how to convert floating number to integer in linux - linux

Question

I have a file that look like this:
#[1]CHROM [2]POS [3]REF [4]ALT [5]GTEX-1117F_GTEX-1117F [6]GTEX-111CU_GTEX-111CU [7]GTEX-111FC_GTEX-111FC [8]GTEX-111VG_GTEX-111VG [9]GTEX-111YS_GTEX-111YS [10]GTEX-ZZPU_GTEX-ZZPU
22 20012563 T C 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
0.0 0.0 0.0 0.0
I want to convert it to look like this:
#[1]CHROM [2]POS [3]REF [4]ALT [5]GTEX-1117F_GTEX-1117F [6]GTEX-111CU_GTEX-111CU [7]GTEX-111FC_GTEX-111FC [8]GTEX-111VG_GTEX-111VG [9]GTEX-111YS_GTEX-111YS [10]GTEX-ZZPU_GTEX-ZZPU
22 20012563 T C 0 0 0 0 0 0 0 0 0 0 0
I basically want to convert the 0.0 or 1.0 or 2.0 to 0,1,2
I tried to use this command but it doesn't give me the correct output:
cat dosage.txt | "%d\n" "$2" 2>/dev/null
Does anyone know how to do this using awk or sed command.
Thank you.

Answer 1

how to convert floating number to integer in linux(...)using awk
You might use int function of GNU AWK, consider following simple example, let file.csv content be
name,x,y,z
A,1.0,2.1,3.5
B,4.7,5.9,7.0
then
awk 'BEGIN{FS=OFS=","}NR==1{print;next}{for(i=2;i<=NF;i+=1){$i=int($i)};print}' file.csv
gives output
name,x,y,z
A,1,2,3
B,4,5,7
Explanation: I inform GNU AWK that , is both field separator (FS) and output field separator (OFS). I print first row as-is and instruct GNU AWK to go to next line, i.e. do nothing else for that line. For all but first line I use for loop to apply int to fields from 2nd to last, after that is done I print such altered line.
(tested in GNU Awk 5.0.1)

Answer 2

This might work for you (GNU sed):
sed -E ':a;s/((\s|^)[0-9]+)\.[0-9]+(\s|$)/\1\3/g;ta' file
Presuming you want to remove the period and trailing digits from all floating point numbers (where n.n represents a minimum example of such a number).
Match a space or start-of-line, followed by one or more digits, followed by a period, followed by one or more digits, followed by a space or end-of-line and remove the period and the digits following it. Do this for all such numbers through out the file (globally).
N.B. The substitution must be performed twice (hence the loop) because the trailing space of one floating point number may overlap with the leading space of another. The ta command is enacted when the previous substitution is true and causes sed to branch to the a label at the start of the sed cycle.

Answer 3

Maybe this will help. This regex saves the whole part in a variable, and removes the rest. regex can often be fooled by unexpected input, so make sure that you test this against all forms of input data. as I did (partially) for this example.
echo 1234.5 345 a.2 g43.3 546.0 234. hi | sed 's/\b\([0-9]\+\)\.[0-9]\+/\1/g'
outputs
1234 345 a.2 g43.3 546 234. hi
It is important to note, that this was based on gnu sed (standard on linux), so it should not be assumed to work on systems that use an older sed (like on freebsd).