[9fans] handling output

[9fans] handling output

[hugo rivera]

I've been wondering for a while if there's some way to multiplex (if
this is the correct term) stdout for a given program:
% ls @ {grep regexp1 > file1 } @ {grep regexp2 > file2}
where @ is an operator that would copy ls stdout to two (maybe more)
different file descriptors. Probably some syntax is required
(something like @[1=3]).
I think a c program can be written to accomplish this, but I'd like to
hear your opinion about it. Am I on the right path or just talking
rubbish?
Saludos

--
Hugo

Re: [9fans] handling output

[roger peppe]

2009/10/1 hugo rivera <uair00@gmail.com>:
> I've been wondering for a while if there's some way to multiplex (if
> this is the correct term) stdout for a given program:

that's what tee does.

e.g.
ls | tee >{grep regexp1 > file1} >{grep regexp2 > file2}

Re: [9fans] handling output

[hugo rivera]

Great, thanks.
Looks like plan 9 guys have thought about everything useful ☺ (and
that I didn't do my homework).

2009/10/1, roger peppe <rogpeppe@gmail.com>:
> 2009/10/1 hugo rivera <uair00@gmail.com>:
>
> > I've been wondering for a while if there's some way to multiplex (if
>  > this is the correct term) stdout for a given program:
>
>
> that's what tee does.
>
>  e.g.
>  ls | tee >{grep regexp1 > file1} >{grep regexp2 > file2}
>
>


-- 
Hugo

Re: [9fans] handling output

[matt]

Tee part of the POSIX standard

http://www.opengroup.org/onlinepubs/9699919799/utilities/tee.html


hugo rivera wrote:

>Great, thanks.
>Looks like plan 9 guys have thought about everything useful ☺ (and
>that I didn't do my homework).
>
>2009/10/1, roger peppe <rogpeppe@gmail.com>:
>
>
>>2009/10/1 hugo rivera <uair00@gmail.com>:
>>
>>
>>
>>>I've been wondering for a while if there's some way to multiplex (if
>>>
>>>
>> > this is the correct term) stdout for a given program:
>>
>>
>>that's what tee does.
>>
>> e.g.
>> ls | tee >{grep regexp1 > file1} >{grep regexp2 > file2}
>>
>>
>>
>>
>
>
>
>

Re: [9fans] handling output

[hugo rivera]

shame on me, I didn't know about it.

2009/10/1, matt <maht-9fans@maht0x0r.net>:
> Tee part of the POSIX standard
>
> http://www.opengroup.org/onlinepubs/9699919799/utilities/tee.html
>
>
>
>  hugo rivera wrote:
>
>
> > Great, thanks.
> > Looks like plan 9 guys have thought about everything useful ☺ (and
> > that I didn't do my homework).
> >
> > 2009/10/1, roger peppe <rogpeppe@gmail.com>:
> >
> >
> > > 2009/10/1 hugo rivera <uair00@gmail.com>:
> > >
> > >
> > >
> > > > I've been wondering for a while if there's some way to multiplex (if
> > > >
> > > >
> > > > this is the correct term) stdout for a given program:
> > >
> > >
> > > that's what tee does.
> > >
> > > e.g.
> > > ls | tee >{grep regexp1 > file1} >{grep regexp2 > file2}
> > >
> > >
> > >
> > >
> >
> >
> >
> >
>
>
>


-- 
Hugo

Re: [9fans] handling output

[Sam Watkins]

I'd like to make a "graphical shell" which could do this sort of thing, it
would also have a textual format for saving and suitable for editing.

I think it would be good to have an equivalent graphical tool for each command
line tool, as powerful as the original one; and an editable graphical
representation of each file format, file-based interface or protocol.

A nodes-and-arcs graph representation is already parsed and more symmetric that
a textual language.

Sam

On Thu, Oct 01, 2009 at 10:26:06AM +0200, hugo rivera wrote:
> I've been wondering for a while if there's some way to multiplex (if
> this is the correct term) stdout for a given program:
> % ls @ {grep regexp1 > file1 } @ {grep regexp2 > file2}
> where @ is an operator that would copy ls stdout to two (maybe more)
> different file descriptors.

Re: [9fans] handling output

[Richard Miller]

> I've been wondering for a while if there's some way to multiplex (if
> this is the correct term) stdout for a given program:
> % ls @ {grep regexp1 > file1 } @ {grep regexp2 > file2}

ls | tee >{grep regexp1 >file1} | grep regexp2 >file2