Login or Register to Ask a Question and Join Our Community

Sponsored Content
Full Discussion: Removing duplicates from a file
Top Forums UNIX for Dummies Questions & Answers Removing duplicates from a file Post 302836269 by Don Cragun on Tuesday 23rd of July 2013 11:59:06 PM
Old 07-24-2013
There are lots of ways to do this. If you want the output sorted by the key you're using to determine duplication, sort -u is the most logical choice. If you want to give preference to the 1st entry found with a given key and have output order match input order, awk provides a simple way to do it:
Code:
awk -F, '!a[$2,$3,$4]++' input1 input2 > mergeNoDup
awk -F, '!a[$2,$3,$4]++' mergeWithDups > mergeNoDup

Use the 1st line if you have your separate files from server 1 and server 2; use the 2nd line if you have already created a merged file and want to remove duplicates from the merged file. Both of these will work with an unlimited number of input files as long as you don't reach ARG_MAX limitations on the number of input files you're feeding into awk.

As always, if you want to try this on a Solaris/SunOS system, use /usr/xpg4/bin/awk, /usr/xpg6/bin/awk, or nawk rather than /bin/awk or /usr/bin/awk.

Other ways to do this include writing a C (or other high level language program), perl, using the associative arrays that are available in some shells, and an endless number of much less efficient combinations using read (to get a list of keys), grep (to get a count of lines containing a key), and an editor (to remove all but one occurrence of duplicated keys).
 

10 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

removing duplicates from a file

i have a file with some 1000 entries it will contain entries like 1000,ram 2000,pankaj 1001,rahim 1000,ram 2532,govind 2000,pankaj 3000,venkat 2532,govind what i want is i want to extract only the distinct rows from this file so my output should contain only 1000,ram... (2 Replies)
Discussion started by: trichyselva
2 Replies

2. Shell Programming and Scripting

Removing duplicates in a sorted file by field.

I have data like this: It's sorted by the 2nd field (TID). envoy,90000000000000634600010001,04/11/2008,23:19:27,RB00266,0015,DETAIL,ERROR, envoy,90000000000000634600010001,04/12/2008,04:23:45,RB00266,0015,DETAIL,ERROR,... (1 Reply)
Discussion started by: kinksville
1 Replies

3. UNIX for Dummies Questions & Answers

removing duplicates of a pattern from a file

hey all, I need some help. I have a text file with names in it. My target is that if a particular pattern exists in that file more than once..then i want to rename all the occurences of that pattern by alternate patterns.. for e.g if i have PATTERN occuring 5 times then i want to... (3 Replies)
Discussion started by: ashisharora
3 Replies

4. Shell Programming and Scripting

Removing duplicates from log file?

I have a log file with posts looking like this: -- Messages can be delivered by different systems at different times. The id number is used to sort out duplicate messages. What I need is to strip the arrival time from each post, sort posts by id number, and reattach arrival time to respective... (2 Replies)
Discussion started by: Ilja
2 Replies

5. Shell Programming and Scripting

Removing Duplicates from file

Hi Experts, Please check the following new requirement. I got data like the following in a file. FILE_HEADER 01cbbfde7898410| 3477945| home| 1 01cbc275d2c122| 3478234| WORK| 1 01cbbe4362743da| 3496386| Rich Spare| 1 01cbc275d2c122| 3478234| WORK| 1 This is pipe separated file with... (3 Replies)
Discussion started by: tinufarid
3 Replies

6. Shell Programming and Scripting

formatting a file and removing duplicates

Hi, I have a file that I want to change the format of. It is a large file in rows but I want it to be comma separated (comma then a space). The current file looks like this: HI, Joe, Bob, Jack, Jack After I would want to remove any duplicates so it would look like this: HI, Joe,... (2 Replies)
Discussion started by: kylle345
2 Replies

7. Shell Programming and Scripting

Removing duplicates depending on file size

Hi all, I am working with a huge amount of files in a Linux environment and I was trying to filter my data. Here's what my data looks like Name............................Size OLUSDN.gf.gif-1.JPEG.......5 kb LKJFDA01.gf.gif-1.JPEG.....3 kb LKJFDA01.gf.gif-2.JPEG.....1 kb... (7 Replies)
Discussion started by: Error404
7 Replies

8. UNIX for Dummies Questions & Answers

Grep from pattern file without removing duplicates?

I have been using grep to output whole lines using a pattern file with identifiers (fileA): fig|562.2322.peg.1 fig|562.2322.peg.3 fig|562.2322.peg.3 fig|562.2322.peg.3 fig|562.2322.peg.7 From fileB with corresponding identifiers in the second column: NODE_0 fig|562.2322.peg.1 peg ... (2 Replies)
Discussion started by: Mauve
2 Replies

9. Shell Programming and Scripting

Removing duplicates from new file

i hav two files like i want to remove/delete all the duplicate lines in file2 which are viz unix,unix2,unix3 (2 Replies)
Discussion started by: sagar_1986
2 Replies

10. Shell Programming and Scripting

Removing duplicates from new file

i hav two files like i want to remove/delete all the duplicate lines in file2 which are viz unix,unix2,unix3.I have tried previous post also,but in that complete line must be similar.In this case i have to verify first column only regardless what is the content in succeeding columns. (3 Replies)
Discussion started by: sagar_1986
3 Replies

LEARN ABOUT MOJAVE

ucblinks

ucblinks(1B)                                         SunOS/BSD Compatibility Package Commands                                         ucblinks(1B)

NAME

       ucblinks - adds /dev entries to give SunOS 4.x compatible names to SunOS 5.x devices

SYNOPSIS

       /usr/ucb/ucblinks [-e rulebase] [-r rootdir]

DESCRIPTION

       ucblinks  creates symbolic links under the /dev directory for devices whose SunOS 5.x names differ from their SunOS 4.x names. Where possi-
       ble, these symbolic links point to the device's SunOS 5.x name rather than to the actual /devices entry.

       ucblinks does not remove unneeded compatibility links; these must be removed by hand.

       ucblinks should be called each time the system is reconfiguration-booted, after any new SunOS 5.x links that are needed have been  created,
       since the reconfiguration may have resulted in more compatibility names being needed.

       In  releases prior to SunOS 5.4, ucblinks used a  nawk rule-base to construct the SunOS 4.x compatible names. ucblinks no longer uses  nawk
       for the default operation, although  nawk rule-bases can still be specifed with the -e option.  The  nawk rule-base equivalent to the SunOS
       5.4 default operation can be found in /usr/ucblib/ucblinks.awk.

OPTIONS

       -e rulebase     Specify rulebase as the file containing nawk(1) pattern-action statements.

       -r rootdir      Specify rootdir as the directory under which dev and devices will be found, rather than the standard root directory /.

FILES

       /usr/ucblib/ucblinks.awk        sample rule-base for compatibility links

ATTRIBUTES

       See attributes(5) for descriptions of the following attributes:

       +-----------------------------+-----------------------------+
       |      ATTRIBUTE TYPE         |      ATTRIBUTE VALUE        |
       +-----------------------------+-----------------------------+
       |Availability                 |SUNWscpu                     |
       +-----------------------------+-----------------------------+

SEE ALSO

       devlinks(1M), disks(1M), ports(1M), tapes(1M), attributes(5)

SunOS 5.10                                                          13 Apr 1994                                                       ucblinks(1B)
All times are GMT -4. The time now is 08:35 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy