grep or sed to find a integer value above a given amount?

Question

This is in reference to another question: How do I use robocopy to list all files over a certain size recursively?

I would like to parse output of a command (or cat a log file) and find a string value, convert it to an integer and then see if that integer value is greater than a given integer.

For instance, given the line:

      *EXTRA File          78223    C:\_Google.Enterprise.Contract.2010-06-01.pdf

I'd like to compare '78223' to '10485760'. Is this possible with grep or sed?

Thanks!

You probably awk, or a scripting language for numerical comparisons. — Zoredache, Sep 02 '11 at 19:48
@JimB : possibly, but Questions [older than 60 days](http://meta.stackexchange.com/questions/10249/what-is-migration-and-how-does-it-work) can't be migrated. — HBruijn, Dec 04 '15 at 10:17

score 14 · Accepted Answer · answered Sep 02 '11 at 19:58

14

Use awk as follows:

$ echo '*EXTRA File     78223    C:\foo.pdf' | awk '$3 > 1048576 {print $0;}'
$ echo '*EXTRA File     78223    C:\foo.pdf' | awk '$3 > 40000 {print $0;}'
*EXTRA File     78223    C:\foo.pdf

answered Sep 02 '11 at 19:58

MikeyB

39,291
10
105
189

Thanks. Not familiar with awk. It looks to be very handy. – brandeded Sep 02 '11 at 20:15

score 4 · Answer 2 · answered Sep 02 '11 at 20:17

4

Also - if you're using a Unix-like userland (like UnxUtils or Cygwin, if you're on Windows), you can use find with the -size parameter to get your file list directly, and then pipe to xargs and do whatever you're trying to do with the selected files.

The general answer to your question (interesting comparisons and other operations) is indeed awk or bash (with bc) or perl - but the specific scenario lends itself to find.

answered Sep 02 '11 at 20:17

mfinni

36,144
4
53
86

Bash can natively perform integer math, check out http://mywiki.wooledge.org/ArithmeticExpression – astrostl Sep 02 '11 at 20:30
find can also -exec rather than piping to xargs. If xargs-like "all arguments in one command" style is preferred, terminate with + rather than ; Check out http://mywiki.wooledge.org/UsingFind#Actions_in_bulk:_xargs.2C_-print0_and_-exec_.2B- :) – astrostl Sep 02 '11 at 20:32

astrostl · Answer 3 · 2011-09-02T20:26:02.267

2

In pure Bash:

while read -r _ _ size _; do ((size > 10485760)) && echo "hit"; done < foo.log

edited Sep 02 '11 at 20:26

answered Sep 02 '11 at 20:19

astrostl

690
4
11

This can be easily modified to report any field(s) or the entire line if there is a match, and/or custom-define the comparison integer at runtime. If you want that, just LMK. – astrostl Sep 02 '11 at 20:22

score 0 · Answer 4 · answered Dec 03 '15 at 01:30

0

Perl can be used similarly to awk:

echo '*EXTRA File 78223 C:\foo.pdf' | perl -ane 'print if $F[2] > 40000'

answered Dec 03 '15 at 01:30

Chris Koknat

111
3

grep or sed to find a integer value above a given amount?

4 Answers4

Linked