Prev: wget -r (recursive) failing to back up a site
Next: how to optionally silence lines in script
From: Ed Morton on 8 Jun 2010 14:55
On Jun 8, 11:40 am, gaze...(a)shell.xmission.com (Kenny McCormack)
wrote:
> In article <d8cbd965-3be1-4c65-b97f-b1a170722...(a)y11g2000yqm.googlegroups..com>,
> Ed Morton  <mortons...(a)gmail.com> wrote:
> ...
>
> >I wouldn't bother with fgrep or egrep. If you have to do anything more
> >complicated than that you may as well learn sed or awk as, with the
> >exception of a couple of cute GNU grep extensions, their syntax is as
> >simple as greps for what grep does but they're extensible to do things
> >grep can't do. The most extensible with the simplest syntax for
> >anything even moderately complex is awk but sed is also very useful
> >for simple searching/subsitutions on a single line.
>
> I agree with your general sentiment, and also your recommendation of AWK
> (as the best balance between high level and low level).  I disagree with
> the (tepid) recommendation of sed - I find that once one knows AWK, sed
> drops off the radar.

YMMV but I use sed for the trivial throw-away things like:

$ echo "foo" | sed 's/o$/x/'
fox
$ echo "foo" | awk '{sub(/o$/,"x")}1'
fox

so I still find it useful to know the sed syntax for simple
substitutions on a single line as it just saves a little bit of typing
compared to awk in that specific case.

Ed.
From: Ed Morton on 8 Jun 2010 15:08
On Jun 8, 12:24 pm, Janis Papanagnou <janis_papanag...(a)hotmail.com>
wrote:
> kaushal wrote:
> > Hi,
>
> > can someone explain me about the usage of grep , fgrep and egrep with
> > examples.
> > In what contexts i should use this.
>
> grep - if you're using simple regexps
> fgrep - if you use literal strings
> egrep - if you use extended regexps
>
> Some samples to illustrate...
>
> echo 'xyz' | grep x.z
> echo 'xyz' | fgrep x.z
>
> echo 'hello' | grep 'hello|world'
> echo 'hello' | egrep 'hello|world'
>
> And of course, as already proposed,  man grep  is your friend.

Right, and for the OP just be aware of this:

$ echo 'hello world' | egrep 'hello|world'
hello world
$ echo 'hello world' | awk '/hello/||/world/'
hello world
$ echo 'hello world' | egrep 'hello&world' <- note no output
$ echo 'hello world' | awk '/hello/&&/world/'
hello world

To simply check for BOTH "hello" and "world" appearing in any order of
the same line, with *grep you need to specify the combination of words
in every possible order:

$ echo 'hello world' | egrep 'hello.*world|world.*hello'
hello world

or create a pipline of separate grep commands:

$ echo 'hello world' | grep 'hello' | grep 'world'
hello world

whereas with awk you can just AND ("&&") the REs:

$ echo 'hello world' | gawk '/hello/&&/world/'
hello world

just like if you wanted to check for either word you could just OR
("||") the REs:

$ echo 'hello world' | gawk '/hello/||/world/'
hello world

Regards,

Ed.
From: Janis Papanagnou on 8 Jun 2010 20:34
Ed Morton wrote:
> On Jun 8, 12:24 pm, Janis Papanagnou <janis_papanag...(a)hotmail.com>
> wrote:
>> kaushal wrote:
>>> Hi,
>>> can someone explain me about the usage of grep , fgrep and egrep with
>>> examples.
>>> In what contexts i should use this.
>> grep - if you're using simple regexps
>> fgrep - if you use literal strings
>> egrep - if you use extended regexps
>>
>> Some samples to illustrate...
>>
>> echo 'xyz' | grep x.z
>> echo 'xyz' | fgrep x.z
>>
>> echo 'hello' | grep 'hello|world'
>> echo 'hello' | egrep 'hello|world'
>>
>> And of course, as already proposed, man grep is your friend.
>
> Right, and for the OP just be aware of this:
>
> $ echo 'hello world' | egrep 'hello|world'
> hello world
> $ echo 'hello world' | awk '/hello/||/world/'
> hello world
> $ echo 'hello world' | egrep 'hello&world' <- note no output
> $ echo 'hello world' | awk '/hello/&&/world/'
> hello world

See also Kornshell's

if [[ $var == @(*hello*&*world*) ]] ; then echo "both any order" ; fi
if [[ $var == @(*hello*|*world*) ]] ; then echo "one or more" ; fi


Janis

>
> To simply check for BOTH "hello" and "world" appearing in any order of
> the same line, with *grep you need to specify the combination of words
> in every possible order:
>
> $ echo 'hello world' | egrep 'hello.*world|world.*hello'
> hello world
>
> or create a pipline of separate grep commands:
>
> $ echo 'hello world' | grep 'hello' | grep 'world'
> hello world
>
> whereas with awk you can just AND ("&&") the REs:
>
> $ echo 'hello world' | gawk '/hello/&&/world/'
> hello world
>
> just like if you wanted to check for either word you could just OR
> ("||") the REs:
>
> $ echo 'hello world' | gawk '/hello/||/world/'
> hello world
>
> Regards,
>
> Ed.
From: Chris F.A. Johnson on 9 Jun 2010 03:33
On 2010-06-08, kaushal wrote:
> Hi,
>
> can someone explain me about the usage of grep , fgrep and egrep with
> examples.
> In what contexts i should use this.

fgrep and egrep are replaced in the standard by grep -F and grep -E
respectively.

fgrep uses the search patterns as strings; egrep as exetended
regular expressions.

--
Chris F.A. Johnson, author <http://shell.cfajohnson.com/>
===================================================================
Shell Scripting Recipes: A Problem-Solution Approach (2005, Apress)
Pro Bash Programming: Scripting the GNU/Linux Shell (2009, Apress)
===== My code in this post, if any, assumes the POSIX locale =====
===== and is released under the GNU General Public Licence =====
First  |  Prev  | 
Pages: 1 2
Prev: wget -r (recursive) failing to back up a site
Next: how to optionally silence lines in script