Right now, adjusting test cases to cover some special edge-cases for a
specific shell is a little problematic.
For instance, if you add a special character that needs escaping
to one of the help strings, not only the tests for the shell you're
currently working on need adjustments, but rather all shell tests!
Contributors can often only work on one specific shell. The
need to adjust shells they don't know anything about discourages
testing or makes it simply impossible for a single contributor.
Arguably, this commit introduces a lot of code duplication, but on
the other hand it also makes testing a lot easier.
the test that was added tests for a panic that would occur
when a global argument had a conflict with another argument
that wasn't present in all the subcommands that the global
argument was present in
this would occur inside the get_arg_conflicts_with function
zsh completion generation would panic if a global argument
had conflicts with another argument which was present in its
own command but not in its subcommands
Adds new method/attribute `Arg::value_hint`, taking a `ValueHint` enum
as argument. The hint can denote accepted values, for example: paths,
usernames, hostnames, commands, etc.
This initial implementation supports hints for the zsh and fish
completion generators, support for other shells can be added later.
compgen takes exactly one "word" argument; if multiple are passed, it
will ignore all but the first. If `$cur` is not quoted and contains
whitespace, it will be expanded into multiple words, so the completions
will be based only on the first word of `$cur`, which is clearly not
correct.
To fix this, quote `$cur` to prevent word splitting, where necessary
(Bash does not perform word splitting inside the extended test command
`[[`, so quotes aren't necessary there).
