There are many more pattern-matching features yet to be documented.
- easy execution of externals commands with return of stdout, stderr and the exitcode of the process
The run... commands use a very basic repl-telemetry system to output more information to the console than just the return value,
but in a way which makes the return value clear. The telemetry only outputs if the command is the first word on the commandline.
- `run <comand> ...`
(return exitcode of process - and allows process writes to stderr/stdout to appear in console as they occur)
- `runout <command> ...`
- `runout [-n] <command> ...`
(return stdout of process - no output until completion)
- `runerr <command> ...`
- `runerr [-n] <command> ...`
(return stderr of process - no output until completion)
- `runx <command> ...`
- `runx [-n] <command> ...`
(return a dict of stdout stderr exitcode - no output until completion)
The run... commands attempt to make it clear if an called process outputs a trailing newline by displaying a trailing blank line.
The optional `-n` argument can be used to suppress a trailing newline.
`runout -n pwd` is thus **similar** to Tcl's `exec pwd`
For simple cases `exec <command>` is fine - but the equivalent `runout -n <command>` when used in the shell will display exitcode and stderr separately (whilst returning only stdout)
exec will return stdout and stderr together.
If you are on a unix-like platform, or a windows platform which has something like msys2 to provide commands like 'which' and 'grep' in the path:
Try `runx -n which grep nonexistant` vs `exec which grep nonexistant` to see the difference in terms of easy access to what was written to stderr vs stdout.
The run... commands are intended as a supplement for rather than a replacement for Tcl's exec/open.
- namespace browser (contextual - allowing running of commands within the active namespace - analogous to 'cd' for directories)
- `n/` - display child namespaces of current namespace (alias `:/`)
