't die
when checksums don't match.
Instead let's build the package anyway, but set a flag and print a
warning with instructions on how to generate the checksums.
We should continue to require APKBUILDs in pmaports to have valid
checksums.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
This got broken during rework. Fix it up, make sure we filter out
packages from to_add that are specified in to_add_local.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
It turns out that apk with the --root argument will still treat local
repositories relative to the host. This makes sense considering the
intended usecase, however as a result the recent change to use
apk.static for all apk commands inside the target root will break the
use of local packages.
To fix this, adjust how we populate /etc/apk/repositories, changing the
default to no longer include the local user repo. Instead we pass
--repository to apk.static with the host path to the repo.
The only time we want to add the user repo to /etc/apk/repositories
inside the chroot is when the user enters the chroot, so a special case
is added to helpers/frontend.py for the chroot command.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
It was the case that packages were being queued in the wrong order,
since the dependency resolver queues the package before descending into
it's dependencies. This was a bit of a goof since the top level package
was always added last. If we add it first then we can just reverse the
queue and now everything is fine...
Additionally, the logic on when a dependency should be built was a bit
wonky. A case is added for when a dependency exists only in the source
repo and the requisite package isn't marked for build. The solver will
now build the dependency regardless. This is surely an edgecase but
somehow I ran into it, I suspect due to a bug elsewhere...
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
Add a function that just detaches all loop devices that are backed by a
file inside the pmbootstrap work dir.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
* Fix details_to_stdout which was broken due to a previous commit.
* Improve support for custom colors with "@COLOR@" (e.g. "@BLUE@"), stop
it from being printed unformatted to the logfile.
* Rework how the logfd is handled so that --details-to-stdout won't
disable all color output.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
This option will make pmbootstrap automatically zap chroots that are
initialized for the wrong channel, making it much faster to switch
between edge and systemd.
Signed-off-by: Caleb Connolly <caleb@postmarketos.org>
