This continues the work started with the fix for #4273. There was
another bug caused by using the subcommand names without considering
their position in the argument list. If the user enters `git diff log
<TAB>`, we build up a string that identifies the subcommand. We ended
up making the string `git__diff__log` in this case because we appended
`__log` without considering the current state. Since `git__diff__log`
does not correspond to an actual command, we wouldn't provide any
suggestions.
This commit restructures the code so we walk subcommands and
subsubcommands in `bash.rs`. While walking those, we build up a list
containing triples of the parent `$cmd` name (e.g. `git__diff`), the
current command's name (e.g. `log`), and the `$cmd` for the current
command. We then build the shell script's case arms based on that
information.
We could instead have fixed#4280 by using the second element in the
pair returned from `utils::all_subcommands()` (a stringified list of
the subcommand path) instead of the first one. However, that would not
have helped us solve #4265.
Closes#4280
Early in the Bash-completion script, we build up a string that
identifies the command or subcommand. When we see the top-level
command's name (e.g. `git`) we set the command so far to that
value. We do that regardless of where in the argument list it
appears. For example, if the argument list is `git diff git`, we set
the current command to `git` when run into it the second time. We
therefore suggest arguments to the top-level command afterwards, which
is not correct.
This patch fixes that by also considering the string that identifies
the command so far, so we only set the overall command to `git` if the
command so far is the empty string.
This is actually just a step on the way to getting completion to work
for aliases of subcommands.
Closes#4273
Adds parser flags to toggle whether to run the
expensive clone logic for completions case.
Help completion will only suggest subcommands, not args.
clap_complete generator sets the flag.
Before we introduced actions, it required specific setups to engage with
claps version and help printing. With actions making that more
explicit, we don't get as much benefit from our multiple, obscure, ways
of users customizing help
Before
- Modify existing help or version with `mut_arg` which would
automatically be pushed down the command tree like `global(true)`
- Create an new help or version and have it treated as if it was the
built-in on (I think)
- Use the same flags as built-in and have the built-in flags
automatically disabled
- Users could explicitly disable the built-in functionality and do what
they want
Now
- `mut_arg` no longer works as we define help and version flags at the
end
- If someone defines a flag that overlaps with the built-ins by id,
long, or short, a debug assert will tell them to explicitly disable
the built-in
- Any customization has to be done by a user providing their own. To
propagate through the command tree, they need to set `global(true)`.
Benefits
- Hopefully, this makes it less confusing on how to override help
behavior. Someone creates an arg and we then tell them how to disable
the built-in
- This greatly simplifies the arg handling by pushing more
responsibility onto the developer in what are hopefully just corner
cases
- This removes about 1Kb from .text
Fixes#3405Fixes#4033
This reduces ambiguity in how the different "multiple" parts of the API
interact and lowrs the amount of API surface area users have to dig
through to use clap.
For now, this is only a matter of cleaning up the public API. Cleaning
up the implementation is the next step.
