Find the matching word in the same file using SORT .

milind suman · New User Joined: 19 Aug 2009 Posts: 55 Location: Pune

Hi,

I have a requirement where i have to identify if a string from 1st column of a file exist anywhere in the same file between column 10 to 80 . How can it be done with DFSORT ..kindly help . The requirement is as follows :

1)Compare the string(word) starting from 1st column for a matching word in the same file in any record between column 10 to 80 ( The word can be a length anywhere between 5 to 10 character )

2) If there is no matching word found than write the record in an output file .

Thanks a lot !

enrico-sorichetti · Posted: Mon Jan 18, 2010 4:51 pm

wondering about the business requirement

Frank Yaeger · Posted: Mon Jan 18, 2010 11:27 pm

It's not clear what you want to do.

Please show an example of the records in your input file (relevant fields only) and what you expect for output. Explain the "rules" for getting from input to output. Give the starting position, length and format of each relevant field. Give the RECFM and LRECL of the input files. If the input file can have duplicates within it, show that in your example.

milind suman · New User Joined: 19 Aug 2009 Posts: 55 Location: Pune

I have to check the word in 1st column of file , If it exist anywhere in the same file then the record should be escaped else I have to write it in an output file .

For example the word VAR123 exists in the file and hence I dont want it in o/p file .

2) Record starting with spaces in column 1 has to be escaped..dont write in o/p file

3) Third record 1st word is VAR9 and it does not exist anywhere in the file . so it has to be written in o/p file .

4) The variables length is not fixed , it can be either delimited by comma , semilcolon , spaces or any special character .

5) File is FB , recrord length = 80

Please let me know if I miss anything to mention .

Frank Yaeger · Posted: Tue Jan 19, 2010 10:36 pm

What is the maximum number of words you will be looking for (10, 100, 1000?).

Does each search word end before the first blank? That is, you would search for 'VAR123' and 'VAR9'? No embedded blanks?

What is the maximum length of a search word?

milind suman · New User Joined: 19 Aug 2009 Posts: 55 Location: Pune

Hi Frank ,

I am rewording my requirement :

Look for the first word of a record ( starting from column 1) in the whole file (between column 10-80) if there is no hit then dump the record in output file .

The maximum number of words in the file would be around 500.
Each search word ends before blank . No embedded blank .
The maxm length of each search word is 10 .

Input file :

Frank Yaeger · Posted: Tue Jan 19, 2010 11:26 pm

Hmmm ... offhand, I don't see a good way to do this with built-in features.

I'd suggest you write a program or exit to do it.

dick scherrer · Posted: Tue Jan 19, 2010 11:59 pm

Hello,

Do you really mean the "whole file"? Or only those records associated with the records having a value beginnnig in position 1?

What if this line:
hoffman steve cameron germany
contained:
hoffman steve james germany
?

What if some record containes jamestown? And so on. . . .

This is surely more suited to code, but even with code, the requirement needs to be completely understood before coding begins. . .

milind suman · New User Joined: 19 Aug 2009 Posts: 55 Location: Pune

word has to be searched in the whole file ( record 1 to end of file and between coulmn 10 to 80)

jamestown can qualify for match to james .