Arguments are learned as-they-go: the cli parsing framework learns each argument upon checking it. For this reason, in order to maximize effectiveness of the parsing it is recommended to parse arguments in the following order:
- flags
- options
- positionals
- subcommands
Note: For a type to be accepted by the command-line it must implement the
std::str::FromStrtrait.
-
long options (flags/options)):
--verbose -
short options (switches):
-v -
positional value placement:
--output a.out,-o a.out -
attached value placement:
--output=a.out,-o=a.out -
combine switches:
-rf -
capture variable flag calls:
--inc --inc --inc... -
capture variable option calls:
--op 10 --op 20 --op 30 -
capture variable positional calls:
10 20 30 -
capture variable limited user-defined caps for flag and option calls:
--verbose --verbose --verbose(ex: up to 3) -
combine switches and give value to final switch:
-rfo=a.out -
ability to prioritize help over any other error
-
ability to enable/disable a help flag with custom long/short flag names
-
subcommand support:
calc add 10 20,calc mult 10 20 -
arguments can be reused in both structs for overall command and nested subcommand
-
zero dependencies (use the
colorfeature for coloring) -
uses dynamic programming sequence alignment algorithm to detect misspelled arguments (with configurable
thresholdfor string comparison) -
ability to set user-defined help text before parsing
-
ability to verify there are no unused arguments passed in the cli
-
verifies all argument type conversions are accepted during parsing
-
preserves unprocessed arguments found after an empty flag
--symbol
See the examples/ folder for practical implementations.