grep for string in multiple files only one instance.

[galger]

if multiple instances exsist in multiple files.. How can I return only one instance.. I did this a long time ago.. looked through man page. could not find anything.. let know.


Thanks in advance.
galger.

 

[PHV]

Something like this ?
grep -q "$string" list of files && echo "$string found"

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[galger]

-q does not work for solaris grep.

 

[PHV]

And this ?
grep "$string" list of files | head -1

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[feherke]

Hi

No idea if is something standard, or just a GNU extension :

-m NUM, --max-count=NUM
Stop reading a file after NUM matching lines.

By the way. I use [tt]grep (GNU grep) 2.5.1[/tt]. Is any official name of that "solaris grep" ? I would like to google for it to read some details.

Feherke.

http://rootshell.be/~feherke/

 

[KenCunningham]

man grep on a Solaris box should give you the vanilla grep man page (I think).

 

[feherke]

Hi

Sorry, forgot to mention that I have no available space to install Solaris just to read one man page.

Feherke.

http://rootshell.be/~feherke/

 

[Annihilannic]

Fortunately, Sun have.

http://docs.sun.com/app/docs/doc/817-0689/6mgfkpcpr?q=grep&a=view

Annihilannic.

 

[PHV]

-q does not work for solaris grep
Courtesy of Annihilannic I see you have to use this version of grep:
/usr/xpg4/bin/grep

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[feherke]

Hi

Thanks. That is a good overview. So no -m on Solaris.

But neighter that man page clarify how the "Solaris [tt]grep[/tt]" introduces itself.

Feherke.

http://rootshell.be/~feherke/

 

[Annihilannic]

Not sure what you mean? Before GNU grep, grep was just grep.

The xpg4 one is the POSIX compliant version I believe.

Annihilannic.

 

[PHV]

I guess it's compliant with:
ISO/IEC DIS 9945 - 2:1992, Information technology - Portable
Operating System Interface (POSIX) - Part 2: Shell and Utilities
(IEEE Std 1003.2 - 1992);

Hope This Helps, PH.
Want to get great answers to your Tek-Tips questions? Have a look at FAQ219-2884 or FAQ181-2886

 

[feherke]

Hi

Thanks. Now I have keywords for googling.

Feherke.

http://rootshell.be/~feherke/

 

[razalas]

Galger,

this is not exactly what you asked for in your original post, but I'm wondering if what you are really after is just a list of the files that contain any occurrence of your search string. If so, try the -l switch for grep:

grep [b]-l[/b] "search-string" <file-list>

I'm on SCO, so I hope the "l" switch is available on Solaris.

Code what you mean,
and mean what you code!
But by all means post your code!

Razalas

 

[galger]

Thanks for all of your help.. appreciated.

 

[kordaff]

# You can also
grep SomeString files* | uniq
# to return just one instance

 

[PHV]

kordaff, in which flavor of *nix does uniq return just one row of an unsorted file ?

 

[kordaff]

It depends on your search string of course:

[phil@localhost phil]$ grep alpha text* | uniq
text:alpha
text2:alpha
text3:alpha just a word on a line?
text3:alpha

[phil@localhost phil]$ perl -e 'for(1..$#ARGV){open(FILE,$ARGV[$_])or die"$!\n";while(<FILE>){if (/$ARGV[0]/){$flag=1}}}if ($flag){print "Found $ARGV[0]\n"}' alpha text text2 text3

Found alpha

Perl does a better job if you need to search for a term that is part of a line in multiple files...