Very Computer

Help,

I am running Solaris 2.6, and have only ever written scripts for the Korn
Shell......and have the following query......

I have a very long text file ( each line represents a CAD model name ).
Somewhere within the model name (i.e. line of the text file ) , is  the
drawing reference number......this number will always match a particular
pattern of letters, hyphens and numbers - which will not be matched
elsewhere within the remainder of the line.
I need a mechanism to search each line in turn, match the pattern of
letters, hyphens and numbers and extract purely that portion of the line and
write it to a separate file. My problem , is that the pattern could appear
anywhere within each line and the number of fields can differ ( if it was
always in the same place or had a constant number of fields, I would use
'cut' )
For example, an extract from the input file could be :-

LD100 WEIGHT DAMPER              W7A   22 Z  535210043-A  USE30SE99MVS
99/10/26
LD100 GROMMET  GUIDE  ROD          W5D   12 Z  545180058-B  USE30SE99MAK
99/10/26
LD100 GROMMET  GUIDE  ROD          W5D   22 Z  545180058-B  USE30SE99MAK
99/10/26
LD100 BRACKET-CABLE,SELECT       W7A   12 Z  535210034-C  USE06SE99LJL
99/10/26
LD100 BRACKET-CABLE,SELECT       W7A   22 Z  535210034-C  USE06SE99LJL
99/10/26

And I need to extract the data matching the following pattern: nine
numbers,a hypen and a letter and put it in an output file - thus the output
I would like to see is :-

535210043-A
545180058-B
545180058-B
535210034-C
535210034-C

Many thanks in advance for your help

Mark Hounslow
LDV Limited

sed -n 's/^.*\([0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9][0-9]-[A-Z]\).*/\1/p'

--

GTE Internetworking, Powered by BBN, Burlington, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.

gawk --posix '{ for (i=1; i<=$NF; i++) if ($i ~ /^[0-9]{9}-[A-Z]$/) print $i }'

I expect someone will provide a shorter and even more efficient one in
perl ...

N.

An answer in perl (which could be squashed to one line if you need to) is:

#!/usr/bin/perl
while (<>)
{
   if(/(^| +)([0-9]{9}\-[A-Z])( +|$)/){
     print "$2\n"
   }

Quote:}

This is case sensitive on the A-Z. You can change the if statement to
   /(^| +)([0-9]{9}\-[A-Z])( +|$)/i
(the extra i on the end) to make things case insensitive.
The expression above also works if the thing your looking for is the first
or last field.

Things can be made even more generic by matching on type of character
(\s for space and \d for digit)

#!/usr/bin/perl
while (<>)
{
   if(/(^|\s+)(\d{9}\-[A-Z])(\s+|$)/){
     print "$2\n"
   }

Quote:}

 - Richard.

--

  _/  _/    _/    _/      Web Page, CV:   http://www.littondale.freeserve.co.uk
 _/_/      _/    _/       Dance (Ballroom, RnR), Hiking, SJA, Linux, ... [ENfP]
_/  _/  _/_/    _/_/_/    PGP2.6 Key ID: 0x0FB084B1     PGP5 Key ID: 0xFA139DA7

Less typing.
sed -n 's/^.*\([0-9]\{9\}-[A-Z]\).*/\1/p' inputfile > outputfile

--

|Linux: Delivers on the promises Microsoft make. | The Zeppelin of   |
|             http://www.linux.org/              | operating systems.|