Sponsored Content
Top Forums UNIX for Dummies Questions & Answers how to deleting except the patern in a file Post 302357976 by Tytalus on Thursday 1st of October 2009 05:12:15 AM
Old 10-01-2009
one way:

Code:
#  cat infile
quality hardware csx pfizet honda kiuty
hardware quality pfizer name
newyork quality mafoi pfizer  working csx

#  nawk '{t=$1" "$2;for (i=3;i<=NF;i++){ if (index("csx pfizer mafoi",$i)){t=t" "$i}} print t;t=""}' infile
quality hardware csx
hardware quality pfizer
newyork quality mafoi pfizer csx

(Not quite the same as your output as it removes the "pfizet" on line 1, which I'm guessing was a minro typo ?)

HTH

Last edited by Tytalus; 10-01-2009 at 06:17 AM.. Reason: minor typo
 

10 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

Deleting lines inside a file without opening the file

Hi, Just consider there are around 10 lines in a file. Now is it possible to delete the first 2 lines in the file without opening the file. No matter whatever the content of the file is, I just wanna delete the first 2 lines without opening the file. Is that possible? If so, please help me out.... (3 Replies)
Discussion started by: toms
3 Replies

2. UNIX for Dummies Questions & Answers

tar archive with including specific patern files

Hi, I need to create recursive tar archive, while I put there only files of type a*.txt. Without file filtering the command is: tar cfzf test.tar.gz test_tar/ How I include the switch for including only files with pattern a*.txt ? Thanks a lot! (1 Reply)
Discussion started by: john.gelburg
1 Replies

3. Shell Programming and Scripting

How to specify 'not case sensitive' in regex (nawk, sed, patern expencions)?

Is it possible to make the search in regular exprecion or in matching parts of sed, nawk and others to IGNORE the case of the search string? I mean, like if used 'grep' with -i option: > grep -i "abc" file I would like to be able to do the same, say, by nawk: > nawk '/abc/ {print $0}'... (4 Replies)
Discussion started by: alex_5161
4 Replies

4. Shell Programming and Scripting

Patern Match Question on file names

I have a script which I use to archive log files and I want to install it on another server. I match any file with a ".log" in the name. Most files end with ".log" or ".log.nnnn". Of course someone has a file on this server that they do not want to archive that has .login.ear in the file name and... (1 Reply)
Discussion started by: prismtx
1 Replies

5. Shell Programming and Scripting

awk - writing matching pattern to a new file and deleting it from the current file

Hello , I have comma delimited file with over 20 fileds that i need to do some validations on. I have to check if certain fields are null and then write the line containing the null field into a new file and then delete the line from the current file. Can someone tell me how i could go... (2 Replies)
Discussion started by: goddevil
2 Replies

6. Shell Programming and Scripting

Deleting file name

I need a way to remove all the file names with they are extension from a line of a document. For example: I have doc named "gara" with the following content: /media/gogo/6651-FEAB/Desktop/Desktop.jpg /media/gogo/6651-FEAB/Desktop.pdf /media/gogo/6651-FEAB/linux/logo1.jpg... (4 Replies)
Discussion started by: gogok_bg
4 Replies

7. UNIX for Dummies Questions & Answers

Deleting a pattern in UNIX without deleting the entire line

Hi I have a file: r58778.3|SOURCES={KEY=f665931a...,fw,221-705}|ERRORS={16_1:T,30_1:T,56_1:C,57_1:T,59_1:A,101_1:A,115:-,158_1:C,186_1:A,204:-,271_1:T,305:-,350_1:C,368_1:G,442_1:C,472_1:G,477_1:A}|SOURCE_1="Contig_1092402550638"(f665931a359e36cea0976db191ff60ff09cc816e) I want to retain... (15 Replies)
Discussion started by: Alyaa
15 Replies

8. UNIX for Dummies Questions & Answers

Grep to find lines matching given patern in a file

When I try with patern matching in a file with below code works cat scj_drive_commands | egrep '/app/oracle/build_lib/pkg32|/app/oracle/build_lib/pkg33' But when I have code with patern searching using of below does not work ! cat scj_drive_commands | egrep... (3 Replies)
Discussion started by: Siva SQL
3 Replies

9. UNIX for Dummies Questions & Answers

Grep to find matching patern and return unique values

Request: grep to find given matching patern and return unique values, eliminate the duplicate values I have to retrieve the unique folder on the below file contents like; /app/oracle/build_lib/pkg320.0_20120927 /app/oracle/build_lib/pkg320.0_20121004_prof... (5 Replies)
Discussion started by: Siva SQL
5 Replies

10. UNIX for Dummies Questions & Answers

Deleting file basing on the timestamp substring in the file name

Hello, I have in my backup folder, files with names convention like this : randomFileNames_13-02-2014_23h13m09+1392333189 randomFileNames_14-02-2014_02h13m09+1392343989 randomFileNames_14-02-2014_04h13m09+1392351189 etc.... Base on timestamp at end of the filename, I would to delete all the... (7 Replies)
Discussion started by: thuyetti
7 Replies
IPSEC_RANBITS(8)						  [FIXME: manual]						  IPSEC_RANBITS(8)

NAME
ipsec_ranbits - generate random bits in ASCII form SYNOPSIS
ipsec ranbits [--quick] [--continuous] [--bytes] nbits DESCRIPTION
Ranbits obtains nbits (rounded up to the nearest byte) high-quality random bits from random(4), and emits them on standard output as an ASCII string. The default output format is datatot(3) h format: lowercase hexadecimal with a 0x prefix and an underscore every 32 bits. The --quick option produces quick-and-dirty random bits: instead of using the high-quality random bits from /dev/random, which may take some time to supply the necessary bits if nbits is large, ranbits uses /dev/urandom, which yields prompt results but lower-quality randomness. The --continuous option uses datatot(3) x output format, like h but without the underscores. The --bytes option causes nbits to be interpreted as a byte count rather than a bit count. FILES
/dev/random, /dev/urandom SEE ALSO
ipsec_datatot(3), random(4) HISTORY
Written for the Linux FreeS/WAN project <http://www.freeswan.org> by Henry Spencer. BUGS
There is an internal limit on nbits, currently 20000. Without --quick, ranbits's run time is difficult to predict. A request for a large number of bits, at a time when the system's entropy pool is low on randomness, may take quite a while to satisfy. Though not a bug of ranbits, the direct use of /dev/hw_random, the Linux hardware random number generator is not supported because it can produce very non-random data. To properly use /dev/hw_random, the rngd daemon should be used to read from /dev/hw_random and write to /dev/random, while performing a FIPS test on the hardware random read. No changes to Openswan are required for this support - just a running rngd. [FIXME: source] 10/06/2010 IPSEC_RANBITS(8)
All times are GMT -4. The time now is 09:04 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy