Hi ....

we are having the below file .Column 1, Column 2 ,column 3 are key fields...
In the below ...for 2 nd , 3 rd row the repeated key column is missing ....

i want the new file to be populated with all missing key columns.


E100,0,5/29/1993,0,E001,E000,A,500000,41666.667,240.384615
_____0,1/23/1994,0,E001,E003,A,125400,10450,60.288462
_____0,6/4/1994,0,E001,E003,A,95000,7916.667,45.673077
E101,0,7/30/1993,0,E001,E003,A,87000,7250,41.826923
______ 1/9/1993,0,E001,E003,A,45200,3766.667,21.730769
E103,0,2/3/1995,0,E001,E003,A,15000,1250,7.211538
E104,0,2/15/1995,0,E001,E001,A,78000,6500,37.5


I want the result to be :


E100,0,5/29/1993,0,E001,E000,A,500000,41666.667,240.384615
E100, 0,1/23/1994,0,E001,E003,A,125400,10450,60.288462
E100, 0,6/4/1994,0,E001,E003,A,95000,7916.667,45.673077
E101,0,7/30/1993,0,E001,E003,A,87000,7250,41.826923
E101,0,1/9/1993,0,E001,E003,A,45200,3766.667,21.730769
E103,0,2/3/1995,0,E001,E003,A,15000,1250,7.211538
E104,0,2/15/1995,0,E001,E001,A,78000,6500,37.5



Please help me ........

Write an awk script which remembers the fields from the previous line, and adds as many to the front as are missing to form a full line. Assume the first line indicates how many fields a full line should have. Read up on array handling in the awk documentation.

I am new to this awk command ...........can you give a sample code on this requirment....that would be a helpful.....

Thanks your response
charan

>cat file
E100,0,5/29/1993,0,E001,E000,A,500000,41666.667,240.384615
0,1/23/1994,0,E001,E003,A,125400,10450,60.288462
0,6/4/1994,0,E001,E003,A,95000,7916.667,45.673077
E101,0,7/30/1993,0,E001,E003,A,87000,7250,41.826923
1/9/1993,0,E001,E003,A,45200,3766.667,21.730769
E103,0,2/3/1995,0,E001,E003,A,15000,1250,7.211538
E104,0,2/15/1995,0,E001,E001,A,78000,6500,37.5

>awk 'NF!=10{for(i=1;i<=(10-NF);i++)printf("%s,",k[i]);print;next}{for(i=1;i<=3;i++)k[i]=$i}1' FS=',' f1
E100,0,5/29/1993,0,E001,E000,A,500000,41666.667,240.384615
E100,0,1/23/1994,0,E001,E003,A,125400,10450,60.288462
E100,0,6/4/1994,0,E001,E003,A,95000,7916.667,45.673077
E101,0,7/30/1993,0,E001,E003,A,87000,7250,41.826923
E101,0,1/9/1993,0,E001,E003,A,45200,3766.667,21.730769
E103,0,2/3/1995,0,E001,E003,A,15000,1250,7.211538
E104,0,2/15/1995,0,E001,E001,A,78000,6500,37.5

Hi Klashxx,

can you explain the logic used in for loop?

awk 'NF!=10{for(i=1;i<=(10-NF);i++)printf("%s,",k[i]);print;next}{for(i=1;i<=3;i++)k[i]=$i}1' FS=',' f1


Thanks
Ashish

It does pretty much what I suggested, except it hard-codes the number of expected fields to 10. You could count the number of fields on the first line and use that on subsequent lines for a slightly more general approach. It remembers $1 in k[1], $2 in k[2], and $3 in k[3]; and if the line has too few fields, prints what's in k[1], k[2], etc from the previous line as many as is required. This explanation is backwards relative to the script because it prints first, and then remembers for the next line.