After `journalctl -D /var/log/journal` "--directory", "--file",
"--machine" and "--root" should not be available for completion, because
they are exclusive. But multiple `--file` arguments are allowed.
In 68c4f6d the following was added:
local -a _modes; _modes=("--user" "--system")
local _sys_service_mgr=${${words:*_modes}[(R)(${(j.|.)_modes})]:---system}
With the following comment:
> If neither are on the line, --system is set; for system services to be
> completed.
But it does not work as documented:
% _modes=(--user --system)
% words=()
% echo ${${words:*_modes}[(R)(${(j.|.)_modes})]:---system}
However, it should not use `--system` in that case anyway, so this patch
removes the part that should cause a default to be used and adds some
comments.
This only completes fields from `journalctl --user` in _journal_fields when `--user`
is used.
It also changes $_sys_service_mgr to include both `--system` and `--user`,
because `journalctl` behaves different from `systemctl` in this regard.
No attempt is made to filter out invalid combinations, e.g. when using both
`--directory` and `--system` (see https://github.com/systemd/systemd/issues/3949).
This uses the same mechanism from _systemctl to inject `--user` into the
`journalctrl -F _EXE` call to list executables.
Before this patch the "commands" section would list executables from
system units always.
Use `$_sys_service_mgr` to handle `--user`, so that `systemctl --user
stop` will correctly filter the active (user) units. Before this patch,
only user units that also exist as system units and are stoppable there
would be listed.
This new output mode formats all timestamps using the usual format_timestamp()
call we use pretty much everywhere else. Timestamps formatted this way are some
ways more useful than traditional syslog timestamps as they include weekday,
month and timezone information, while not being much longer. They are also not
locale-dependent. The primary advantage however is that they may be passed
directly to journalctl's --since= and --until= switches as soon as #3869 is
merged.
While we are at it, let's also add "short-unix" to shell completion.
This patch renames Read{Write,Only}Directories= and InaccessibleDirectories=
to Read{Write,Only}Paths= and InaccessiblePaths=, previous names are kept
as aliases but they are not advertised in the documentation.
Renamed variables:
`read_write_dirs` --> `read_write_paths`
`read_only_dirs` --> `read_only_paths`
`inaccessible_dirs` --> `inaccessible_paths`
Support for net_cls.class_id through the NetClass= configuration directive
has been added in v227 in preparation for a per-unit packet filter mechanism.
However, it turns out the kernel people have decided to deprecate the net_cls
and net_prio controllers in v2. Tejun provides a comprehensive justification
for this in his commit, which has landed during the merge window for kernel
v4.5:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=bd1060a1d671
As we're aiming for full support for the v2 cgroup hierarchy, we can no
longer support this feature. Userspace tool such as nftables are moving over
to setting rules that are specific to the full cgroup path of a task, which
obsoletes these controllers anyway.
This commit removes support for tweaking details in the net_cls controller,
but keeps the NetClass= directive around for legacy compatibility reasons.
But also keep the old name as (undocumented) compatibility around.
The reload-or-try-restart was documented to be a NOP if the unit is not running, since the previous commits this is
also implemented. The old name suggests that the "try" logic only applies to restarting. Fix this, by moving the "try-"
to the front, to indicate that the whole option is a NOP if the service isn't running.
Escape colons and backslashes in unit names.
This gives correct completions for units with names like
systemd-backlight@backlight:acpi_video0.service
and
systemd-fsck@dev-disk-by\x2duuid-...
The current code is not compatible with current dkr protocols anyway,
and dkr has a different focus ("microservices") than nspawn anyway
("whole machine containers"), hence drop support for it, we cannot
reasonably keep this up to date, and it creates the impression we'd
actually care for the microservices usecase.
This directive allows passing environment variables from the system
manager to spawned services. Variables in the system manager can be set
inside a container by passing `--set-env=...` options to systemd-spawn.
Tested with an on-disk test.service unit. Tested using multiple variable
names on a single line, with an empty setting to clear the current list
of variables, with non-existing variables.
Tested using `systemd-run -p PassEnvironment=VARNAME` to confirm it
works with transient units.
Confirmed that `systemctl show` will display the PassEnvironment
settings.
Checked that man pages are generated correctly.
No regressions in `make check`.
