Shell Programming and Scripting

Hey guyz,

I have a table like this:


# are just some numbers. I want to have the output like this:


So basically, I wanna know in each of the rows in my input (which gonna be the columns in output) what is the number for each factor (A,B,C,...). sometimes there are repeating of one factor which I like to take the maximum number in that case. And sometimes the factor is missing which it turns NA in corresponding cell in output.

There are not the same number of factor in each row which makes it problematic.

Can anyone help me with this?
Thanks in advance

Code:

awk -F "[ ;]" '{!X[$1]++;for(i=2;i<=NF;i++){split($i,b,"=");a[$1,b[1]]=b[2];!Y[b[1]]++;}}END{
for(i in X){s=s?s" "i:"* "i}
print s;
for(j in Y){printf j" "
for(i in X){
printf a[i,j]?a[i,j]" ":"NA "
}print ""}}' file

With some assumptions:

Code:

perl -lane '
@temp=split /[=;]/,$F[1];
for ($i=0;$i<@temp;$i+=2) {
 if(not exists $val{$temp[$i] . " " . $F[0]}) {
  $val{$temp[$i]  . $F[0]}=$temp[$i+1]
 }
 else {
  if ($temp[$i+1] >= $val{$temp[$i] . " " . $F[0]}) {
   $val{$temp[$i] . $F[0]}=$temp[$i+1]
  }
 }
 $exists{$temp[$i]}=1;
 $cols{$F[0]}=1;
}
END{
@cols=sort keys %cols;
unshift @cols,"*";
print "@cols";
for (sort keys %exists) {
 @temp=();
  for ($i=1;$i<@cols;$i++) {
   push @temp,((exists $val{$_  . $cols[$i]})?$val{$_ . $cols[$i]}:"NA")
  }
 unshift @temp,$_;
 print "@temp";
}
}' file

Thank you both pamu and elixir_sinari! Your posts are always helpful! I need some time to go through your scripts and find out the logic and then applying them on my data.
I will let you know...

cheers!

Ok. I went through the scripts and applied it on my data. elixir_sinari it seems your script does not work for all factors. It doesn't give the information for all factors. With minor knowledge that I have about perl I couldn't find what is the problem.
pamu, your scripts seems to work just fine. There is just problem with those factors at the very end of each line. for some reasons the output is not true for those ones.

To make it more clear for you guys I write the first 5 lines of my input with some simplicity:



And in this case the desired output will be:


If you can find out the solution you will make me so happy!

P.S. I have some factors which their names start with similar letters (for exmaple: zfh, z, zld). But I don't think that's the problem.
Thanks again

##before posting this I just applied pamu's script on this exmaple here and it works just fine! But on my own data which is rather big and compicated I face the problem that I mentioned! Help me! :| :|

Made some changes to my earlier post. Check now.

Yes. If on sample file it is working fine then we don't have any chance of checking what could be the problem.

And please clarify - above highlighted in RED there are two D present is first row. What to do here..?