Commit graph

273 commits

Author SHA1 Message Date
Will-Low
41c0d514bc Update doctring for BoolishValueParser::new 2024-04-23 12:14:42 -07:00
Ed Page
93d40450f7 docs: Fix typo 2024-04-09 12:05:12 -05:00
Ed Page
f65d421607 chore: Release 2024-03-06 10:52:18 -06:00
Amaan Qureshi
2aea9504c4
fix: Use SCREAMING_SNAKE_CASE for static variable authors 2024-03-05 23:52:48 -05:00
Ben Wiederhake
0076cac7cb fix(builder): Don't doc-link to undocumented item 2024-02-23 01:40:29 +01:00
Ed Page
8a7a13a561 chore: Release 2024-02-16 07:08:54 -06:00
Ed Page
446328a8d3 fix(error): Include -- in more cases
Inspired by rust-lang/cargo#12494.
Part of this is that our "did you mean" does prefix checks so it can be
overly aggressive in providing suggestions.
To avoid providing needless suggestions I limited this change to `last`
/ `trailing_var_arg` as those convey that `--` is more likely a valid
suggestion.
2024-02-16 06:54:57 -06:00
Ed Page
323345b3bc fix: Add missing clap::builder::OsStr impl
Found via rust-lang/cargo#10554
2024-02-10 19:59:40 -06:00
Ed Page
a751c5fe65 chore: Release 2024-02-08 10:34:06 -06:00
Ed Page
9ec6c942b8 chore: Release 2024-02-08 10:33:13 -06:00
renovate[bot]
cd0ca3cc13 chore(deps): update rust crate strsim to 0.11.0 2024-02-01 01:06:03 +00:00
Ed Page
0134f45ff0 chore: Release 2024-01-16 14:28:07 -06:00
Ed Page
2f1890907e
Merge pull request #5314 from epage/required
fix(parser): Don't duplicate requireds when usage disabled
2024-01-16 14:26:40 -06:00
Ed Page
0a635b9a20 fix(parser): Don't duplicate requireds when usage disabled
Skipping a test for now because we don't have a mixed-feature testing
strategy.

Fixes #5312
2024-01-16 14:15:48 -06:00
Martin Pool
8c83971b8c
docs: Link to exit code info 2024-01-15 14:30:11 -08:00
Ed Page
ef98da27cb chore: Release 2024-01-15 10:20:03 -06:00
Ed Page
50c59f63bb fix(parser): Don't panic with conflicts and groups
With #5298, I had overlooked that `matcher.arg_ids()` includes
`ArgGroup`s.  I had assumed I could always find a present `id` among
`Arg`s and `unwrap`ed.

I skipped a test for this because the use case is a bit strange that the
long term value for the test would likely be low.
If/when we add derive support for `args_conflicts_with_subcommands`, it
will then cover this case.

Fixes #5304
2024-01-15 09:30:19 -06:00
Ed Page
cbc9c9dd44 chore: Release 2024-01-11 20:52:38 -06:00
Ed Page
2a3356a535 fix: Update anstream 2024-01-11 20:33:39 -06:00
Ed Page
12b5c39d46 chore: Release 2024-01-11 10:37:53 -06:00
Ed Page
f529ec398c fix(parser): Ensure subcommand flags can conflict
Fixes #5297
2024-01-11 10:25:57 -06:00
Ed Page
a7e04a53e4 fix(parser): Improve subcommand conflict error 2024-01-11 10:20:19 -06:00
Ed Page
ea00ef3051 refactor(error): Allow more conflict sources 2024-01-11 10:17:35 -06:00
Ed Page
e47d8a2a66 refactor(parser): Clarify arg error 2024-01-11 09:47:00 -06:00
Ed Page
514f28bc92 chore: Release 2024-01-08 10:39:47 -06:00
Ed Page
3eaf1af052 fix(help): Correctly show help for fake flags 2024-01-08 09:58:42 -06:00
Ed Page
148e102ac3 fix(builder): Allow custom flag definitions
The goal is to allow getting the position of repeated flags for #5287
2024-01-08 09:49:43 -06:00
Ed Page
6d601e6f31 chore: Release 2023-12-28 12:07:48 -06:00
Sean Allred
05cd057978
perf: Avoid retrieving possible_values unless used
In some sophisticated situations, these may be expensive to calculate.
One example might be a '--branch' option accepting any single Git
branch that exists on the remote -- in such a case, the remote would
need to be queried for all possible_values. The cost is ultimately
unavoidable at runtime since this validation has to happen eventually,
but there's no need to pay it when generating help text if
`is_hide_possible_values_set`.

To keep '-h' fast, avoid collecting `possible_values` during '-h'
unless we're actually going to use the values in display.

This optimization is repeated for the manpage renderer.

This is trivially based on the short-circuiting logic at [1], which at
least supports the idea that actually consuming the iterator is not
generally-guaranteed behavior when `hide_possible_values` is set.

Note on the 'expensive' mod: This keeps all the possible_values tests
in one file but allows the entire set of tests to be controlled by the
'strings' feature (which is required to be able to use String rather
than str for each possible value).