Snapshots were never useful or used for anything. Many systemd
developers that I spoke to at systemd.conf2015, didn't even know they
existed, so it is fairly safe to assume that this type can be deleted
without harm.
The fundamental problem with snapshots is that the state of the system
is dynamic, devices come and go, users log in and out, timers fire...
and restoring all units to some state from the past would "undo"
those changes, which isn't really possible.
Tested by creating a snapshot, running the new binary, and checking
that the transition did not cause errors, and the snapshot is gone,
and snapshots cannot be created anymore.
New systemctl says:
Unknown operation snapshot.
Old systemctl says:
Failed to create snapshot: Support for snapshots has been removed.
IgnoreOnSnaphost settings are warned about and ignored:
Support for option IgnoreOnSnapshot= has been removed and it is ignored
http://lists.freedesktop.org/archives/systemd-devel/2015-November/034872.html
'set-property' was missing from the list of known command.
Also a list of unit names will be proposed as next argument.
However no support on property names is provided since it would
require a hard coded list of them.
awk is an external program, and it is better to stick to shell built-ins.
Also, even with external awk, sort -u is redundant, because the shell does
this on its own.
"machinectl list" only lists running machines while many of the MACHINES
commands use names of images; both running and non-running.
List machines from both "list" and "list-images" and use sort -u to
avoid duplicates.
_loginctl: respects the verbose style. which allows a user to get
the pre d5df0d950f behavior of not showing a description for sessions
and users, by default they aren't shown.
zstyle ':completion:*' verbose true
or
zstyle ':completion:*:loginctl*:*' verbose true # or similar
Will show the descriptions.
zstyle ':completion:*' verbose true
and
zstyle ':completion:*:loginctl*:*' verbose false # or similar
Won't show descriptions for loginctl only
_systemd: complete pids for systemd-notify's --pid option.
display a message of the expected argument for other options.
_systemd-inhibit: complete block & delay for --mode
display a message of the expected argument for --who/--why
filenames will be completed for --image/-i/--bind/--bind-ro/--tmpfs
network interfaces for --network-(interface|macvlan|ipvlan|bridge)
users for --user/-u, yes & no for --register, x86 * x86-64 for
--personality
display a message of the expected argument for --machine/-M/--uuid
--slice/-S/--port/-p/--selinux-*/-Z/-L/--setenv
Allow completing commands(and their options) of the host system for COMMAND
1) the iterator `fun' has an local scope. after running the completer,
it will no longer be defined.
2) use _describe instead of calling compadd. Using compadd without
calling _description or something similar before, restricts the
user's ability to customize what is presented to them.
zstyle ':completion:*' format 'Completing %d'
- now displays an header showing what is being completed.
zstyle ':completion::complete:loginctl-*::users' users user1 user2
- allows the user to manually specify which users is offered
zstyle :completion::complete:loginctl-kill-user:\* \
ignored-patterns '(100<0-4>|user1)'
- selectively ignore some users when completing loginctl kill-user
<tab>
Sessions, UIDs now have descriptions when selecting them.
3) removed the call to _loginctl_all_seats in _loginctl_attach(), since
_loginctl_seats calls it a second time, right before adding matches.
There isn't a noticeable difference doing this.
Optimize _filter_units_by_property by calling `systemctl` only once with
a list of units, and not once per unit.
I could not reproduce the "Unknown unit" error mentioned in a FIXME,
which might have made this necessary previously.
using _wanted instead of calling compadd directly. this allows the user to customize
possible matches.
An example being, grouping units by type:
autoload -Uz compinit; compinit
zstyle ':completion:*' menu select
zstyle ':completion:*' group-name ''
zstyle ':completion:*' format 'Completing %d'
zstyle -e ':completion:*:*:systemctl-(((re|)en|dis)able|(*re|)start|reload*):*' \
tag-order 'local type; for type in service template target socket;
reply+=( systemd-units:-${type}:${type} ); reply=( "$reply systemd-units:-misc:misc" )'
zstyle ':completion:*:systemd-units-template' ignored-patterns '^*@'
zstyle ':completion:*:systemd-units-target' ignored-patterns '^*.target'
zstyle ':completion:*:systemd-units-socket' ignored-patterns '^*.socket'
zstyle ':completion:*:systemd-units-service' ignored-patterns '^*.service'
zstyle ':completion:*:systemd-units-misc' ignored-patterns '*(@|.(service|socket|target))'
also, <poke> http://lists.freedesktop.org/archives/systemd-devel/2015-May/032012.html
By the time __systemctl is called, --user/--system are shifted out of
`words' by _arguments. This patch queries the array sooner.
