Created a unit test that tests the unknown flag
error message when the unknown flag is located
in different arg positions.

)

Fixes spf13#1847

Don't require contributors to install richgo but keep it as an option and for CI

…pf13#1851)

Bumps [golangci/golangci-lint-action](https://github.com/golangci/golangci-lint-action) from 3.2.0 to 3.3.1.
- [Release notes](https://github.com/golangci/golangci-lint-action/releases)
- [Commits](golangci/golangci-lint-action@v3.2.0...v3.3.1)

---
updated-dependencies:
- dependency-name: golangci/golangci-lint-action
  dependency-type: direct:production
  update-type: version-update:semver-minor
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

Signed-off-by: David Wertenteil <dwertent@armosec.io>

Remove testing for go 1.15 to allow CI to pass, but don't force projects to upgrade.

`template` is an import in `cobra.go` file and also used as a variable
name, which masks the library in the scope of that function.

Based on
badges/shields#8671

Corrected the function name at the start of doc strings, as per the convention
outlined in official go documentation: https://go.dev/blog/godoc

…pf13#1817)

Fixes spf13#1816

Previously, arguments with a dash as the second character (e.g., 1-ff00:0:1)
were detected as a flag by mistake. This resulted in auto completion misbehaving
if such an argument was last in the argument list during invocation.

align documentation with the code : completions.go:452

…pf13#1902)

Signed-off-by: John McBride <jpmmcbride@gmail.com>

This allows programs to request the shell to maintain the order of completions that was returned by the program

PowerShell 7.2 has changed the way arguments are passed to executables.
This was originally an experimental feature in 7.2, but as of 7.3 it is
built-in. A simple "" is now sufficient for passing empty arguments, no
back-tick escaping is required.

Fixes spf13#1849

Signed-off-by: Oldřich Jedlička <oldium.pro@gmail.com>
Co-authored-by: Oldřich Jedlička <oldrich.jedlicka@rohlik.cz>

Signed-off-by: Florent Poinsard <florent.poinsard@outlook.fr>

Signed-off-by: Luiz Carvalho <lucarval@redhat.com>

Although it is not the recommended approach, sourcing a completion
script is the simplest way to get people to try using shell completion.
Not allowing it for zsh has turned out to complicate shell completion
adoption.  Further, many tools modify the zsh script to allow sourcing.

This commit allows sourcing of the zsh completion script.

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>

Fixes spf13#1880

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>
Co-authored-by: Deleplace <deleplace2015@gmail.com>

…pf13#1971)

The use in generated bash completion files is getting flagged by
Lintian (the Debian package linting tool).

Signed-off-by: Taavi Väänänen <hi@taavi.wtf>

Use temporary files instead of assuming the current directory is
writable. Also, if creating a temporary file still returns an error,
prevent the test from failing silently by replacing `log.Fatal` with
`t.Fatal`.

This fixes an issue with program names that include a dot, in our case
`podman.exe`. This was caused by the change in commit 6ba7ebb.

Fixes spf13#1853

Signed-off-by: Paul Holzinger <pholzing@redhat.com>

…pf13#1976)

…3#1983)

…pf13#2021)

)

spf13#2047)

Currently, only one of the persistent pre-runs and post-runs is executed.
It is always the first one found in the parents chain, starting at this command.
Expected behavior is to execute all parents' persistent pre-runs and post-runs.

Dependent projects implemented various workarounds for this:
- manually building persistent hook chains (in every hook).
- applying some kind of monkey-patching on top of Cobra.

This change eliminates the necessity for such workarounds
by allowing to set a global variable EnableTraverseRunHooks.

Tickets:
- spf13#216
- spf13#252

Signed-off-by: Volodymyr Khoroz <volodymyr.khoroz@foundries.io>

When using golangci-lint v1.55.0 some new errors were being reported.

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>

Fixes spf13#2060

When a command sets `DisableFlagParsing = true` it requests the
responsibility of doing all the flag parsing. Therefore even the
`--help/-f/--version/-v` flags should not be automatically completed
by Cobra in such a case.

Without this change the `--help/-h/--version/-v` flags can end up being
completed twice for plugins: one time from cobra and one time from the
plugin (which has set `DisableFlagParsing = true`).

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>

Different problems have been reported about flag completion registration.
These two tests are the cases that were not being verified but had been
mentioned as problematic.

Ref:
- spf13#1320
- spf13#1438 (comment)

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>

…pf13#1940)

* Replace all non-alphanumerics in active help env var program prefix

There are other characters besides the dash that are fine in program
names, but are problematic in environment variable names. These include
(but are not limited to) period, space, and non-ASCII letters.

* Another change in docs to mention non-ASCII-alphanumeric instead of just dash

In this case the executable is `kubectl-plugin`, but we run it as:

    kubectl plugin

And the help text should reflect the actual usage of the command.

To create a plugin, add the cobra.CommandDisplayNameAnnotation:

    rootCmd := &cobra.Command{
        Use: "plugin",
        Annotations: map[string]string{
            cobra.CommandDisplayNameAnnotation: "kubectl plugin",
        }
    }

Internally this change modifies CommandPath() for the root command to
return the command display name instead of the command name. This is
used for error messages, help text generation, and completions.

CommandPath() is expected to have spaces and code using it already
handle spaces (e.g replacing with _), so hopefully this does not break
anything.

Fixes: spf13#2017

Signed-off-by: Nir Soffer <nsoffer@redhat.com>

The new API is simpler and matches the `c.RegisterFlagCompletionFunc()`
API.  By removing the global function `GetFlagCompletion()` we are more
future proof if we ever move from a global map of flag completion
functions to something associated with the command.

The commit also makes this API work with persistent flags by using
`c.Flag(flagName)` instead of `c.Flags().Lookup(flagName)`.

The commit also adds unit tests.

Signed-off-by: Marc Khouzam <marc.khouzam@gmail.com>