[core] Enable interactive prompting for arguments

[forfudan]

This PR adds an interactive prompting capability to ArgMojo so arguments marked for prompting can be requested from the user when missing at parse time.

Changes:

• Add Argument.prompt() / Argument.prompt["custom text"]() builder methods and store prompt metadata on Argument.
• Prompt for missing prompt-enabled arguments during Command.parse_arguments() before defaults/validation.
• Add a new test suite plus docs and demo updates describing the feature.

[Copilot]

Pull request overview

Adds an interactive prompting capability to ArgMojo so arguments marked for prompting can be requested from the user when missing at parse time.

Changes:

• Add Argument.prompt() / Argument.prompt["custom text"]() builder methods and store prompt metadata on Argument.
• Prompt for missing prompt-enabled arguments during Command.parse_arguments() before defaults/validation.
• Add a new test suite plus docs and demo updates describing the feature.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated 13 comments.

Show a summary per file

File	Description
tests/test_prompt.mojo	New tests covering the prompt builder API and “prompt skipped when provided” behavior.
src/argmojo/command.mojo	Hooks prompting into parse_arguments() and implements _prompt_missing_args().
src/argmojo/argument.mojo	Adds _prompt / _prompt_text fields and the .prompt() builder overloads.
pixi.toml	Runs the new prompt test with stdin redirected to avoid blocking.
examples/demo.mojo	Adds a login subcommand demonstrating prompting and updates the feature list.
docs/user_manual.md	Documents interactive prompting and updates builder method references/tables.
docs/changelog.md	Adds an unreleased changelog entry for interactive prompting.
docs/argmojo_overall_planning.md	Marks interactive prompting as done and lists the new test file/feature.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

[Copilot]

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated 4 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

You can also share your feedback on Copilot code review. Take the survey.

[Copilot]

The _prompt_missing_args() docstring says it’s called “before _apply_defaults()”, but parse_arguments() currently calls it after defaults. Either update the docstring or (preferably) reorder the call sites to match, otherwise the documented default/empty-input behavior isn’t achievable.

Suggested change

	Called after the parsing loop but before ``_apply_defaults()``, so
	that user-provided prompt responses take priority and defaults
	fill in only remaining gaps. For flag arguments, the prompt
	accepts ``y``/``n`` (case-insensitive). For arguments with
	choices, valid options are shown. For arguments with a default,
	the default is displayed in parentheses and used when the user
	enters nothing.
	Prompting is skipped entirely when no arguments have
	``.prompt()`` enabled. When stdin is not a terminal (e.g.,
	This is called after the main parsing loop so that user-provided
	prompt responses can take priority over any existing values in
	the parse result. For flag arguments, the prompt accepts
	``y``/``n`` (case-insensitive). For arguments with choices,
	valid options are shown. For arguments with a default, the
	default is displayed in parentheses and used when the user
	enters nothing.
	Prompting is skipped entirely when no arguments have
	``.prompt()`` enabled. When stdin is not a terminal (e.g.,

[Copilot]

The prompt path treats a._number_of_values (nargs) as an append-style single value by calling _store_append_value(a, value, result) once. That diverges from normal parsing, where nargs consumes exactly N separate tokens and stores N list entries. Consider prompting N times (or parsing/splitting a single line into N values and enforcing the count) so number_of_values[N]() behaves consistently under prompting.

Suggested change

	self._store_append_value(a, value, result)
	# For arguments that append or accept multiple values (nargs),
	# ensure we store one entry per value, matching normal parsing.
	if a._number_of_values > 1:
	let parts = value.split()
	if len(parts) != a._number_of_values:
	self._warn(
	"Invalid number of values for '"
	+ a.name
	+ "': expected "
	+ String(a._number_of_values)
	+ ", got "
	+ String(len(parts))
	)
	# Do not store partial/invalid data; allow defaults
	# or other validation to handle this argument.
	continue
	for part in parts:
	self._store_append_value(a, part, result)
	else:
	self._store_append_value(a, value, result)

[Copilot]

The test suite here doesn’t exercise the “happy path” where input() returns a non-empty response (i.e. _prompt_missing_args() actually stores prompted values, validates choices, and interacts with defaults). Since prompting is the core feature being added, it’d be valuable to add at least one integration-style test that runs with stdin preloaded (e.g., redirect from a fixture file) and asserts the parsed results.

[Copilot]

parse_arguments() applies defaults before calling _prompt_missing_args(). Because _apply_defaults() writes defaults into result and ParseResult.has() treats any stored value as “provided”, prompt-enabled args that also have .default[]() will never prompt (and users can’t press Enter to accept the default or override it interactively). Consider reordering to prompt before defaults (e.g. parse → apply implications → prompt → apply defaults → re-apply implications → validate) so prompting works as documented for defaulted args.

Suggested change

	# Apply defaults, propagate implications, prompt for remaining
	# values, re-apply implications (in case prompts triggered new
	# ones), then validate constraints.
	self._apply_defaults(result)
	self._apply_implications(result)
	self._prompt_missing_args(result)
	# Propagate implications from explicitly provided args, prompt for
	# remaining values, apply defaults for anything still missing,
	# re-apply implications (in case prompts/defaults triggered new
	# ones), then validate constraints.
	self._apply_implications(result)
	self._prompt_missing_args(result)
	self._apply_defaults(result)