Re: how to output data as null separated

[TSHIMANGA Minkoka]

Hello Assaf,

You are right, the sed option -z also sets the null as record separator
in the output.

The documentation is not explicit on it, I do not know what distracted
me to not notice that while I was testing.

Thanks.

Tshimanga

On Sat, 2019-09-14 at 01:12 -0600, Assaf Gordon wrote:
> Hello,
> 
> On 2019-09-11 1:35 a.m., TSHIMANGA Minkoka wrote:
> > I am aware of the sed -z option, that reads input with null
> > separated
> > "lines".
> > But how can I also similarly output null separated "lines".
> 
> The "-z" option does exactly that:
> uses NUL as *both* input and output line separator.
> 
> > My need is somehow a filenames piped processing, as outlined below.
> > 
> > find /path -print0 | sort -z | sed -z --
> > needed_sed_option_to_output_null_separated_lines -e 'sed process 1'
> > -e 'sed process 2' | perl -n0 script.pl
> 
> The about should "just work" (at least as far as sed is concerned).
> 
> For example: the following uses "sed" to add a number before each
> line 
> (still, NUL separated).
> Then perl reads each line and prints it.
> Note that perl's "-0" indeed only affects the input separator
> (unlike 
> sed's "-z"):
> 
>       find . -print0 | sort -z | sed -z = \\
>             | perl -n -0 -e 'print "line = $_\n";'
> 
> regards,
>   - assaf
>