I am attempting to grep for terms that contain letters followed by a single digit. Example svls3 or svls4. I've tried 'egrep 'svls[1-9]' but it returns any 'svls' sequence with any numerical characters such as svls16 or svls59. It even returns svls60 and I don't have a 0 in my expression.

According to the docs I've read "A range of single-digit numbers can be specified as: [0-9]

Obviously I'm doing something wrong but I can't get it.

Maybe try egrep 'svls[0-9]\b'
\b is word boundary

Your problem is that grep says "true" and prints the line as soon as it sees 's' 'v' 'l' 's' followed by a number between 1 and 9.

So, the regular expression engine starts at the left. It sees 's-', then 'sl-', then 'svl-', then 'svls', then 'svls1', at which point "Ding ding ding, we have a winner!"

Then it prints the line. It doesn't matter if another number follows, since you didn't tell it anything about what has to happen next. As Keefaz says, you could add \b for a word boundary or \s for whitespace or $ to specify that the item needs to be at the end of a line or...

Thanks for all the suggestions. Some docs I had suggested putting a space at the end of the expression: egrep "svls[1-9] ". This did not work for me because the target was at the end of the line. So I changed the line input which was easy since it was being piped through awk. I put the target in the middle of the line. Problem solved.

Out of curiosity I went back and tried with the \b with and the target at the end of the line. It work like a charm. The $ character did not work: egrep "svls[1-9]$" returns 'Illegal variable name.'

Thanks again for the help.

Single quotes are required to protect "$" from being interpreted by the shell instead of by GREP. So this should work:

egrep 'svls[1-9]$'

Moin,

a way to do it is to tell grep to look for "svls" + 1 digit, followed by a non digit or linefeed:
I used extended regex (option -E), that makes things easier.

Jan

My KDE menu has an application KRegExpEditor.
Just now for the first time I had some time to understand it (there is help) and have constructed some regular expressions.

I remember that in the past I have lost hours because of incorrect regular expressions (say in apache config files etc.). When there are so many parts in your task that could go wrong then it is helpful if you can isolate smaller parts.

I am confident that with KRegExpEditor I will get at last the regular expressions right the first time.
I am grateful to the developers of this program.