NAME

shcompgen - Generate shell completion scripts

VERSION

This document describes version 0.321 of shcompgen (from Perl distribution App-shcompgen), released on 2018-02-08.

SYNOPSIS

Initialize (this will create completion scripts directory, create shell script to initialize completion system):

 % shcompgen init

Generate shell completion scripts for all detectable programs in PATH:

 % shcompgen generate

Note that this distribution automatically runs 'init' and 'generate' the first time it is installed, or when you upgrade from an older version. So normally you don't have to do this manually.

Generate some programs only, replace if previously already exists, be verbose:

 % shcompgen generate --verbose --replace prog1 prog2 ./bin/prog3

List all shell completion scripts generated by us:

 % shcompgen list
 % shcompgen list --detail

Remove some shell completion scripts:

 % shcompgen remove prog1 prog2

Remove all generated shell completion scripts:

 % shcompgen remove

DESCRIPTION

Some shells, like bash/fish/zsh/tcsh, supports tab completion for programs. They are usually activated by issuing one or more complete (zsh uses compctl) internal shell commands. The completion scripts which contain these commands are usually put in (e.g., for fish) /etc/fish/completion/PROGNAME.fish (if one wants to install globally) or ~/.config/fish/completions/PROGNAME.fish (if one wants to install per-user).

This utility, shcompgen, can detect how to generate shell completion scripts for some programs and then install the completion scripts into the abovementioned location (the default is to per-user directory, but if running as root or with --global switch will install to the global directory).

It can also list all completion scripts generated by it, and be instructed to uninstall them again.

It supports several shells, currently: bash, fish, zsh, and tcsh. Shell-specific information can be found below.

Shell-specific information

bash

This script can work with the bash-completion package (and uses the same global completion directory: /etc/bash_completion.d). At the time of this writing, bash-completion (at version 2.1) does not yet look at per-user completion scripts directory. This script picks ~/.config/bash/completions/PROGNAME as location for per-user completion scripts. If later on bash-completion package decides on a different per-user location, this script will probably be adjusted too.

fish

Known issues include proper escaping of completion answer (e.g. when an answer contains a whitespace). To demonstrate this problem, try peri-eg-complete-fruits-any and type:

 % peri-eg-complete-fruits-any --fruit <tab>

The answer butternut squash is returned as butternut\\\ squash.

tcsh

So far I couldn't get fallback (a.k.a catch-all) completion mechanism to work in tcsh. For example if I do:

 complete '*' 'p/*/`helper`/'

then it will eclipse the other existing completion definitions.

So in tcsh, activating (or deactivating) completion is currently less convenient compared to the other shells. Instead of the complete definitions being put on a per-command basis in ~/.config/tcsh/completions/ directory, the init script ~/.config/shcompgen.tcsh will directly contain all the complete definitions. This script must be sourced to update the definitions. So after you

 % shcompgen generate foo

you will need to re-source the init script (or logout from + login back to the shell). And after you remove a completion script, you will need to uncomplete + re-source the init script (or logout from + login back to the shell).

Known issues include proper escaping of completion answer (e.g. when an answer contains a whitespace). To demonstrate this problem, try peri-eg-complete-fruits-any and type:

 % peri-eg-complete-fruits-any --fruit <tab>

The answer butternut squash is only returned as butternut\\ .

zsh

Known issues include proper escaping of completion answer (e.g. when an answer contains a whitespace). To demonstrate this problem, try peri-eg-complete-fruits-any and type:

 % peri-eg-complete-fruits-any --fruit <tab>

The answer butternut squash is only returned as butternut\\ .

Another known issue is still having to compinit after shcompgen generate or shcompgen remove.

And yet another known issue is having to compinit in the init script (~/.config/shcompgen.zsh) which is slowing down the shell startup.

Program detection

Below are the types/kinds of programs that can be detected. Expect the list to expand as more methods are added.

  • Scripts which are tagged with hints of what completion program to use

    You can put this line in a script, e.g. in a script called foo:

     # FRAGMENT id=shcompgen-hint command=bar

    The above line tells shcompgen that the script should be completed using an external program called bar. This will construct this completion script, e.g. for bash:

     complete -C bar foo
  • Completion programs which are tagged with hints of what programs they complete

    You can create a completion script in Perl (or other language, actually), e.g. _foo and tag it with hints of what programs they complete, e.g.

