How do I skip garbage lines from input file

[LittlesasMaster]

Hi There,

i few sample data placed in datalines, I just wanted to skip those lines which has got "name" and "age" values,

i just wanted to read only real data.

I tried this below code, but it's not cool, I am looking for some better options, like #N or / options to skip certain lines, but

I do not know how to use above tools?

data have;
input value $ @;
if value eq "name" then delete;
else; input name $ age;
drop value;
datalines;
name age
Dharmu 20
Vivian 18
name age
Tina 20
Kath 21
name age
John 30
Tim 30
;
run;

Thanks!!

[jklaverstijn]

What's not to like about what you have? I think it's cool enough.

A shorter alternative would be:

data have;
input name $ @;
if value eq "name" then delete;
else; input age;
datalines;
name age
Dharmu 20
Vivian 18
name age
Tina 20
Kath 21
name age
John 30
Tim 30
;
run;

[jklaverstijn]

By the way, your original code has an issue. Be aware that if you hold the input record with a trailing '@' the pointer will not be at the beginning of the record for the next input statement. You will have to reset it:

...
else input @1 name $ age;
...

Hope this helps,

- Jan.

[LittlesasMaster]

Yes Thanks, I figured it out when saw the output, thanks for pointing that out,

but can we use those line pointer to skip the lines, I mean just curius to learn that.

[jklaverstijn]

---

@LittlesasMaster wrote:

but can we use those line pointer to skip the lines, I mean just curius to learn that.

---

Yes you can. Use the forward slash in your INPUT statement. Or simply use another INPUT statement.

See http://support.sas.com/documentation/cdl/en/basess/68381/HTML/default/viewer.htm#p0s16wvzu0z9q7n0zmx...

- Jan

[PGStats]

I would do:

data have;
input @;
if _infile_ eq "name age" then input;
else do;
    input name $ age;
    output;
    end;
datalines;
name age
Dharmu 20
Vivian 18
name age
Tina 20
Kath 21
name age
John 30
Tim 30
;

[Astounding · Accepted solution]

If you are positive of the pattern (one garbage line followed by two good lines), you can use:

data have;
input / name $ age;

output;

input name $ age;

output;
datalines;
name age
Dharmu 20
Vivian 18
name age
Tina 20
Kath 21
name age
John 30
Tim 30
;

But if you don't know the pattern ahead of time, there's no way for you to tell SAS the pattern (through your programming statements) ahead of time.