_precommand lets zsh complete other commands and their arguments
e.g. it can complete grep with "systemd-cat gr" and complete grep options
with "systemd-cat grep -"
* fix error
* remove options that are no longer supported
* add missing options
* stop completion if an option `--help` or `--version` is supplied
[[[
zjs: a note for the reader:
zshcompsys(1) in the section about optspecs in _arguments says:
> Each of the forms above may be preceded by a list in parentheses of option names and argument num‐
> bers. If the given option is on the command line, the options and arguments indicated in parentheses
> will not be offered. For example, ‘(-two -three 1)-one:...' completes the option ‘-one'; if this ap‐
> pears on the command line, the options -two and -three and the first ordinary argument will not be
> completed after it. ‘(-foo):...' specifies an ordinary argument completion; -foo will not be com‐
> pleted if that argument is already present.
>
> Other items may appear in the list of excluded options to indicate various other items that should
> not be applied when the current specification is matched: a single star (\*) for the rest arguments
> (i.e. a specification of the form ‘\*:...'); a colon (:) for all normal (non-option-) arguments; and a
> hyphen (-) for all options. For example, if ‘(\*)' appears before an option and the option appears on
> the command line, the list of remaining arguments (those shown in the above table beginning with
> ‘\*:') will not be completed.
The intended effect of the change is to remove irrelevant completion matches from the completion.
tl;dr: (- : ) prevents further completion
]]]
Show hidden images in the completion results, but only if the current
word starts with ".", such that
- `machinectl clone <Tab>` will only offer non-hidden images, but
- `machinectl clone .<Tab>` will offer both hidden and non-hidden images
bootctl is rather useful to have, even if on a system without UEFI,
as it has a number of verbs that are unrelated to UEFI (e.g kernel-identify),
and more importantly, it supports --root to operate on directory trees
(which could be intended to be deployed on UEFI) so let's make sure we
always build it.
I'm not quite sure what the original intent of this line was, but it
doesn't work in the one call-site the "required" argument is actually
used. The "writable" flag was indexed as a scalar leaving only the
"e" to compare against.
Instead, let's just sort the parsed flags and compare the whole thing.
Also substitute "required" as a pattern, so that pattern based
comparisons may be supported.
The systemctl completion previously made use of PREFIX as a pattern
argument to list-unit-files and list-units. This had the problem of
erroneously filtering the results that were stored in the cache, and
erroneously filtering results that might have been requested according
to the users configuration (e.g. _correct completer, certain
matcher-lists or tag-orders, etc.).
Unfortunately, the runtime of list-unit-files increases when no pattern
argument is provided, and systemctl show, used to filter those units,
can become unacceptably slow when provided with too many units to
describe.
Let's re-introduce the pattern argument to list-unit-files and
list-units where necessary in order to alleviate these bottlenecks
without poisining the cache. A 'use-pattern' style is introduced that
may be used to disable this behavior if it is undesired. We can still
expect that certain completions, like `systemctl start <TAB>` will be
slow, like before. To fix this we will need systemd to learn a more
efficient way of filtering the units than parsing systemctl show.
The systemctl invocations used for these completions match the ones used
for the _sys_really_all_units parameter, so we should really just use
the cached parameter rather than recomputing the result.
Template names can be learned from the filesystem, so there isn't a need
to parse the output of systemctl list-unit-files in this case. This
should accelerate the completion of some verbs like enable.
The existing caching policy isn't very sensible for this cache. We could
write a different policy, but I don't think there is much value in
caching these values, as in my experience the command used to generate
them is quick.
The existing caching policy was completely bogus.
In the first stanza, despite the comment, the pattern given would
consider the cache invalid if it was more than 1 hour old.
The second stanza was also incorrect, since the output of `systemctl
--all` is not unit file paths, but unit names. When they were being
tested against the cachefile mtime, the test would always fail becuase
of the nonexistant file (hopefully).
In fact it's not very useful to test if the unit files have newer mtime
in this case anyway, since we are only caching their names. Also,
`systemctl --all` is an unfortunately slow operation to be used in
testing for the cache validity — we want this operation to at least be
faster than rebuilding the cache.
I've rewritten this stanza with my best guess at its original intent. It
now checks against the mtime of the parent directories in the search
path, which should be updated and cause the cache to rebuild when we
add, remove, or rename any unit files.
Rebuilding whenever the cached parameter is not set forces each new
shell to rebuild the cache, which often defeates the purpose of caching
in the first place.
This used to work correctly, before the change was reverted in
e09d0d46c2. In fact it is important to specify the manager explicity
in the completion because the argument is reused in the caching
policies. An empty argument here caused the completion to create
separate caches with and without the --system parameter. We can simplify
the given pattern a little here too.