     # FRAGMENT id=shcompgen-hint completer=1 for=foo,foo-this-host

    This will add completion script for foo:

     complete -C _foo foo

    as well as for foo-this-host:

     complete -C _foo foo-this-host
  • Getopt::Long::Complete-based CLI scripts

    If a script foo is detected as a Perl script using Getopt::Long::Complete, we know that it can complete itself. Thus, shcompgen will generate this completion script (e.g. for bash):

     complete -C foo foo
  • Getopt::Long::Subcommand-based CLI scripts

    If a script foo is detected as a Perl script using Getopt::Long::Subcommand, we know that it can complete itself. Thus, shcompgen will generate this completion script (e.g. for bash):

     complete -C foo foo
  • Perinci::CmdLine-based CLI scripts

    If a script like foo is detected as a Perl script using Perinci::CmdLine (or its variant like Perinci::CmdLine::Lite or Perinci::CmdLine::Any) we know that it can complete itself. Thus, shcompgen will add this completion script e.g. for bash:

     complete -C foo foo
  • Other methods

    Other methods will be added in the future, e.g. by parsing manpage or POD, and so on.

SUBCOMMANDS

detect-prog

(Utility) Guess running shell.

generate

Generate shell completion scripts for detectable programs.

guess-shell

(Utility) detect a program.

init

Initialize shcompgen.

This subcommand creates the completion directories and initialization shell script, as well as run generate.

list

List all shell completion scripts generated by this script.

remove

Remove shell completion scripts generated by this script.

OPTIONS

* marks required options.

Common options

--config-path=filename

Set path to configuration file.

Can be specified multiple times.

--config-profile=s

Set configuration profile to use.

--debug

Shortcut for --log-level=debug.

--fish

Shortcut for --shell=fish.

See --shell.

--format=s

Choose output format, e.g. json, text.

Default value:

 undef
--help, -h, -?

Display help message and exit.

--json

Set output format to json.

--log-level=s

Set log level.

--naked-res

When outputing as JSON, strip result envelope.

Default value:

 0

By default, when outputing as JSON, the full enveloped result is returned, e.g.:

    [200,"OK",[1,2,3],{"func.extra"=>4}]

The reason is so you can get the status (1st element), status message (2nd element) as well as result metadata/extra result (4th element) instead of just the result (3rd element). However, sometimes you want just the result, e.g. when you want to pipe the result for more post-processing. In this case you can use `--naked-res` so you just get:

    [1,2,3]
--no-config

Do not use any configuration file.

--no-env

Do not read environment for default options.

--quiet

Shortcut for --log-level=error.

--shell=s

Override guessing and select shell manually.

Valid values:

 ["bash","fish","zsh","tcsh"]
--subcommands

List available subcommands.

--tcsh

Shortcut for --shell=tcsh.

See --shell.

--trace

Shortcut for --log-level=trace.

--verbose

Shortcut for --log-level=info.

--version, -v

Display program's version and exit.

--zsh

Shortcut for --shell=zsh.

See --shell.

Options for subcommand detect-prog

--prog=s*

Options for subcommand generate

--per-option

Create per-option completion script if possible.

If set to true, then attempt to create completion script that register each option. This creates nicer completion in some shells, e.g. fish and zsh. For example, option description can be shown.

This is possible for only some types of scripts, e.g. <pm:Perinci::CmdLine>- (that does not have subcommands) or <pm:Getopt::Long::Descriptive>-based ones.

--prog-json=s

Program(s) to generate completion for (JSON-encoded).

See --prog.

--prog=s@

Program(s) to generate completion for.

Can contain path (e.g. `../foo`) or a plain word (`foo`) in which case will be searched from PATH.

Can be specified multiple times.

--remove

Remove completion for script that (now) is not detected to have completion.

The default behavior is to simply ignore existing completion script if the program is not detected to have completion. When the `remove` setting is enabled, however, such existing completion script will be removed.

--replace

Replace existing script.

The default behavior is to skip if an existing completion script exists.

--stdout

Output completion script to STDOUT.

Options for subcommand init

--per-option

Create per-option completion script if possible.

If set to true, then attempt to create completion script that register each option. This creates nicer completion in some shells, e.g. fish and zsh. For example, option description can be shown.

This is possible for only some types of scripts, e.g. <pm:Perinci::CmdLine>- (that does not have subcommands) or <pm:Getopt::Long::Descriptive>-based ones.

Options for subcommand list

--detail, -l
--per-option

Create per-option completion script if possible.

If set to true, then attempt to create completion script that register each option. This creates nicer completion in some shells, e.g. fish and zsh. For example, option description can be shown.

This is possible for only some types of scripts, e.g. <pm:Perinci::CmdLine>- (that does not have subcommands) or <pm:Getopt::Long::Descriptive>-based ones.

Options for subcommand remove

--per-option

Create per-option completion script if possible.

If set to true, then attempt to create completion script that register each option. This creates nicer completion in some shells, e.g. fish and zsh. For example, option description can be shown.

