Printable View
I have a large csv file that I need to break apart the file, looks
like this.
How do you search for a specific pattern with in a specific field?
"text1","text2","text3","text4"
i can do:
grep ',"text3",' foo > newfoo
however i need to make sure that the 3rd field is always searched not
just an occurrance anywhere within the file.
help appreciated.
wj
willjay typed (on Sat, Sep 15, 2007 at 03:32:06PM -0700):
| I have a large csv file that I need to break apart the file, looks
| like this.
|
| How do you search for a specific pattern with in a specific field?
|
| "text1","text2","text3","text4"
|
| i can do:
|
| grep ',"text3",' foo > newfoo
|
| however i need to make sure that the 3rd field is always searched not
| just an occurrance anywhere within the file.
|
| help appreciated.
grep '.*,.*,.*pattern_to_find.*,' file
Haven't tried it, but seems logical as I type it...
--
JP
==> [url]http://www.frappr.com/cusm[/url] <==
On Sat, 15 Sep 2007, willjay wrote:
[color=blue]
> I have a large csv file that I need to break apart the file, looks
> like this.
>
> How do you search for a specific pattern with in a specific field?
>
> "text1","text2","text3","text4"
>
> i can do:
>
> grep ',"text3",' foo > newfoo
>
> however i need to make sure that the 3rd field is always searched not
> just an occurrance anywhere within the file.[/color]
I believe you can use the "cut" command to pull out the third field. But
if you want to output the whole line, I don't think it would work.
Regards,
.....Bob Rasmussen, President, Rasmussen Software, Inc.
personal e-mail: [email]ras@anzio.com[/email]
company e-mail: [email]rsi@anzio.com[/email]
voice: (US) 503-624-0360 (9:00-6:00 Pacific Time)
fax: (US) 503-624-0760
web: [url]http://www.anzio.com[/url]
willjay wrote:[color=blue]
> I have a large csv file that I need to break apart the file, looks
> like this.
>
> How do you search for a specific pattern with in a specific field?
>
> "text1","text2","text3","text4"
>
> i can do:
>
> grep ',"text3",' foo > newfoo
>
> however i need to make sure that the 3rd field is always searched not
> just an occurrance anywhere within the file.
>
> help appreciated.
>
> wj
>
>[/color]
Looks to me like a job for AWK and not grep.
awk -F "," ' $3 ~ /pattern you want/ { print $0 } ' foo > newfoo
The above will print the entire record where field 3 contains "pattern you want."
If you only want field 3, then change "print $0" to "print $3."
One problem is the ","text with, in it","
Its better to get rid of "," and change it to just | with a sed
command:
sed 's/","/|/g' foo | \
awk -F "|" ' $3 ~ /pattern you want/ { print $0 } ' | \
sed 's/|/","/g' > newfoo
--
Steve Fabac
S.M. Fabac & Associates
816/765-1670
On Sep 16, 3:09 am, "Steve M. Fabac, Jr." <smfa...@att.net> wrote:[color=blue]
> willjay wrote:[color=green]
> > I have a large csv file that I need to break apart the file, looks
> > like this.[/color]
>[color=green]
> > How do you search for a specific pattern with in a specific field?[/color]
>[color=green]
> > "text1","text2","text3","text4"[/color]
>[color=green]
> > i can do:[/color]
>[color=green]
> > grep ',"text3",' foo > newfoo[/color]
>[color=green]
> > however i need to make sure that the 3rd field is always searched not
> > just an occurrance anywhere within the file.[/color]
>[color=green]
> > help appreciated.[/color]
>[color=green]
> > wj[/color]
>
> Looks to me like a job for AWK and not grep.
>
> awk -F "," ' $3 ~ /pattern you want/ { print $0 } ' foo > newfoo
>
> The above will print the entire record where field 3 contains "pattern you want."
> If you only want field 3, then change "print $0" to "print $3."
>
> One problem is the ","text with, in it","
>
> Its better to get rid of "," and change it to just | with a sed
> command:
>
> sed 's/","/|/g' foo | \
> awk -F "|" ' $3 ~ /pattern you want/ { print $0 } ' | \
> sed 's/|/","/g' > newfoo
>
> --
> Steve Fabac
> S.M. Fabac & Associates
> 816/765-1670- Hide quoted text -
>
> - Show quoted text -[/color]
awk balks with
awk: input record `"1","TEXT DATA...' too long
source line 1 of program << $24 ~ /"GA"/ { pri ... >>
wj
willjay wrote:[color=blue]
> On Sep 16, 3:09 am, "Steve M. Fabac, Jr." <smfa...@att.net> wrote:[color=green]
>> willjay wrote:[color=darkred]
>>> I have a large csv file that I need to break apart the file, looks
>>> like this.
>>> How do you search for a specific pattern with in a specific field?
>>> "text1","text2","text3","text4"
>>> i can do:
>>> grep ',"text3",' foo > newfoo
>>> however i need to make sure that the 3rd field is always searched not
>>> just an occurrance anywhere within the file.
>>> help appreciated.
>>> wj[/color]
>> Looks to me like a job for AWK and not grep.
>>
>> awk -F "," ' $3 ~ /pattern you want/ { print $0 } ' foo > newfoo
>>
>> The above will print the entire record where field 3 contains "pattern you want."
>> If you only want field 3, then change "print $0" to "print $3."
>>
>> One problem is the ","text with, in it","
>>
>> Its better to get rid of "," and change it to just | with a sed
>> command:
>>
>> sed 's/","/|/g' foo | \
>> awk -F "|" ' $3 ~ /pattern you want/ { print $0 } ' | \
>> sed 's/|/","/g' > newfoo
>>
>> --
>> Steve Fabac
>> S.M. Fabac & Associates
>> 816/765-1670- Hide quoted text -
>>
>> - Show quoted text -[/color]
>
> awk balks with
>
> awk: input record `"1","TEXT DATA...' too long
> source line 1 of program << $24 ~ /"GA"/ { pri ... >>
>
> wj
>
>[/color]
Try gawk:
[color=blue]
> Many versions of awk have various implementation limits, on things such as:
>
> Number of fields per record
>
> Number of characters per input record
>
> Number of characters per output record
>
> Number of characters per field
>
> Number of characters per printf string
>
> Number of characters in literal string
>
> Number of characters in character class
>
> Number of files open
>
> Number of pipes open
>
> The ability to handle 8-bit characters and characters
> that are all zero (ASCII NUL)
>
> gawk does not have limits on any of these items, other
> than those imposed by the machine architecture and/or
> the operating system.[/color]
/usr/gnu/bin/gawk -> /opt/K/SCO/gnutools/5.0.7Kj/usr/gnu/bin/gawk
--
Steve Fabac
S.M. Fabac & Associates
816/765-1670