--- Begin Message ---
Subject: |
grep -o fails to count empty lines (Debain Bug #532541) |
Date: |
Thu, 3 Aug 2017 15:28:14 +0200 |
(Sorry if it's already filed, but I am unable to find it in the
bug archives)
Hi,
I'd like to forward this bug reported to Debian some years ago.
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=532541
It still happens in 3.1.
Quoting the original report:
When grepping for empty lines in a text file, grep works correctly in
most cases:
~$ grep ^$ myfile.txt
~$ grep -n ^$ myfile.txt
11:
15:
19:
21:
22:
However the -o option, which is supposed to return only the matching
parts of the search, fails:
~$ grep -o ^$ myfile.txt
~$ grep -no ^$ myfile.txt
~$
Thanks,
Santiago
--- End Message ---
--- Begin Message ---
Subject: |
Re: bug#27931: grep -o fails to count empty lines (Debain Bug #532541) |
Date: |
Thu, 3 Aug 2017 13:16:52 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.2.1 |
On 08/03/2017 06:28 AM, Santiago R.R. wrote:
the -o option, which is supposed to return only the matching
parts of the search, fails:
It's not failing. It's behaving as documented: -o outputs only nonempty
matches. Otherwise, commands like 'grep -o "a*"' would output a separate
line for each byte in the input. Although this behavior for -o is
longstanding and is documented in the manual, it's not in the grep
--help output so that's an oversight. I installed the attached to fix
grep --help, and am closing the bug report on the GNU side.
Users who want to match empty lines can use 'grep "^$"', which is what
I'd expect them to do anyway (-o would be superfluous there even if it
included empty matches).
0001-doc-improve-o-help.patch
Description: Text Data
--- End Message ---