Shell Programming and Scripting

Hi,

I have put a similar question in one of the other threads through which I got the solution shown below but I have some more condition to add to it, hence have further queries on it. I appologies if I should be putting this with the old thread.

I have a file which perform a grep on the original file and put the result into another file File2.

grep -E "^SH1|^RD1" Original_File1 > File2 ( which has the data as shown below)

cat File2:
SH1,QC,12334
RD1,MO,898909,35476
SH1,BC,34556
RD1,FG,2341212,909090
SH1,TR,787878
RD1,GH,12345,676767
SH1,YO,565656
RD1,GO,7878604,23978

Then I combined the two records (SH1 AND RD1) using sed command and put it into File3.

sed 'N;s/\n/,/' File2 > File3

cat File3:
SH1,QC,12334,RD1,MO,898909,35476
SH1,BC,34556,RD1,FG,2341212,909090
SH1,TR,787878,RD1,GH,12345,676767
SH1,YO,565656,RD1,GO,7878604,23978

New Condition:
I am trying to do is: if the SH1 record is missing the corresponding RD1 record below it then append 8 (commas) after the end SH1 record else append the RD1 record(if existis). Also, vice versa...if RD1 record is missing the SH1 record above it, then prepend the record with 4 (commas).

All I trying to do is to create a file without combining SH1 record with SH1 record or RD1 record with RD1 record. Can we perform a loop to check ths.

If the record is something like this

SH1,QC,12334
RD1,MO,898909,35476 --- let say if this record is missing then in that case I want to add 6 commas in the above record.
SH1,BC,34556
RD1,FG,2341212,909090
SH1,TR,787878 ---let say this record is missing
RD1,GH,12345,676767
SH1,YO,565656
RD1,GO,7878604,23978

Output:
SH1,QC,12334,,,,,,,
SH1,BC,34556,RD1,FG,2341212,909090
,,,,RD1,GH,12345,676767
SH1,YO,565656,RD1,GO,7878604,23978


I will really appreciate your advice on this.

Thanks

Given these conditions, I think your data should have a record separator (a "pair-separator" if you will). If you have this, for example:

Code:

SH1,QC,12334
RD1,MO,898909,35476

then it could mean -
(a) the RD1 record corresponds to the SH1 record, or
(b) the SH1 record is missing its corresponding RD1 record and the RD1 record is missing its corresponding SH1 record as well.

And each of these assumptions yield different outputs.

This ambiguity could've been removed had there been a record separator/pair-separator.

tyler_durden