My point was, if #shellCommand: accepts a single string for a whole
command, then surely it passes it to a system shell already, so why nest a
bash in between? (effectively running the equivalent of sh -c "bash -c \"ls
~\"')

On 15 May 2017 at 16:28, Eliot Miranda <[email protected]> wrote:

> Hi Damien,
>
> On May 15, 2017, at 6:44 AM, Damien Pollet <[email protected]>
> wrote:
>
> On 15 May 2017 at 15:26, Eliot Miranda <[email protected]> wrote:
>
>> Try something like
>>
>>         shellCommand: 'bash -c ''ls ~''';
>>
>
> But then that would run ls inside of bash inside of the system shell
> (/bin/sh), wouldn't it? What's the point?
>
>
> -c's argument is parsed by the shell so one gets full expansion.  Further,
> if there are arguments after the string, they are assigned to the
> positional parameters, so that
>
>     sh -c 'echo ~/$0' foo
>
> prints /Users/eliot/foo
>
> That may be useful.
>
>
> --
> Damien Pollet
> type less, do more [ | ] http://people.untyped.org/damien.pollet
>
>


-- 
Damien Pollet
type less, do more [ | ] http://people.untyped.org/damien.pollet

Reply via email to