In the case that both --user and --system are on the line when compsys runs,
_sys_service_mgr is set to the latter. Which is seemingly how systemctl behaves.
If neither are on the line, --system is set; for system services to be completed.
compadd's -a option treats non-option arguments as arrays. So
$(_systemctl_get_template_names) expands to some words that aren't
legal array names. Even if there were, they would be empty; thus adding
nothing.
deduplicated a few functions too.
This makes all functions that rely on _filter_units_by_property() (like
_systemctl_{stop,kill,try_restart}) work with unit names that contain backslash
escaped sequences (like automount units with spaces that are escaped to
"\x20").
*Autocompletion for dirs, doesn't leave until you press space.
*Added tmpfs, volatile and network-macvlan options.
I tried with the SELinux options with seinfo(setools-console), but too
messy to get it right. Even Daniel Walsh haven't done it yet. :)
We really don't want to get lost in adding fridge, car, plane, drone, or
whatever else, hence add a generic term "embedded" cover all the cases
where the computer is just part of something bigger, and not at the
focus of things.
I tried to use 'systemctl --all list-units' to filter unit files, but
this always filters out unit files which are not loaded. We want to complete
systemctl start with those units too, so this approach is not going to work.
New version is rather slow, but hopefully correct.
Templates can be [re]enabled, on their own if the have DefaultInstance set,
and with an instance suffix in all cases. Propose just the template name
ending in @, to underline the instance suffix may have to be appended.
Likewise for start/restart.
This means that sometimes superflous units that one will not really
want to operate on will be proposed, but this seems better than
proposing a very incomplete set of names.
https://bugs.freedesktop.org/show_bug.cgi?id=66912
removed pointless index sort of bootids.
use `compadd -a' to add each array, instead of expanding possibly hundreds of words needlessly.
optional completion of -b
Since c6a373a263, we might encounter unit templates via the
'list-units' verb. These aren't restartable (and we throw errors), so
make sure they're filtered out of the completion options.
fixes downstream bug: https://bugs.archlinux.org/task/41719
This fixes the issue noted by Zbigniew in most cases.
if a unit's name is enclosed in single quotes completion still
will not happen after the first `\'.
https://bugs.freedesktop.org/show_bug.cgi?id=78388
I think that it is better to return good results slightly more slowly,
than partial quickly. Also reading from disk seems fast enough. Even
the delay on first try with completely cold cache is acceptable.
This is just for bash, 'cause zsh was already doing this.
https://bugzilla.redhat.com/show_bug.cgi?id=790768
Since this is a shortopt with an optional argument, assume the user
knows what they're doing. The longopts --boot and --this-boot will
continue to offer boot IDs as completions.
-- fix grammar and reword some descriptions for clarity
-- add a useful description of what --follow does
-- fix the description for --after-cursor
-- properly introduce the FSS acronym for "Forward Secure Sealing" in
both sections
-- clarify the --disk-usage command
[zj: perform similar changes to zsh completions]
squash! journalctl: fix several issues in --help message text
Instead of having two different listings of machines, use an autoloaded
function that can be used by other shell completions in the future. It
will also allow editing a single file to change the way machinectl and
systemd-run completion for machines.
Suggested by David Wilkins <dwilkins@maths.tcd.ie> in
https://bugzilla.redhat.com/show_bug.cgi?id=967521:
> [Specific boot ID is a] bit of a palaver to obtain. I consulted the
> verbose dump of the journal to discover the _BOOT_ID for the
> timestamp, and then generated the journal dump for that boot using
> journalctl _BOOT_ID=foo -o short-monotonic.
_SYSTEMD_USER_UNIT in the --user-unit flag argument should instead be
USER_UNIT. It should also have an optional `=` between the flag and the
argument.
Things like -n to specify the lines to show with systemctl and
journalctl accepts syntax like:
journalctl -n4
systemctl -n14
Previously, typing `-nXX <tab>` where XX is a number, zsh would try to
complete an integer. Now it will see the XX and use the _journalctl_none
completion. This is also how any of the single letter options that take
arguments work as well.
