removing redundant entries

anasuya has asked for the wisdom of the Perl Monks concerning the following question:

I have multiple entries in a file which look like this.

1HDZ-B    3NT5-A    0.489917    0.400067
1I30-B    2B35-E    0.879363    0.400068
1K0U-E    1UWK-A    0.969722    0.400072
1BPW-A    1EFL-C    0.556795    0.400077
1EFL-C    1BPW-A    0.556795    0.400077
1GIQ-B    3QJ5-B    0.997880    0.400077
1GEE-A    2B4R-O    0.829231    0.400080
[download]

for the 4th and 5th line,the first two columns are interchanged, i.e.1st column on line4 =2nd column on the subsequent line. so what if i want to keep only one line among lines 4 and 5. say the line 4, so that my file looks like this:

1HDZ-B    3NT5-A    0.489917    0.400067
1I30-B    2B35-E    0.879363    0.400068
1K0U-E    1UWK-A    0.969722    0.400072
1BPW-A    1EFL-C    0.556795    0.400077
1GIQ-B    3QJ5-B    0.997880    0.400077
1GEE-A    2B4R-O    0.829231    0.400080
[download]

also, there are multiple instances of this occurrence in my file. so how do i remove such redundant entries?

Comment on removing redundant entries Select or Download Code

Replies are listed 'Best First'.
Re: removing redundant entries by ikegami (Patriarch) on Dec 01, 2011 at 10:32 UTC
`perl -nae'print if !$seen{ join " ", sort @F[0,1] }++' infile >outfile` [download]	[reply] [d/l]
Re: removing redundant entries by MidLifeXis (Monsignor) on Dec 01, 2011 at 14:01 UTC
ikegami has given you a solution that works in your case, but in general, the steps to take to solve the "how do I remove redundant data" question are along these lines: Identify the information that determines if a set of data is unique. This is the key. It may need to be normalized (the `sort @F[0,1]` call in ikegami's response, for example). See if the key has already been used. Hashes are useful for this (the `!$seen{...}++` construct in ikegami's response). See How can I remove duplicate elements from a list or array? (or `perldoc -q duplicate`) for more information. Process the duplicate data for that key. In some cases it you are only interested in the last data associated with the key, in others you may want to group all of the data together, in still others, you may want to run some function on all of the data associated with that key. Return the processed results. If you step through your question using this framework, it can be easier to come up with a solution on your own. Update: Added references to ikegami's response --MidLifeXis	[reply] [d/l] [select]