We pass this check now.

They should definitely not be None here, thus assert it so we don't need
to keep checking for None later.

It doesn't really make sense to check the blacklist against None, so
just skip the value if it's None. Appeases mypy.

Otherwise, if pmb.parse.deviceinfo().cgpt_kpart is an empty string, for
example, we end up passing an empty string to get_partition_layout()
instead of False.

pmaports_cfg.get() can return other types, too. As such, we need to
assert that it's a list to appease mypy.

Helps mypy understand the code better and also makes it easier to read
as a human in my opinion.

I think the "these types make no sense" comment were about
output_return_buffer, which defaulted to False but was to be set to
list[bytes] if used. I changed it so that it defaults to None, which
probably is more conventional. Other than that I didn't notice anything
weird about the types here.

Also check for None as necessary to appease mypy.

img_path_2nd is a Path, not a str, so we can't just rely on string
concatenation here.

img_path_2nd is a Path as that's what create_second_storage() returns
since 198f302a.

This could be done better with a real class instead of a TypedDict, but
it's better than a regular dict.

Useful for annotating places where a parsed APKBUILD is expected.

This makes the behaviour match the docstring which specifies that this
function returns integers.

This is equivalent to the previous definition, but more concise.

We don't want an actual Arch object in the environment object.

Can't construct a path by using the / operator on strings, so use the
Path constructor instead.

This is no longer used as of 1299c2ef,
so just remove this dead code.

If flavor is None here, something has gone wrong and the line after will
fail. Explicitly assert that it has the right value to appease mypy.

Reusing this variable might lead to confusing bugs later if this code is
changed in the future, and also leads to mypy being confused about the
type of size_mb as it then gets assigned both str and int values.

At some point, this function was updated to use the Arch type instead of
plain strings, but the return type was left omitted and the docstring
outdated.

No parameter called "now" exists for this function, so don't document
it.

This is already the default: https://mypy.readthedocs.io/en/stable/config_file.html#confval-color_output

And remove --check-untyped-defs from mypy.sh script to avoid
duplication.

Flush the progress bar displayed by the function before logging a
warning message for 404 not found responses from the server. Without
flushing, this message gets lost.

In addition to the warning, display a NOTE and ERROR at the end to
explain what went wrong:

  [20:16:50] Update package index for x86_64 (4 file(s))
  [20:16:51] WARNING: file not found: http://mirror.postmarketos.org/postmarketos_get_404_test/edge/main/x86_64/APKINDEX.tar.gz
  [20:16:51] NOTE: check the [mirrors] section in 'pmbootstrap config'
  [20:16:51] ERROR: getting APKINDEX from binary package mirror failed!

Without handling this properly, pmbootstrap would fail slightly later
with a much less useful error:

  [15:14:28] ERROR: expected str, bytes or os.PathLike object, not NoneType

Fixes: pmbootstrap issue 2424

Don't update the git URL again when coming from 7-2.x where this step
was already done. This didn't fail, but was potentially confusing for
the user since the git URLs were already upgraded from 7 -> 7-2x.

Casey authored 5 months ago and Oliver Smith committed 5 months ago

Some packages in aports have their own subpackages in their
dependencies, for some reason. This causes our resolver to assume there
is a cyclical dependency.

Adjust the resolver to safely ignore these dependencies, since they'll
be built as a part of the package that depends on them.

For example, modemmanager depends on libmm-glib which is a subpackage of
modemmanager. Currently to build modemmanager in pmaports it would be
necessary to manually modify the APKBUILD after copying it from aports.
This change avoids that.

Signed-off-by: Caleb Connolly <caleb@postmarketos.org>

args.autoinstall attribute might not exist, use getattr like for others,
and add a fixme about the default value...

Fixes: #2476


Signed-off-by: Caleb Connolly <caleb@postmarketos.org>

Newbyte authored 5 months ago and Oliver Smith committed 5 months ago

All call sites that use this have been removed. This was only introduced
as a stopgap until we could refactor more code to not use args
everywhere, and that has now been achived. As such, remove this so it
doesn't get used in more locations.

Casey authored 5 months ago and Oliver Smith committed 5 months ago

See
#2473



Signed-off-by: Caleb Connolly <caleb@postmarketos.org>

Casey authored 5 months ago and Oliver Smith committed 5 months ago

Add a basic test to ensure that the cache does get hit as we expect.
This doesn't test for cache conflicts.

Signed-off-by: Caleb Connolly <caleb@postmarketos.org>

Casey authored 5 months ago and Oliver Smith committed 5 months ago

Use a different way of keying the apkindex path cache.

Signed-off-by: Caleb Connolly <caleb@postmarketos.org>

Casey authored 5 months ago and Oliver Smith committed 5 months ago

It may be desirable to test older versions of packages, allow them to be
built when --force is specified and print a warning that they won't be
installed automatically.

Signed-off-by: Caleb Connolly <caleb@postmarketos.org>

Vikram Alagh authored 5 months ago and Oliver Smith committed 5 months ago

Seems to be ChromeOS specific:

> this happens every time on my machine. I am on ChromeOS 129.0.6668.112
> using a Debian bookworm chroot through Crouton. I am not sure why
> this is happening nor which layer is responsible for this. I deleted
> the chroot and created another one but the problem persists.

Commit msg tweaked by Oliver.

Provides kpartx.

Looks like an unintended leftover from debugging.

---

This was needlessly reverted by
1ec55fc1. While revreting
0975d064 was good, this commit did not
cause any harm.

And fix type errors.

---

This was needlessly reverted by
1ec55fc1. While revreting
0975d064 was good, this commit did not
cause any harm.

Running these inside the chroot is tricky, and we ended up reverting
previous commits that attempted to make them run there (see
1ec55fc1). As such, just accept defeat
and ensure that the host has these.

Also see postmarketOS/pmbootstrap#2465

We fully expect this to happen if a user hasn't configured their host
system properly. As such, don't treat it like a programming error.

Luca Weiss authored 5 months ago and Casey committed 5 months ago

We now run the job in chroot, so remove the unnecessary check. If
sphinx-build is missing for some reason, it will fail a bit further in
the script anyways.