Add the suggested changes
Fix test issue_1050_num_vals_and_defaults
It used a required which isn't needed for the purpose of the test.
Add tests for the default+required assertions
Add test for positional args with long or short
Some more fixes in addition to 594c535ba2
* Some I noticed manually.
* Some were found by Topy (https://github.com/intgr/topy), either new
rules from Typo or new code in clap.
* `etc.` and `e.g.` rules were disabled during the last run.
Argument prefixed by `-` will never inferred as a subcommand
Suggest using subcommand when supplied after `--`
Cargo fmt, adding test for wrongly using subcommand after `--`
Fix test
Fix clippy
This commit introduces a new feature called `"regex"`. It adds a new
function `validator_regex` to `Arg` and was inspired by the discussion
in #1968. The name `validator_regex` was chosen instead of
`regex_validator` to make sure that the developer keeps in mind that
there may only be a single `Validator` on an `Arg`.
The feature can be used with YAML files, however there is no proper
pattern in `clap_app!` (yet).
2109: fix: detect if space should be added before spec_vals r=CreepySkeleton a=knidarkness
Co-authored-by: knidarkness <knidarkness@gmail.com>
Co-authored-by: Sergey Dubovyk <knidarkness@gmail.com>
Using `App::get_matches` from the integration tests meant that any
CLI arguments passed to libtest would also be captured by clap, often
causing the tests to fail.
For example, running `cargo test --test help -- --nocapture` would
result in several failed tests, even though `cargo test --test help`
worked fine. This was very surprising/confusing.
This commit makes the tests no longer implicitly rely on the value of
`env::args_os()`, which means developers can now provide arguments to
libtest without failures.
The YAML null value is called `null` in YAML, not `Null`. yaml-rust
handles those values according to spec, so we should use the correct
capitalization.
See https://yaml.org/spec/1.2/spec.html#id2803362 for more
information.
While reading the code for the yaml translation, I've noticed that
there is a bug in the macro `yaml_opt_str` as well as a wrong `null`
value in the test fixture.
These tests add the expected behaviour on the given fixture, e.g.
prog
and
prog <WRONG>
where `<WRONG>` is **not** `other` should yield `None` for
`"positional2"`, whereas
prog other
should yield "something" and
prog --flag ARBITRARY_VALUE
should yield "some". The first two tests
- default_value_if_not_triggered
- default_value_if_not_triggered_by_argument
fail, as the second positional arguments *gets set*, although its
conditions aren't fulfilled.
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.
Tests to check for conflicts between flag subcommands long and short and
their aliases and args both long and short and their aliases. Tests to
handle self conflicts, where a flag subcommand short or long with have a
corresponding alias with the same value.
Instead of a `FlagSubCommand` struct the addition of two simple methods
to `App`. `App::long_flag` and `App::short_flag` that cover all the
bases of the many methods that were provided in `FlagSubCommand`. This
API is far simpler to use and more akin to the present `Arg::long` and `Arg::short`.
this implements Arg::short_alias, Arg::short_aliases,
Arg::short_visible_alias, and Arg::short_visible_aliases in addition to
adding their associated tests
These are similar to the tests backported to the 2.x branch in
https://github.com/clap-rs/clap/pull/1907. These are more comprehensive,
because the OsStr splitting methods are more robust now.
In particular, fix macros that take an enum of one of the the following forms:
- `#[...] enum { ... , }`
- `pub enum { ... , }`
- `enum { ... , }`
Previously, these expansions would result in an error message like "error: no
rules expected the token `:`".
Add extensive tests for each pattern. Only two of the patterns had tests
before, so these errors did not surface automatically.
1612: Use about() with help() and long_about() with long_help() r=pksunkara a=TheLostLambda
I was going through the clap documentation and was under the impression that calling `help()` would call `about()` and `long_help()` would call `long_about()`, but I've actually discovered this not to be the case. Instead, the `long_about()` was always shown when it existed, rendering the output (in the about section) of programs called with `-h` and `--help` identical. Issue #1472 shows this and that is fixed here.
Note this doesn't remove the ability to use the same about in both cases: if `long_about()` is unset, then `about()` is used in both cases.
I've changed the implementation here to use `is_some()` and `unwrap()` as opposed to `if let` because it ultimately allows for less repetitive code. Ideally, I'd be able to pair `if let` with a secondary condition (namely `self.use_long`), but to my dismay, let-chains are not stabilized yet.