This is possible for only some types of scripts, e.g. <pm:Perinci::CmdLine>- (that does not have subcommands) or <pm:Getopt::Long::Descriptive>-based ones.

--prog-json=s

Program(s) to remove completion script of (JSON-encoded).

See --prog.

--prog=s@

Program(s) to remove completion script of.

Can contain path (e.g. `../foo`) or a plain word (`foo`) in which case will be searched from PATH.

Can be specified multiple times.

COMPLETION

This script has shell tab completion capability with support for several shells.

bash

To activate bash completion for this script, put:

 complete -C shcompgen shcompgen

in your bash startup (e.g. ~/.bashrc). Your next shell session will then recognize tab completion for the command. Or, you can also directly execute the line above in your shell to activate immediately.

It is recommended, however, that you install modules using cpanm-shcompgen which can activate shell completion for scripts immediately.

tcsh

To activate tcsh completion for this script, put:

 complete shcompgen 'p/*/`shcompgen`/'

in your tcsh startup (e.g. ~/.tcshrc). Your next shell session will then recognize tab completion for the command. Or, you can also directly execute the line above in your shell to activate immediately.

It is also recommended to install shcompgen (see above).

other shells

For fish and zsh, install shcompgen as described above.

CONFIGURATION FILE

This script can read configuration files. Configuration files are in the format of IOD, which is basically INI with some extra features.

By default, these names are searched for configuration filenames (can be changed using --config-path): ~/.config/shcompgen.conf, ~/shcompgen.conf, or /etc/shcompgen.conf.

All found files will be read and merged.

To disable searching for configuration files, pass --no-config.

To put configuration for a certain subcommand only, use a section name like [subcommand=NAME] or [SOMESECTION subcommand=NAME].

You can put multiple profiles in a single file by using section names like [profile=SOMENAME] or [SOMESECTION profile=SOMENAME] or [subcommand=SUBCOMMAND_NAME profile=SOMENAME] or [SOMESECTION subcommand=SUBCOMMAND_NAME profile=SOMENAME]. Those sections will only be read if you specify the matching --config-profile SOMENAME.

You can also put configuration for multiple programs inside a single file, and use filter program=NAME in section names, e.g. [program=NAME ...] or [SOMESECTION program=NAME]. The section will then only be used when the reading program matches.

Finally, you can filter a section by environment variable using the filter env=CONDITION in section names. For example if you only want a section to be read if a certain environment variable is true: [env=SOMEVAR ...] or [SOMESECTION env=SOMEVAR ...]. If you only want a section to be read when the value of an environment variable has value equals something: [env=HOSTNAME=blink ...] or [SOMESECTION env=HOSTNAME=blink ...]. If you only want a section to be read when the value of an environment variable does not equal something: [env=HOSTNAME!=blink ...] or [SOMESECTION env=HOSTNAME!=blink ...]. If you only want a section to be read when an environment variable contains something: [env=HOSTNAME*=server ...] or [SOMESECTION env=HOSTNAME*=server ...]. Note that currently due to simplistic parsing, there must not be any whitespace in the value being compared because it marks the beginning of a new section filter or section name.

List of available configuration parameters:

Common for all subcommands

 format (see --format)
 log_level (see --log-level)
 naked_res (see --naked-res)
 shell (see --shell)

Configuration for subcommand detect-prog

 prog (see --prog)

Configuration for subcommand generate

 per_option (see --per-option)
 prog (see --prog)
 remove (see --remove)
 replace (see --replace)
 stdout (see --stdout)

Configuration for subcommand guess-shell

Configuration for subcommand init

 per_option (see --per-option)

Configuration for subcommand list

 detail (see --detail)
 per_option (see --per-option)

Configuration for subcommand remove

 per_option (see --per-option)
 prog (see --prog)

ENVIRONMENT

SHCOMPGEN_OPT => str

Specify additional command-line options.

FILES

~/.config/shcompgen.conf

~/shcompgen.conf

/etc/shcompgen.conf

HOMEPAGE

Please visit the project's homepage at https://metacpan.org/release/App-shcompgen.

SOURCE

Source repository is at https://github.com/perlancar/perl-App-shcompgen.

BUGS

Please report any bugs or feature requests on the bugtracker website https://rt.cpan.org/Public/Dist/Display.html?Name=App-shcompgen

When submitting a bug or request, please include a test-file or a patch to an existing test-file that illustrates the bug or desired feature.

SEE ALSO

Dist::Zilla::Plugin::GenShellCompletion

AUTHOR

perlancar <perlancar@cpan.org>

COPYRIGHT AND LICENSE

This software is copyright (c) 2018, 2017, 2016, 2015, 2014 by perlancar@cpan.org.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.