Before, validating UTF-8 was all-or-nothing and would cause a `panic` if
someone used the right API with non-UTF-8 input.
Now, all arguments are validated for UTF-8, unless opted-out. This
ensures a non-panicing path forward at the cost of people using the
builder API that previously did `value_of_os` need to now set this flag.
Fixes#751
Previously, we paniced. This is one less reason people have to call
lower level `get_matches` to get the commonly expected behavior, making
it more likely for the Rust community to "do the right thing"
Fixes#2659
Instead they should behave like `default_value`/`default_values`.
In implementingt this, I didn't see any reason to be using a `VecMap`.
In fact, this helped simplify the code / make intent clearer.
With this, we are also able to simplify the derive macro work from #2633.
Fixes#2634
BREAKING CHANGE: `value_name`/`value_names` always overwrite, rather
than append. We expect the impact to be minimal.
Rename needs_valueof to parse_result
Use ParseResult more
Less predicting, more fallback
Remove non-sense ParsingResult::NotFound
Merge FlagSubCommand and FlagSubCommandShort
Merge NoMatchingLongArg and NoMatchingShortArg
Better documentation for pos_counter bumping
Denoise of pos_counter
Split ParseState from ParseResult
Remove ParseResult::Flag
small cleanup
When using `#[clap(flatten)]` inside of a `Subcommand`, we would do a
`from` instead of an `update`.
The challenge is knowing when we are
going into a flattened subcommand vs changing the variant. To resolve
this, I added a `Subcommand:has_subcommand(name)` trait method that we
generate, so we can ask.
Improved the `impl`s of `From` for `&'help Yaml` to use expects with useful messages instead of unwraps. Also made changes to avoid potentially fetching redundant data from YAML hashes and unwrapping the same data multiple times. Finally, there are tests to ensure the panics are correct.
PR #1637 switched clap to report `64` on errors and then #1653 switch it
to `2`, but both missed a case. This also documents the reason why inline
since I had to go and dig through the history to re-discover the
motivation.
Before, partial command lines would panic at runtime. Now it'll be a
compile error
For example:
```
pub enum Opt {
Daemon(DaemonCommand),
}
pub enum DaemonCommand {
Start,
Stop,
}
```
Gives:
```
error[E0277]: the trait bound `DaemonCommand: clap::Args` is not satisfied
--> clap_derive/tests/subcommands.rs:297:16
|
297 | Daemon(DaemonCommand),
| ^^^^^^^^^^^^^ the trait `clap::Args` is not implemented for `DaemonCommand`
|
= note: required by `augment_args`
```
To nest this, you currently need `enum -> struct -> enum`. A later
change will make it so you can use the `subcommand` attribute within
enums to cover this case.
This is a part of #2005
While having convinience derives can be helpful, deriving traits that
are not used in similar situations (`Clap` and `ArgEnum`) can make
things harder
- From a user, derives are opaque and create uncertainty on how to use
the API if not kept crystal clear (deriving a name gives you the trait
by that name)
- This makes documentation harder to write and read
- You can use types in unintended places, which is made worse for crate
APIs because changing this breaks compatibility.
Fixes#2584
* add a new arg option for the max_occurrences
* check ErrorKind in tests
* Updated grammer in doc comments
Co-authored-by: Pavan Kumar Sunkara <pavan.sss1991@gmail.com>
* assert is_err() before unwraping
Co-authored-by: Pavan Kumar Sunkara <pavan.sss1991@gmail.com>
* docs: fix links to Subcommand
These were broken by 03333800fe
* docs: remove crate links from body content
Addresses code review
* docs: allow some room to breath
Addresses code review
* refactor(validator): use filtermap style for `gather_conflicts` and `validate_exclusive`
* refactor(validator): remove nested `filter`s in `build_conflict_err_usage`
* refactor(validator): add more code beautifying
* refactor(validator): revert some unnecessary changes
* refactor(validator): revert some unnecessary changes, take 2
* fix(wip): add necessary debugging prints for analysis
* fix(wip): implement the fix, take 1
* docs: add docstring for `Parser::flag_subcmd_at` and `Parser::flag_subcmd_skip`
* test: make `flag_subcommand_short_after_long_arg` test more realistic
Refactoring and better test cases
Refactored SubcommandHelpShowsLongForm to
UseLongFormatForHelpSubcommand.
Tests and docuemntation examples use about and long_about instead of
(before/after)_help.
Removed commented out tests
Linting: Fix trailing new line
Updated change log, refactored tests and doc str
Reordered items in the Changelog
New test added and old tests removed that were redundant
Doc string for AppSettings::UseLongFormatForHelpSubcommand fixed
Change default help template:
- The new template introduce new lines before and after
author/about sections.
- Add help template placeholders:
- about-section
- author-section
- Documentation of new placeholders in clap::App::help_template
- Update all unit tests by incorporating new lines
Following changes are implemented:
- Add 'clap::Arg::get_visible_aliases(&self)' method
This new method provides access to visible argument aliases
- Add 'clap::Arg::get_short/long_and_visible_aliases
This new method provides access to the short/long arguments and its
visible aliases
- Add visible aliases completions in clap_generate for following shells:
- Bash
- Elvish
- Fish
- Powershell
- Zsh
- Add test fixtures for testing visible alias completions in clap_generate:
'clap_generate/tests/completions/<shell>::<shell>_with_aliases()'
This removes the direct dependency on unicode-width and delegates the
complexity of computing the displayed width of text to the Textwrap
crate.
The `display_width` function handles characters like “æøå” (Danish),
“äöü” (German), and “😂✨😍” (emojis) – even if the unicode-width
Cargo feature is disabled.
This is an improvement of the former `str_width` function which would
over-estimate the width of emojis and non-ASCII characters (since they
are several bytes wide).
* docs: Correct method name in example
This looks like it was copied from the documentation for
`required_unless_present_all` but not updated accordingly for this
method.
* docs: Add note re `required_unless_present_all` to `required_unless_present_any`
There's a note in the documentation for `required_unless_present_all`
telling users to check out `required_unless_present_any` if that's what
they want. I figured it might be useful to have a similar note in the
documentation for `required_unless_present_any` pointing to that method
as well.
* docs: Fix `required` link in docs for `required_unless_present_all`
* docs: Correct "if" to "unless"