For a second opinion, here is the code a settled on:
```
if self.use_long && parser.meta.long_about.is_some() {
debugln!("Help::write_default_help: writing long about");
write_thing!(parser.meta.long_about.unwrap())
} else if parser.meta.about.is_some() {
debugln!("Help::write_default_help: writing about");
write_thing!(parser.meta.about.unwrap())
}
```
Here is the alternative:
```
if self.use_long {
if let Some(about) = parser.meta.long_about {
debugln!("Help::write_default_help: writing long about");
write_thing!(about)
} else if let Some(about) = parser.meta.about {
debugln!("Help::write_default_help: writing about");
write_thing!(about)
}
} else {
if let Some(about) = parser.meta.about {
debugln!("Help::write_default_help: writing about");
write_thing!(about)
}
}
```
Co-authored-by: Brooks J Rady <b.j.rady@gmail.com>
> incidentally, how do we feel about adding a rustfmt check to the CI(s)?
yes we should be doing that. you can send another pr that adds the check to the Ci
The example code.
~~~rust
use clap::{App, Arg};
fn main() {
let matches = App::new("My Super Program")
.arg(
Arg::with_name("verbose")
.help("Sets the level of verbosity")
.short('v')
.long("verbose")
.takes_value(false)
.multiple_occurrences(true)
.env("VERBOSE"),
)
.get_matches();
match matches.occurrences_of("verbose") {
0 => println!("0 No verbose info"),
1 => println!("1 Some verbose info"),
2 => println!("2 Tons of verbose info"),
3 | _ => println!("3 >= Don't be crazy"),
}
}
~~~
It code use multiple_occurrences with env.
But it do not work.
`env` method set require take value.
It result see under.
~~~console
% cargo run -- -v
Finished dev [unoptimized + debuginfo] target(s) in 0.01s
Running `target/debug/foo -v`
error: The argument '--verbose <verbose>...' requires a value but none was supplied
USAGE:
foo [OPTIONS]
For more information try --help
~~~
And, structopt or clap_derive may be create similar code.
So I am confused by structopt.
This to fix code small.
- Manually fix some problems
- Run 'cargo fix --clippy'
Commits taken from similar PRs open at that time:
- Replace indexmap remove with swap_remove
Resolves#1562 and closes#1563
- Use cognitive_complexity for clippy lint
Resolves#1564 and closes#1565
- Replace deprecated trim_left_matches with trim_start_matches
Closes#1539
Co-authored-by: Antoine Martin <antoine97.martin@gmail.com>
Co-authored-by: Brian Foley <bpfoley@users.noreply.github.com>
This patch:
* Removes the `ArgSettings::Global` variant, and replaces all
users of it to `Arg::global(...)`. The variant itself is lifted up
into a field on Arg. This was deprecated in clap 2.32.0.
* Removes AppFlags::PropagateGlobalValuesDown. This was deprecated in
clap 2.27.0.
* Removes `Arg::empty_values`. This was deprecated in clap 2.30.0.
* Removes `ArgMatches::usage`. This was deprecated in clap 2.32.0.
subcommands
This commit changes the internal ID to a u64 which will allow for
greater optimizations down the road. In addition, it lays the ground
work for allowing users to use things like enum variants as argument
keys instead of strings.
The only downside is each key needs to be hashed (the implementation
used is an FNV hasher for performance). However, the performance gains
in faster iteration, comparison, etc. should easily outweigh the single
hash of each argument.
Another benefit of if this commit is the removal of several lifetime
parameters, as it stands Arg and App now only have a single lifetime
parameter, and ArgMatches and ArgGroup have no lifetime parameter.
Args can now be added to custom help sections. This breaks up the builder pattern a little by adding help section declarations inline, but it's the most intuitive method and doesn't require strange nesting that feels awkward.
```rust
app::new("foo")
.arg(Arg::with_name("arg1")) // under normal headers
.help_heading("SPECIAL")
.arg(Arg::with_name("arg2")) // under SPECIAL: heading
```
Closes#805
Add logic to filter based on hidden long/short.
There is still an issue with the logic in parser.rs use_long_help. This
causes invalid evaluation of whether to show/hide based on long or short help
Complete check for use_long_help, add tests
For version 3, we want the args section to immediately follow
the usage section in the default help message.
One change that I am unhappy with is needing to make "write_arg"
in app/help.rs accept an extra param that makes it suppress the
extra line. This is to prevent an extra blank line from appearing
between args and options in the default help, and seems necessary,
but there might be a better way.
Optional positionals mixed with subcommands will still break this, since
I can't see how to tell which element of $line is the command than.
Mixing optional positionals with subcommands is a bit weird and awkward
though...
This changes the way we complete positionals to complete them using
_arguments, as should be done, instead of completing their uppercase name
as a string.
Currently I made it offer _files completion for all positional arguments.
This can be improved to complete actual possible values of the arguments
and only complete files if the argument truly takes them. But this will
require further changes in clap to actually have the required
functionality to get this information.
