Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!mailrus!tut.cis.ohio-state.edu!bloom-beacon!athena.mit.edu!wesommer
From: wesommer@athena.mit.edu (William Sommerfeld)
Newsgroups: comp.unix.wizards
Subject: Re: grep replacement
Keywords: Long Lines
Message-ID: <5742@bloom-beacon.MIT.EDU>
Date: 12 Jun 88 08:17:26 GMT
References: <7882@alice.UUCP> <5630@umn-cs.cs.umn.edu> <6866@elroy.Jpl.Nasa.Gov> <2312@bgsuvax.UUCP> <7869@ncoast.UUCP> <144@insyte.UUCP>
Sender: daemon@bloom-beacon.MIT.EDU
Reply-To: wesommer@athena.mit.edu (William Sommerfeld)
Organization: Massachusetts Institute of Technology
Lines: 27

In article <144@insyte.UUCP> jad@insyte.UUCP writes:
>A missing feature in UNIX is the ability to deal with files with
>very long lines - the kind of file you get from a data base tape like
>Compustat.  The standard data base tape contains very long lines.
>Instead of separating each record with a newline, all the records may
>be on the same line.  There is a defined record size which is used to
>determine when a record ends - instead of a newline.
>
>There are two specific things that it would be nice to do with UNIX, 
>instead of having to write a c program:

As usual with UNIX, you _don't_ have to write a C program.  Use `dd'
instead.  If we're talking about the canonical IBM "80 column card
image", then the following should work just fine to convert it to a
"normal" file:

dd conv=unblock cbs=80 <in >out

Adding conv=ascii will also convert EBCDIC into ascii.

>2) To be able to specify a field range (ie columns 20-30) for the 
>program to search - instead of the entire line/record.  

grep '^........foo' 

picks up any line which has `foo' in columns 9-11..

					- Bill