This has been clearly lacking while making the previous commit
but the implementation isn't that clear so let it be a separate
step.

The problem requiring the change in subsequent processors
is that these relied upon "@arch" as a flag to be inspected,
and "pkg@!arch1,arch2" on arch2 needs to take out *all* of that
fragment *including* arch1 mention as well.

Part of the cause is difference in handling: "positive" multi-match
would explode its "client" line into multiple lines to filter down
the pipeline, while "negative" multi-match *has* to keep that line
on a similarly single line (otherwise we'd end up with N-1 of those
slipping past the filter for particular architecture thus defeating
the whole purpose of "negative" matching semantics):

$ echo 'pkg@!E2K,mipsel,riscv64' |
  sed -r  ':loop; s/^((([^@]+@!)[^,]+)+),([a-zA-Z0-9_]+)/\1@!\4/; t loop'
pkg@!E2K@!mipsel@!riscv64

I've tried my best to test this specific change but it still might
introduce a regression in some corner case; feel free to report;
looks like there's a space for improvement in m-p's automated
tests department as well.

So now we can do:

  pkg@!ARCHES1,ARCHES2,arch3,arch4

and have pkg excluded on arches mentioned; the previous approach
could only offer explicit whitelists (not that it was entirely
wrong but then again, we have both ExclusiveArch and ExcludeArch
rpmtags in our spec files).

This has been inspired by a few commits that cared
for package availability reasons on a particular
architecture; the problem at hand is that pkglists
might need to include groups of packages that are
(un)available on groups of arches, and tackling that
with plain pkg@arch just results in combinatorial
explosion of that matrix.

Arches are handled one-by-one with a few hardcoded
macro substitutions.

Exploding a "pkg@arch1,arch2" string into a set of:

pkg@arch1
pkg@arch2

with subsequent archdep pruning would do the trick;
so here's another sed oneliner that does just that:

$ echo 'pkg@X86,ARM,ppc64le' |
  sed -r ':loop; s/^((([^@]+@)[^,]+)+),([^,]+)/\1\n\3\4/; t loop'
pkg@X86
pkg@ARM
pkg@ppc64le

See-also: 9601a9e7
See-also: 5581dc91
See-also: http://stackoverflow.com/a/55781741/561921

The parameter determines use of QEMU or not, if the target architecture
does not correspond to the host architecture.
By default, the parameter is on (Value 1).
For architectures that do not support QEMU (e2k), the option is turned off.

See also commit: 61f66b08

This allows you to reset the BRANCH variable by specifying the
'BRANCH=' argument on the command line.

The commit 99695713 used this variable
even though it didn't exist yet. A good reason to add this variable.

Aptbox is not initialized, so build time is reduced.

The @META suffix is used to expand the metapackage into a list.
apt asks for the dependencies of such a package and adds them to the
package list after this metapackage.

Now it cannot be set from the config ~/mkimage/profiles.mk

It was fixed in the commit:
187d0a5d

NO_SYMLINK disables the creation of symlinks for the builded image.

.fop.xconf needed for building pdf.

The BRANCH parameter indicates that the target is a starterkit and
not a regular build for Sisyphus.

Previously, you had to specify two parameters MKIMAGE_PREFIX and
GLOBAL_PREFIX with the same value - the path to mkimage. And this
behavior has not been documented. The GLOBAL_PREFIX variable is
defined in mkimage in config.mk and rules.mk.

The parameter is used to save the intermediate tarball used to
build the virtual machine image.

At the moment, the ability to select the kernel with which to boot is
implemented only for grub (grub-pc, grub-efi, ieee1275boot).
note that renamed STAGE1_KFLAVOUR to STAGE1_KFLAVOURS, as multiple
kernels can now be added.

Those willing to reuse archdep-filter elsewhere
will likely have to figure it out from Makefiles
anyway, so probably not worth overdocumenting...

It was misleading already, and riscv64 is going to join the gang
(but not there yet) following aarch64, mipsel, ppc64le and more
e2k* subarches...

doc/archdep.txt was the reason to look closer,
and some other files needed a bit of facelift
by now too (would be better to do that in sync).

Use it in the image filename, so we'll know what version of a distro
it is.

See commit 2d257a42
for details, this just brings those to pkg.in/profiles/
(along with the generic archdep support, of course).

pauli@ has proposed slightly different setup to what is provided
by livecd-qemu-arch; but the use case might be different either.
At least provide the link to those interested.

The former ("proper 32-bit x86 package form") has been suggested
by zerg@ quite some time ago but the desired interface wasn't clear
at the moment IIRC; a quiet morning helped me realize that

  ICAClient-preinstall@IA32

is rather more readable than

  ICAClient-preinstall@i586 i586-ICAClient-preinstall@x86_64

so here's the (trivial) implementation; and I actually needed
the latter, @X86 ("any-x86") to mark x86-only packages so

  xorg-drv-intel@X86

is now equivalent to

  xorg-drv-intel@i586 xorg-drv-intel@x86_64

Another minor facelift:
- s/ALT Linux/ALT/gc (where applicable);
- s/7.0/8.0/g;
- 2017;
- e2k;
- links.

It's at least removing the very obvious user->root
attack through (maliciously) modifying bin/tar2fs
and waiting for it to be run; if mkimage-profiles
is installed system-wide as a package, the script
from /usr/share/mkimage-profiles will be tried so
those willing to allow vm/* build to themselves
can provide for a passwordless sudo (as described
in doc/vm.txt) to run a root-only writable script,
not user-writable.

Still not perfect but a step away from the abyss.

This performs distclean upon successful image build.
Suggested-by: Andrey Cherepanov <cas@altlinux.org>

It wasn't clear enough what the whole tracing affair is for;
thanks ykos@ for spelling it aloud.

A fellow got confused by the original header somehow,
let's spell this a lil' bit differently.

Thanks cas@ and Ivan Khakhaev for feedback!
My docs are clearly not the best in town,
please help bring those into shape :-)

This initial page might lack a lot, don't be shy to blame me!
(thanks ldv@ for having skimmed it through though)

Merge commits aren't going to be merged upstream
for these break all sorts of git tools including
log -p, blame, bisect, etc (or make their use less
convenient).

Sometimes we tend to "just" translate the code to human
which doesn't really help; what usually helps is noting
down what the code author intended to implement or fix,
and what was the case for that.

The new archdep part has been initially included into
"The Basics" chapter which it doesn't really belong to,
let's move it into Addendum.

NB: I'd better try building a package and skimming over
    at least the index upon having modified docs next time.