[1]: clap_builder/src/builder/command.rs:long_help_exists_
2023-12-28 10:06:35 -06:00
Ed Page
d092896d61 chore: Release 2023-12-04 12:17:32 -06:00
Ed Page
37917be0b7 feat: Add Command::mut_group
Fixes #5038
2023-12-04 12:03:05 -06:00
Ed Page
21b671f689 chore: Release 2023-11-27 11:57:26 -06:00
Ed Page
b13f6d9862 fix(help): Hide 'help' if only flattened subcommand 2023-11-27 10:49:21 -06:00
Ed Page
a9322cabf3 fix(help): Hide subcommands in flattened usage 2023-11-27 10:43:16 -06:00
Ed Page
22130e3057 fix(help): Use right about when flattening
Fixes #5226
2023-11-27 09:28:30 -06:00
Kelly Thomas Kline
7568d5fa67
Update command.rs
Correct wording.
2023-11-16 21:06:03 -08:00
Ed Page
3aeea916e8 chore: Release 2023-11-10 16:24:46 -06:00
Ed Page
9c0f7a7253 fix(help): Recurse help flattening 2023-11-09 15:34:05 -06:00
Ed Page
4bef91ca3c refactor(help): Pull out flat subcommands 2023-11-09 15:34:03 -06:00
Ed Page
c9a7ef06e1 fix(help): Gloss over globals with flatten
When using globals, people tend to make all of the top-level arguments
global and cascading them through would just bloat the output.
2023-11-09 15:33:18 -06:00
Ed Page
9e5f93d43f fix(help): Be consistent in long/short help 2023-11-09 15:33:18 -06:00
Ed Page
66d2bcbdd4 feat(help): Allow flattening help 2023-11-09 15:32:57 -06:00
Ed Page
caf5cdcfa0 feat(help): Allow flattening usage 2023-11-09 13:31:22 -06:00
Ed Page
a1fd922bda feat(help): Allow controlling flattening 2023-11-09 13:31:11 -06:00
Ed Page
2142b136b8 refactor: Pull out usage name fallback 2023-11-09 13:27:59 -06:00
Ed Page
4b60cefbf1 refactor: Pull out bin name fallback 2023-11-09 13:26:20 -06:00
Ed Page
c1c2e95ab6 refactor(help): Pull out subcommand usage logic 2023-11-09 12:52:08 -06:00
Ed Page
d9685bcc71 refactor(help): Bypass outer usage layer 2023-11-09 12:52:08 -06:00
Ed Page
25e3a87d3c refactor(help): Consolidate arg line usage generation 2023-11-09 12:14:25 -06:00
Ed Page
83981a7f7b refactor: Pull out arg usage line 2023-11-09 12:08:24 -06:00
Ed Page
0c668c3915 refactor(help): Pull out bin name logic 2023-11-09 12:03:35 -06:00
Ed Page
7472aba6ce refactor(help): Clarify usage condition 2023-11-09 12:00:52 -06:00
Ed Page
676e934a61 refactor(help): More directly write args 2023-11-09 11:57:15 -06:00
Ed Page
acdd2c33f0 refactor(help): Make usage argument order consistent 2023-11-09 11:50:46 -06:00
Ed Page
da98eb1a05 refactor(help): Pull out top-level usage logic 2023-11-08 11:49:07 -06:00
Ed Page
f50800f763 refactor(help): Only trim the end on usage 2023-11-08 11:42:54 -06:00
Ed Page
f04f04111b refactor(help): Switch usage creation to writing 2023-11-08 11:25:43 -06:00
Ed Page
6f7174cc7c refactor(help): Pull out usage separator 2023-11-08 11:10:55 -06:00
Ed Page
28425e484f style(docs): Resolve warnings 2023-11-07 13:44:41 -06:00
Ed Page
9bfa5a338c chore: Release 2023-10-24 13:25:07 -05:00
Ed Page
38b5a2f956 chore: Release 2023-10-24 11:45:21 -05:00
Alex Crichton
9a9aabc178 refactor: Reduce code size of testing tokens if they're a number
This commit is a tiny win in compiled code size of a final binary
including `clap` which shaves off 19k of compiled code locally.
Previously tokens were checked if they were a number by using
`.parse::<f64>().is_ok()`, but parsing floats is relatively heavyweight
in terms of code size. This replaces the check with a more naive "does
this string have lots of ascii digits" check where the compiled size of
this check should be much smaller.
2023-10-24 09:17:45 -07:00
Kevin Reid
71c1e59334 docs: Fix doc link to Arg::trailing_var_arg 2023-10-14 18:11:33 -07:00
Ed Page
dfebb54423 style: Make clippy happy 2023-10-12 08:32:59 -05:00
Josh Triplett
416b1a8da7 fix: Remove unused dev-dependencies
These dependencies are used elsewhere in clap, but not in these crates.
`cargo test` and `cargo check --examples` both still pass with these
dependencies removed.
2023-10-10 20:11:33 +08:00
Ed Page
29f22c193c
Merge pull request #5160 from epage/docs
docs: Provide custom version/help flag examples
2023-10-04 17:01:16 -05:00
Ed Page
da7ebca04e docs: Provide custom help flag example 2023-10-04 16:25:31 -05:00
Ed Page
e74864f88a docs: Provide custom version flag example 2023-10-04 16:22:46 -05:00
renovate[bot]
4054a0c1a5
chore(deps): update compatible (dev) (#5151)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2023-10-01 01:52:25 +00:00
Ed Page
39f5e807af chore: Release 2023-09-28 14:36:28 -05:00
Ed Page
485b957c4b chore: Upgrade anstream 2023-09-28 14:13:54 -05:00
Ed Page
c298f6a52c chore: Release 2023-09-25 15:59:01 -05:00
Ed Page
3ac44040ef
Merge pull request #5025 from SUPERCILEX/resolve-alias-conflicts
fix: Resolve conflicting name inference if from aliases
2023-09-25 15:56:28 -05:00
Alex Saveau
a76789eb8b
fix: Make long subcommand flag inference consistent
Signed-off-by: Alex Saveau <saveau.alexandre@gmail.com>
2023-09-25 13:28:27 -07:00
Alex Saveau
c2b8ec3bd3
fix: Resolve conflicting name inference if from aliases
Signed-off-by: Alex Saveau <saveau.alexandre@gmail.com>
2023-09-25 13:18:54 -07:00
Ed Page
221177b9cb fix(assert): Call out the action in positional assert
Brought up in #5135
2023-09-25 10:18:37 -05:00
Ed Page
c344edf8b9 docs: Clarify most asserts are for debug builds 2023-09-19 12:38:21 -05:00
Ed Page
e6e539660f chore: Release 2023-09-18 09:28:47 -06:00
Colin Walters
84f99ff979 chore(builder): Bump terminal_size to 0.3
For some reason, terminal_size bumped semver recently; they
also bumped the rustix version to 0.38.