Some of the options in systemd can take multiple arguments, such as
systemctl's --type option. Previously, you would only be able to
complete a single type after the -t, but now zsh will continue to
complete the types, separating them by commas.
systemd-inhibit's --what command has colon (:), and that has been taken
into account.
_hosts_or_user_at_host was used by 6 different completions, and
previously was in all 6 of those files. I moved it out to its own file,
_sd_hosts_or_user_at_host. This will be autoloaded for use in other
completion functions. It also allows external completions to use this
function by simply calling _sd_hosts_or_user_at_host as in the systemd
completions.
Splitting things unnecessarily at newlines causes tab completion to take
an extremely long time. Also add a note saying that caching is not good
for journalctl's completion.
Moved zsh shell completion to shell-completion/zsh/_systemd for
automake's sake. Also allow users to specify where the files should go
with::
./configure --with-zshcompletiondir=/path/to/some/where
and by default going to `$datadir/zsh/site-functions`
The AA is unnecessary and only adds needless complexity. Replace it
with a case statement instead of repeatedly calling __contains_word to
overglorify string equalities.
- scope the iterator var
- use the correct, quoted, non-expansion prone positional parameter
notation
- prevent expansion on RHS of comparison
- remove unneeded explicit returns.
This really should be defined only once...
Hi,
I redid the boot ID look up to use enumerate_unique.
This is quite fast if the cache is warm but painfully slow if
it isn't. It has a slight chance of returning the wrong order if
realtime clock jumps around.
This one has to do n searches for every boot ID there is plus
a sort, so it depends heavily on cache hotness. This is in contrast
to the other way of look-up through filtering by a MESSAGE_ID,
which only needs about 1 seek + whatever amount of relative IDs
you want to walk.
I also have a linked-list + (in-place) mergesort version of this
patch, which has pretty much the same runtime. But since this one
is using libc sorting and armortized allocation, I prefer this
one.
To summarize: The MESSAGE_ID way is a *lot* faster but can be
incomplete due to rotation, while the enumerate+sort will find
every boot ID out there but will be painfully slow for large
journals and cold caches.
You choose :P
Jan
--user basically gives messages from your own systemd --user services.
--system basically gives messages from PID 1, kernel, and --system
services. Those two options are not exahustive, because a priviledged
user might be able to see messages from other users, and they will not
be shown with either or both of those flags.
Instead of completing the whole line, which doesn't work, only complete
the pid, but still show the whole line so the user can see which command
was which.
Users can also let the parameter expansion sort the completion by date
instead of by pid, by setting
zstyle ':completion:*:*:systemd-coredumpctl:*' sort no
so that the zshcompsys doesn't sort the _describe function for only
systemd-coredumpctl.
This mirrors --property, and is generally useful.
New functionality is used in bash completion.
In case of zsh completion, new functionality is less useful
because of caching. Nevertheless, zsh completion for restart
is made to behave more-or-less the same as bash completion.
At least sockets can be restarted.
It is faster to use a bash built-in, then to invoke an external
program. The problem of unit names starting with a dash is solved
by prepending a space. Spaces are ignored anyway.
For zsh, replace echo "$unit", which is vulnerable to dashes,
with echo " $unit".
Split the large bash completion script into separate, smaller files each
named after the binary it is used for and move the files to
/usr/share/bash-completion/completions. This way the completions can be
loaded on demand and we only install the completions for the tools we
actually build. The old path /etc/bash_completion.d/ is deprecated and
will disappear in the future.
The approach taken is different between the two:
- since there are many files in /usr, but messages appear
only for a tiny subset, the completion is performed
only for stuff shown by journalctl -F _EXE. This makes
sense because the list is already in proper form.
- since it is hard to convert _KERNEL_DEVICE to device
file name, simply all files in /dev/ are used as possible
completions.
Unfortunately zsh completion requires more work and is not
covered by this commit.
Adds is-failed to join is-active and is-enabled.
I grabbed this one from the todo list. Most of the functionality was
already there for is-active. I just needed to make check_one_unit take
the states to check for as an argument instead of the hardcoded
"active" and "reloading".
is-failed will return 1 if none of the units given are failed. This is
different from is-active which will return 3 if none of the units
given are active. It returns 3 with this comment:
/* According to LSB: "program is not running" */
As that does not make sense when looking for failed units I simply
chose 1 instead.
Yu Watanabe