BASH : Sum size of same name directories

tjerich · 09-24-2013, 11:26 AM

EDITED: Added more information

First of all, I am a bash noob, so please be gentle

I am trying to sum the size of folders that are in different places but have the same name. It looks like this :

root
--- directory 1

------ folder 1
--------subfolder 1
--------subfolder 2
------ folder 2
--------subfolder 3
--------subfolder 4
------ folder 3
--------subfolder 5
--------subfolder 6

--- directory 2

------ folder 1
--------subfolder 1
--------subfolder 2
------ folder 2
--------subfolder 3
--------subfolder 4
------ folder 3
--------subfolder 5
--------subfolder 6

I am trying to sum the size of subdirectories 1 to 6 and output that to a .csv

At the moment I am simply outputting sizes of subdirectories in two seperate CSV files. One for directory 1 and one for directory 2

At the moment I have this to output sizes of subfodlers that I run where I need them :

du -h --max-depth=1 --block-size=GB * | grep "[\/]" | sort -n -r > ~/lists/disks/RC_job.csv

The output look like this :

40GB folder1/subfolder1

15GB folder1/subfolder2

10GB folder2/subfolder 3
...

I have one output for directory 1 and one for directory 2. I would like to sum the size of subfolders from directory one and two and have an output that looks like this

60GB subfolder1

25GB subfolder2

10GB subfolder3

Where subfolder1 is directory1/folder1/subfolder1 + directory2/folder1/subfolder1

This is my first post here I do not know if this enough info. I would be pleased to provide more if necessary. I am pretty sure this can be done with awk, but I haven't really used that yet.

Cheers !

unSpawn · 09-24-2013, 12:27 PM

Something like?:

Code:

for DIRECTORYNAME in name0 name1 name2; do
 RETVAL_BT=$(echo "scale=0;$(find ${DIRECTORYNAME} -type f -printf "%s+"|sed -e "s|\+$||g")"|bc -l)
 RETVAL_GB=$(echo "scale=2;$RETVAL_BT/1024/1024/1024"|bc -l)
 echo "${RETVAL_GB:=ERROR}";${DIRECTORYNAME};
done

schneidz · 09-24-2013, 12:38 PM

i wouldnt use spaces in filenames...
quick-and-dirty:

Code:

[schneidz@hyper ~]$ awk '{print $2}' dir1.lst dir2.lst | while read dir
do
 size=`cat dir1.lst dir2.lst | grep -w $dir | awk '{size = size + $1} END {print size}'`
 echo $size"gb" $dir
done

grail · 09-24-2013, 06:16 PM

Untested, but how about something like:

Code:

find dir1 dir2 -type d -maxdepth 2 -mindepth 2 -exec du -sBG {} \; | awk -F"[ /]*" '{_[$NF]+=gensub(/G/,"","g",$1)}END{for(i in _)print i,_[i]}'