Create udev rules file from two input file - linux

Question

I am looking for a solution to create Oracle ASM udev rules file for linux. I have two input file. file1 has info of ASM disk requirement and file2 has disk information.
For example, line 2 of file1 is showing DATA12 need 3 disk(DATA12_01,DATA12_02,DATA12_03) of each 128G. file2 has all disk info with size. From these two input file I need to create output file shown bellow.
cat file1
Count - size - name
3 - 128 GB DATA12
1 - 128 GB TEMP02
2 - 4 GB ARCH03
2 - 1 GB ARCH04
1 - 3 GB ORAC01
cat file2
UUID Size
360060e80166ef70000016ef700006700 128.00 GiB
360060e80166ef70000016ef700006701 128.00 GiB
360060e80166ef70000016ef700006702 128.00 GiB
360060e80166ef70000016ef700006703 128.00 GiB
360060e80166ef70000016ef700006730 4.00 GiB
360060e80166ef70000016ef700006731 4.00 GiB
360060e80166ef70000016ef700006733 1.00 GiB
360060e80166ef70000016ef700006734 1.00 GiB
360060e80166ef70000016ef700006735 3.00 GiB
Output File
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006700", SYMLINK+="udevlinks/DATA12_01"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006701", SYMLINK+="udevlinks/DATA12_02"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006702", SYMLINK+="udevlinks/DATA12_03"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006703", SYMLINK+="udevlinks/TEMP02_01"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006730", SYMLINK+="udevlinks/ARCH03_01"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006731", SYMLINK+="udevlinks/ARCH03_02"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006733", SYMLINK+="udevlinks/ARCH04_01"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006734", SYMLINK+="udevlinks/ARCH04_02"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006735", SYMLINK+="udevlinks/ORAC01_01"

Answer 1

Here is one in AWK:
$ cat > test.awk
BEGIN {FS="([.]| +)"} # field separator do deal with "." in file2 128.00
FNR==1 {next} # skip header
NR==FNR { # read available disks to pool from file1
for(i=1; i<=$1; i++)
a[$5"_"0i]=$3 # name and set the disks into pool
next}
{
for(i in a) { # look for right sized disk
if(a[i]==$2) { # when found, print...
printf "%s%s%s%s%s", "ACTION==\"add|change\", ENV{DM_NAME}==\"",$1,"\",\"SYMLINK+=\"udevlinks/",i,"\"\n"
delete a[i] # ... and remove from pool
break
}
} # if no device was found:
old=len; len=length(a); if(old==len) {print "No device found for ",$0}
}
$ awk -f test.awk file1 file2
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006700","SYMLINK+="udevlinks/DATA12_01"
ACTION=="add|change", ENV{DM_NAME}=="360060e80166ef70000016ef700006701","SYMLINK+="udevlinks/DATA12_02"
...
No device found for THIS_IS_AN_EXAMPLE_OF_MISSING_DISK 666.00 GiB
Due to disk search using for(i in a) no order in which disks are read from the pool is guaranteed.