Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.networking > #2071

Re: Mental-model of multiprocessing?

Cross-posted to 2 groups.

Show all headers | View raw

In comp.os.linux.misc Avoid9Pdf@gmail.com wrote:
> Which you would never write [nor echo "hello" | cat | cat | cat ]

As CFAJ mentioned, this would work. Follow the (simplified) set of rules
I gave you and you'll see that they apply: the "hello" is written the
first cat, which writes to the second, which writes to the third. That
final cat writes to the terminal because no other output target has
been specified.


> So then:
>  cat>/tmp/mesgFile
> (
>    read line 0<&3 ; echo "$line">$log
>    case "$line" in 2*) ;; *) echo "QUIT" 1>&3 ; exit 0 ;; esac
>    echo "HELO $local_name" 1>&3
> ) 3<>/dev/tcp/$smtp_server/25  

> would mean 
> "stdin [since none is provided]" writeTo /tmp/mesgFile
> the sequential output of the bracketed-statements
> [which also write to there OWN destinations]
> .... ??

No.

The shell applies redirection for stdout to /tmp/mesgFile. Cat is then
invoked, and it reads from stdin since no files are present on the command
line, and writes to its stdout. Remember that the shell has already
redirected stdout to a file, so cat effectively writes to the file. But
cat doesn't know this - all it knows is that it's writing to its stdout.

The bit in brackets is grouped as a single item. But before it's
executed the shell opens fd 3 and attaches its input/output to the
network socket tcp/25.

Now as you work through the lines inside the brackets you have this
additional descriptor available to you.


> Does: 3<>/dev/tcp/$smtp_server/25
> mean: let fd3 have input AND output to that-device?
> 
> *THAT* is the 'fraud'(:-;) !!
> Only AFTER it's been used is it defined.

No. The shell applies redirection before the command (or subshell)
is executed. This is all documented in the man page.


> It would be good if the *nix docos put UP FRONT that
> 'it's not parsed sequentially left-to-right'; like computers
> work or mathematical theorems are proved.

It's in the shell man page under the heading REDIRECTION. This is
referenced several times from the section SHELL GRAMMAR that explains
the basic shell syntax.

Chris

Back to comp.os.linux.networking | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

Mental-model of multiprocessing? Avoid9Pdf@gmail.com - 2013-03-27 11:13 +0000
  Re: Mental-model of multiprocessing? Joe Beanfish <joebeanfish@nospam.duh> - 2013-03-27 14:02 +0000
    Re: Mental-model of multiprocessing? Richard Kettlewell <rjk@greenend.org.uk> - 2013-03-27 14:13 +0000
      Re: Mental-model of multiprocessing? The Natural Philosopher <tnp@invalid.invalid> - 2013-03-27 14:35 +0000
        Re: Mental-model of multiprocessing? Chris Davies <chris-usenet@roaima.co.uk> - 2013-03-27 15:30 +0000
          Re: Mental-model of multiprocessing? The Natural Philosopher <tnp@invalid.invalid> - 2013-03-27 18:57 +0000
        Re: Mental-model of multiprocessing? Richard Kettlewell <rjk@greenend.org.uk> - 2013-03-27 21:15 +0000
          Re: Mental-model of multiprocessing? The Natural Philosopher <tnp@invalid.invalid> - 2013-03-27 21:36 +0000
          Re: Mental-model of multiprocessing? Aragorn <stryder@telenet.be.invalid> - 2013-03-28 06:56 +0100
      Re: Mental-model of multiprocessing? Jorgen Grahn <grahn+nntp@snipabacken.se> - 2013-03-27 21:47 +0000
    Re: Mental-model of multiprocessing? Avoid9Pdf@gmail.com - 2013-03-29 14:28 +0000
      Re: Mental-model of multiprocessing? The Natural Philosopher <tnp@invalid.invalid> - 2013-03-29 15:32 +0000
        Re: Mental-model of multiprocessing? J G Miller <miller@yoyo.ORG> - 2013-03-29 22:36 +0000
      Re: Mental-model of multiprocessing? Chris Davies <chris-usenet@roaima.co.uk> - 2013-03-29 17:31 +0000
        Re: Mental-model of multiprocessing? Avoid9Pdf@gmail.com - 2013-03-29 21:10 +0000
          Re: Mental-model of multiprocessing? "Chris F.A. Johnson" <cfajohnson@gmail.com> - 2013-03-29 18:55 -0400
            Re: Mental-model of multiprocessing? Unknown <dog@gmail.com> - 2013-03-30 06:01 +0000
          Re: Mental-model of multiprocessing? Chris Davies <chris-usenet@roaima.co.uk> - 2013-03-30 21:28 +0000
          Re: Mental-model of multiprocessing? Joe Beanfish <joebeanfish@nospam.duh> - 2013-04-01 15:32 +0000
            Re: Mental-model of multiprocessing? Unknown <dog@gmail.com> - 2013-04-03 08:00 +0000
              Re: Mental-model of multiprocessing? Joe Beanfish <joebeanfish@nospam.duh> - 2013-04-08 14:47 +0000
            Re: Mental-model of multiprocessing? Unknown <dog@gmail.com> - 2013-04-03 08:02 +0000

csiph-web