Problem getting entire record when using grep on a file

[rahill13]

I'm using grep to read a file and strip the header records out and feed it to another file. I'm not getting all of the record; grep is chopping off the tailend of the record. It looks like I only get 250 bytes of the record. Any suggestions???????

grep -v "***TOTAL***" file1 > file2

 

[bjverzal]

Post a few input and output records.

 

[Yegolev]

Not trying to sound un-helpful but maybe awk would work. Alternatively, I think perhaps you should try your line again with single-quotes instead of double-quotes. I like either awk or egrep because of the control that regular expressions give me.

I'm assuming that you are looking for asterisks rather than using them as wildcards. IBM Certified -- AIX 4.3 Administration
IBM Certified -- AIX 4.3 Support
IBM Certifiable!

 

[ECEBOB]

If you're using the built in AIX version of grep, try downloading and installing the Gnu version. I found that other programs such as tar, cpio, etc. had issues with lengthy path names and the Gnu versions of these programs did not have the same issue.

 

[Yegolev]

Good point ECEBOB. rahill13, see earlier postings about line length limits in AIX.

If it is number of records, sed would work fine as it is a streaming sort of tool. Of course it is a bit arcane. IBM Certified -- AIX 4.3 Administration
IBM Certified -- AIX 4.3 Support
IBM Certifiable!

 

[trifo]

Hi!

If the task itself is no more than just filtering out some well known lines. grep will do. Only you have to give grep a correct expression.

As * has special meanings in regular expressions, you have to escape it like this:

grep -v \*\*\*TOTAL\*\*\* file1 > file2

and this way there is no need to use quotes. Or you may use single quotes which will suppress any special meanins of characters.

--Trifo