On 2/12/2008 8:18 AM, Minalba Jordao wrote:
> On 12 fev, 11:33, Ed Morton <mor...@[EMAIL PROTECTED]
> wrote:
>
>>On 2/12/2008 7:23 AM, Minalba Jordao wrote:
>>
>>
>>
>>
>>>On 12 fev, 11:15, Ed Morton <mor...@[EMAIL PROTECTED]
> wrote:
>>
>>>>On 2/12/2008 6:57 AM, Minalba Jordao wrote:
>>>
>>>>>Hello!
>>>>
>>>>>I am a newbie.
>>>>
>>>>>I have a few problems.
>>>>
>>>>>I have a file that looks like:
>>>>>"
>>>>>! comments
>>>>>! comments
>>>>>! comments
>>>>
>>>>>@[EMAIL PROTECTED]
text
>>>>>X
>>>>>Y
>>>>>Z
>>>>>@[EMAIL PROTECTED]
>>>>
>>>>>column1 column2 column3 column4
>>>>>column1 column2 column3 column4
>>>>>... ... ...
>>>>>... ... ...
>>>>>column1 column2 column3 column4
>>>>>column1 column2 column3 column4
>>>>>"
>>>>
>>>>>My questions:
>>>>>I would to get the
>>>>>- "! coments"
>>>>>- what is between the @[EMAIL PROTECTED]
(at sign).
>>>>>and
>>>>>column1 column2 -column3
>>>>>for each line..
>>>>
>>>>>could anyone help me?
>>>>
>>>>>Mina
>>>>
>>>>This MAY be what you want, but it's not clear:
>>>
>>>>awk '
>>>>/^@[EMAIL PROTECTED]
{ a=!a; b=1; next }
>>>>a || /^!/ { print; next }
>>>>b { print $1,$2,$3 }
>>>>' file
>>>
>>>>If not, post some sample input and expected output.
>>>
>>>> Ed.
>>>
>>>that worked!
>>>But it did`t include the @[EMAIL PROTECTED]
sign.
>>>' comments
>>>! comments
>>>! comments
>>>!
>>> X
>>> Y
>>> Z
>>>1865.16833 6639.33984 -3278.68018
>>>1865.21423 6539.07471 -3284.02954
>>>1889.54529 6414.59814 -3287.10986
>>
>>>and it should be like
>>>' comments
>>>! comments
>>>! comments
>>>!
>>>@[EMAIL PROTECTED]
>>> X
>>> Y
>>> Z
>>>@[EMAIL PROTECTED]
>>>1865.16833 6639.33984 -3278.68018
>>>1865.21423 6539.07471 -3284.02954
>>>1889.54529 6414.59814 -3287.10986
>>
>>See my other post in this thread after you'd clarified that requirement
in a
>>followup to your original post.
>>
>> Ed.
>
>
> Ed,
> It did work. Thank you.
> But I didn`t understand acctually what you did.
>
> In the first line, you look for the @[EMAIL PROTECTED]
sign in the begining of of the
> line, then you print what you search. OK, then a=!a and b=1, I didn`t
> understand what does this means.
I'm just setting a couple of flags which I use later to print the records
when
those flags are set. When the script starts "a" has the value "zero" (or
"null"
if it were treated as a string) so when the first "@[EMAIL PROTECTED]
" is found, "a=!a"
sets a to
"1" so that on the "a || ..." line, "a" is set and so the print occurs.
WHen
the second "@[EMAIL PROTECTED]
" is found later in the file, "a=!a" flips the value of "a"
back to
zero so that on the "a || ..." line the print does not occur (at least,
not due
to being insid the "@[EMAIL PROTECTED]
" section). Setting "b=1" just makes sure that
after
the final "@[EMAIL PROTECTED]
" is seen, printing of the first 3 fields will occur at the "b
{
...." lines of the script.
> /^@[EMAIL PROTECTED]
{ print; a=!a; b=1; next }
> a || /^!/ { print; next }
> b { print $1,$2,"-"$3 }
>
> What can I read to understand it better?
The archives of this NG.
> Thank you again.
You're welcome.
> If I need to to that, with many files? Could I do it with awk? Or will
> I need to use some shell scripting?
You can do it with awk:
awk ' FNR==1 { outfile=FILENAME ".new" }
/^@[EMAIL PROTECTED]
{ print > outfile; a=!a; b=1; next }
a || /^!/ { print > outfile; next }
b { print $1,$2,"-"$3 > outfile}' *.dat
but your OS is typically better suited for the job, especially if you want
to
replace the original files with the modified ones.
> $
> for file in *.dat; do ./trans.sh $file > $file.new; done
>
> transf.sh
> #!/bin/sh
> awk ' /^@[EMAIL PROTECTED]
{ print; a=!a; b=1; next }
> a || /^!/ { print; next }
> b { print $1,$2,"-"$3 }' $1
>
>
> Any suggestions?
Anything other than the awk language is OT for this NG so you might get
better
help on the shell script at comp.unix.shell, but to start with, always
quote
your shell variables unless you know exactly what you're doing and have a
very
specific reason not to.
Ed.
|
