Using AWK to merge unique rows based on column one

问题

I am trying to write an AWK script to summarize data on a large text file. The order of the resulting data is important so i can't use sort.

I have tried different variations of FNR==NR but haven't had any luck

Input file

Height 3.5
Weight 12.3
Age 23
: 
:
Height 4.5
Weight 15.5
Age 31
:
:

Expected Output

Height 3.5 4.5
Weight 12.3 15.5
Age 23 31

回答1:

With awk:

awk '{a[$1]=a[$1] FS $2} END{for(i in a) print i a[i]}' file

Output:

Weight 12.3 15.5
Height 3.5 4.5
:     
Age 23 31

Derived from: how to merge rows that share unique IDs into a comma separated table

---

See: 8 Powerful Awk Built-in Variables – FS, OFS, RS, ORS, NR, NF, FILENAME, FNR

回答2:

Using awk's array you can group the values.

This is a simplified version:.

BEGIN {
summary["Weight"] = "Weight";
...
}
{
summary[$1] = summary [$1] " " $2
}
END {
print summary["Weight"];
...
}

回答3:

This is not awk but might work for you (GNU sed):

sed -E 'H;g;s/((\n\S+)[^\n]*)(.*)\2(.*)/\1\4\3/;h;$!d;x;s/.//' file 

Use the hold space to gather up the results for each key and at end-of-file, remove the introduced newline and print the results.

来源：https://stackoverflow.com/questions/56716292/using-awk-to-merge-unique-rows-based-on-column-one