Is there a way to 'uniq' by column?

问题:

I have a .csv file like this:

[e-mail],2009-11-27 01:05:47.893000000,example.net,127.0.0.1
[e-mail],2009-11-27 00:58:29.793000000,example.net,255.255.255.0
[e-mail],2009-11-27 00:58:29.646465785,example.net,256.255.255.0
...

I have to remove duplicate e-mails (the entire line) from the file (i.e. one of the lines containing [e-mail] in the above example). How do I use uniq on only field 1 (separated by commas)? According to man, uniq doesn't have options for columns.

I tried something with sort | uniq but it doesn't work.

回答1:

sort -u -t, -k1,1 file

• -u for unique
• -t, so comma is the delimiter
• -k1,1 for the key field 1

Test result:

[e-mail],2009-11-27 00:58:29.793000000,xx3.net,255.255.255.0 
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1 

回答2:

awk -F"," '!_[$1]++' file

• -F sets the field separator.
• $1 is the first field.
• _[val] looks up val in the hash _(a regular variable).
• ++ increment, and return old value.
• ! returns logical not.
• there is an implicit print at the end.

回答3:

To consider multiple column.

Sort and give unique list based on column 1 and column 3:

sort -u -t : -k 1,1 -k 3,3 test.txt

• -t : colon is separator
• -k 1,1 -k 3,3 based on column 1 and column 3

回答4:

or if u want to use uniq:

<mycvs.cvs tr -s ',' ' ' | awk '{print $3" "$2" "$1}' | uniq -c -f2

gives:

1 01:05:47.893000000 2009-11-27 [e-mail]
2 00:58:29.793000000 2009-11-27 [e-mail]
1

回答5:

If you want to retain the last one of the duplicates you could use

 tac a.csv | sort -u -t, -r -k1,1 |tac

Which was my requirement

here

tac will reverse the file line by line

回答6:

Here is a very nifty way.

First format the content such that the column to be compared for uniqueness is a fixed width. One way of doing this is to use awk printf with a field/column width specifier ("%15s").

Now the -f and -w options of uniq can be used to skip preceding fields/columns and to specify the comparison width (column(s) width).

Here are three examples.

In the first example...

1) Temporarily make the column of interest a fixed width greater than or equal to the field's max width.

2) Use -f uniq option to skip the prior columns, and use the -w uniq option to limit the width to the tmp_fixed_width.

3) Remove trailing spaces from the column to "restore" it's width (assuming there were no trailing spaces beforehand).

printf "%s" "$str" \
| awk '{ tmp_fixed_width=15; uniq_col=8; w=tmp_fixed_width-length($uniq_col); for (i=0;i<w;i++) { $uniq_col=$uniq_col" "}; printf "%s\n", $0 }' \
| uniq -f 7 -w 15 \
| awk '{ uniq_col=8; gsub(/ */, "", $uniq_col); printf "%s\n", $0 }'

In the second example...

Create a new uniq column 1. Then remove it after the uniq filter has been applied.

printf "%s" "$str" \
| awk '{ uniq_col_1=4; printf "%15s %s\n", uniq_col_1, $0 }' \
| uniq -f 0 -w 15 \
| awk '{ $1=""; gsub(/^ */, "", $0); printf "%s\n", $0 }'

The third example is the same as the second, but for multiple columns.

printf "%s" "$str" \
| awk '{ uniq_col_1=4; uniq_col_2=8; printf "%5s %15s %s\n", uniq_col_1, uniq_col_2, $0 }' \
| uniq -f 0 -w 5 \
| uniq -f 1 -w 15 \
| awk '{ $1=$2=""; gsub(/^ */, "", $0); printf "%s\n", $0 }'

回答7:

well, simpler than isolating the column with awk, if you need to remove everything with a certain value for a given file, why not just do grep -v:

e.g. to delete everything with the value "col2" in the second place line: col1,col2,col3,col4

grep -v ',col2,' file > file_minus_offending_lines

If this isn't good enough, because some lines may get improperly stripped by possibly having the matching value show up in a different column, you can do something like this:

awk to isolate the offending column: e.g.

awk -F, '{print $2 "|" $line}'

the -F sets the field delimited to ",", $2 means column 2, followed by some custom delimiter and then the entire line. You can then filter by removing lines that begin with the offending value:

 awk -F, '{print $2 "|" $line}' | grep -v ^BAD_VALUE

and then strip out the stuff before the delimiter:

awk -F, '{print $2 "|" $line}' | grep -v ^BAD_VALUE | sed 's/.*|//g'

(note -the sed command is sloppy because it doesn't include escaping values. Also the sed pattern should really be something like "[^|]+" (i.e. anything not the delimiter). But hopefully this is clear enough.

回答8:

By sorting the file with sort first, you can then apply uniq.

It seems to sort the file just fine:

$ cat test.csv
[e-mail],2009-11-27 00:58:29.793000000,xx3.net,255.255.255.0
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 00:58:29.646465785,2x3.net,256.255.255.0 
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1

$ sort test.csv
[e-mail],2009-11-27 00:58:29.646465785,2x3.net,256.255.255.0 
[e-mail],2009-11-27 00:58:29.793000000,xx3.net,255.255.255.0
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1

$ sort test.csv | uniq
[e-mail],2009-11-27 00:58:29.646465785,2x3.net,256.255.255.0 
[e-mail],2009-11-27 00:58:29.793000000,xx3.net,255.255.255.0
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1

You could also do some AWK magic:

$ awk -F, '{ lines[$1] = $0 } END { for (l in lines) print lines[l] }' test.csv
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 01:05:47.893000000,xx2.net,127.0.0.1
[e-mail],2009-11-27 00:58:29.646465785,2x3.net,256.255.255.0