Den 2011-02-17 19:18 skrev Stefano Lattarini:
> Hello Peter.
>
> On Thursday 17 February 2011, Peter Rosin wrote:
>> Den 2011-02-15 21:19 skrev Ralf Wildenhues:
>>> It's a bit of a shame this uses a temp file though. awk
>>> should be able to cope without (but be sure to try old
>>> Solaris awk if you go for it).
>>
>> Ahh, who needs awk? :-)
>>
>> [CUT]
>>
> Wow, this seems really clever! (I say "seems" because I must
> admit I haven't tried to read it in detail, nor to really
> understand it).
I think the terseness of the sed language is the biggest
difficulty.
> However ...
>
>> Too much of a maintenance hassle? You tell me.
>>
> ... I do think it's really too much of a maintenance hassle
> *for the use of the function we expect to do*. If we were
> writing a library of general-purpose shell functions, I'd
> definitely choose your implementation over mine, but in the
> present situation I see no real advatange. Sorry.
It's that thing with parallel execution, but I can agree that
there are no current users needing that.
Anyway, when I wrote the program I looked for a way to just
skip ahead to the next cycle, but didn't find how to do it.
Since then, I have realized that 'b' w/o argument does exactly
the thing I was looking for, so a slightly shorter program
follows (I also added a few anchors):
unindent ()
{
sed '
x
:useit
/./{
G
h
s/\n.*$//
x
s/^x\(.*\)\n\1//
s/^.*\n//
b
}
g
/^$/b
s/^\([ ]*\).*/x\1/
b useit'
}
> Anyway, Ralf has the final word here, so let's wait to hear
> from him too.
Yes, I will try my best not to spam the list with more versions of
this baby...
Cheers,
Peter