I'd like to reduce the number of copies of OS integration
crates (rustix and nix) in my dependency chain.  Let's bump
to the new version so I can aim to drop rustix 0.37.
2023-09-18 10:31:19 -04:00
renovate[bot]
dc63cba772
chore(deps): update compatible (dev) (#5108)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2023-09-01 01:47:45 +00:00
Ed Page
3cd571e654 chore: Release 2023-08-31 13:19:11 -05:00
Ed Page
7a4f7d3f2c
Merge pull request #5106 from epage/once
perf: Drop a dep for faster builds
2023-08-31 13:18:15 -05:00
Ed Page
087224a486 perf: Drop a dep for faster builds
`OnceLock` became available as of 1.70 which is older than our new MSRV.
We can easily get away without `Lazy` variants.
2023-08-31 12:54:45 -05:00
Joe Chen
4012b195fd fix(builder): Update documentation comments 2023-08-31 13:22:35 +08:00
Ed Page
7126f78421 chore: Release 2023-08-28 11:45:19 -05:00
Ed Page
eb53db89f8 feat(help): Stabilize styling support
Fixes #3234
2023-08-28 09:49:48 -05:00
Ed Page
2e49645418 chore: Release 2023-08-24 11:08:02 -05:00
Ed Page
0b196a165c perf(builder): Speed up builds by dropping is-terminal 2023-08-24 10:19:28 -05:00
Ed Page
88cf306216 chore: Release 2023-08-23 09:21:22 -05:00
Ed Page
bf3f25ebb2 fix(help): Ensure padding isn't stripped
Fixes #5083
2023-08-22 16:02:33 -05:00
Ed Page
b9df80c2ce chore: Release 2023-08-18 16:07:31 -05:00
Ed Page
56135f3ff3 fix(builder): UnknownValueParser shouldn't error on flag absense
Fixes #5079
2023-08-18 14:29:37 -05:00
Ed Page
6720240577 feat(parser): Report source to value parsers 2023-08-18 14:28:15 -05:00
Ed Page
df337de701 chore: Release 2023-08-17 08:59:45 -05:00
Ed Page
9f65eb0c9a refactor(error): Give caller control over suggestion 2023-08-16 14:54:45 -05:00
Ed Page
8413c155d2 feat(builder): Allow injecting known unknowns
Fixes #4706
2023-08-16 14:50:08 -05:00
Ed Page
ee1388c0a3 chore: Release 2023-08-07 20:36:44 -05:00
Ed Page
3156e1a35c
Merge pull request #5066 from 9999years/export-trymapvalueparser
feat: Allow referencing `TypedValueParser` in`ValueParserFactory` associated type
2023-08-07 20:35:49 -05:00
Ed Page
b96cbafcbe chore: Release 2023-08-07 20:33:13 -05:00
Rebecca Turner
c86172e891
feat: Export builder::TryMapValueParser
Partial fix for #5065, allows using TypedValueParser::try_map to
implement a ValueParserFactory.
2023-08-03 15:53:38 -07:00