Separating the Columns based on the ColumnSize

SeenuGuddu · #1 (**permalink**) 10-28-09, 08:20

i had Client Table with the colummns

Code:

col1 		 Decimal(3),
col2 		 Decimal(2),
col3 		 String(1),
col4_blank1* string(2),
col5_blank2 *string(10),
col6	     string(3);


the flat file is

89522D            085NOVOLIN BRAND                 888888ALL OTHER

i want to dispaly the output by seperating the columns with ~ based on the column size is 3 like col1 Decimal(3) and col2 having the size 2

895~22~D~    ~          ~085~NOVOLIN

Thanks in Advance

mike_bike_kite · #2 (**permalink**) 10-28-09, 08:56

This starts you off. It passes your file through a small awk program. You'll need to expand the awk program to do all your fields - alter the 2nd param to be the postion in the line and the 3rd param to be the number of characters to use. In your code the 3rd field is 1 character but your example seems to hold many characters. You might also be able to use the cut with the -c option to get certain parts and then perhaps use the output-delimiter option to put in the ~.

Code:

#!/bin/sh

cat your_file | \
awk '{ print    substr( $0,1,3 ) "~" substr( $0,4,2 ) "~" substr( $0,6,1 ); }'

exit

SeenuGuddu · #3 (**permalink**) 10-29-09, 06:06

in Temp table having the column size Value

Code:

Id Column_name     Col_size 
1    col1                   3
2    col2                   2
3    col3                   1

the records of the file  is 

89522D            085

based on the column size.. i have to delimitate the records of file into 

895~22~D~            085

can u give some example on this

mike_bike_kite · #4 (**permalink**) 10-29-09, 06:52

Quote:

Originally Posted by SeenuGuddu

can u give some example on this

This example just runs your example data:
Code:
#!/bin/sh
echo "89522D            085NOVOLIN BRAND                 888888ALL OTHER" | \
awk '{ print    substr( $0,1,3 ) "~" substr( $0,4,2 ) "~" substr( $0,6,1 ); }'

exit
Running the above produces:
Code:
895~22~D

SeenuGuddu · #5 (**permalink**) 10-29-09, 07:37

Hi,

Thanks for your reply.

i had 100 records in each record having 50 columns of Data

Code:

sams11Y
johns12Y
ravi13N

 how would i get the position of the start string in substr( $0,1,3 )

in that loop of each record i have to print with ~ delimated  like 
sams~11~Y
johns~12~Y

Thanks

mike_bike_kite · #6 (**permalink**) 10-29-09, 08:24

Quote:

Originally Posted by SeenuGuddu

how would i get the position of the start string in substr( $0,1,3 )

I'd suggest counting. The first character in the line is 1, the next is 2 and so fourth. You then put this value in the 2nd field above. The length of the field goes into the 3rd field. Repeat. This measuring device might also help:
Code:
         1         2         3         4         5         6
123456789012345678901234567890123456789012345678901234567890
89522D            085NOVOLIN BRAND                 888888ALL OTHER
895 starts at position 1.
22 starts at position 4.
D starts at position 6.
...

Mike

SeenuGuddu · #7 (**permalink**) 10-29-09, 08:41

Thanks Mike

Can u share the Example for this start String and End String in substr( $0,1,3 )

mike_bike_kite · #8 (**permalink**) 10-29-09, 09:59

it generally doesn't hurt if you try a little experimenting for yourself ...

SeenuGuddu · #9 (**permalink**) 10-30-09, 09:15

sure Mike .. i will get back to u.

Thanks

SeenuGuddu · #10 (**permalink**) 11-20-09, 09:37

hi mike

here is the code but i have some problem with that

${COL_SIZE_DAT its a flat file of records of column size

Code:

	normalfile="john_789_has_2007-11-30_s"
	
    j=1
    export delim=""
    while read col_size_dat
    do 
        export col_size=$(echo "${col_size_dat}" | cut -f2 -d"|")
        export delim_temp="${delim} substr( $0,$j,$col_size) \"~\"" 
        j=`expr $j + ${col_size}`   
    done < ${COL_SIZE_DAT}
   
   export delim_temp1=$(echo "$delim_temp" | sed 's/\(.*\)"~"/\1/')   
   export delim_cmd=`echo " awk '{ print ${delim_temp1}; }' " ` 
   echo $delim_cmd
 
  cat normalfilename | $delim_cmd > delimatedfile

instead of normalfile="john_789_has_2007-11-30_s"
u want me to give normalfile for $0 or any other way we had the substr( $0,$j,$col_size)

wat should i use for redirecting the filewith delimated
is that ok

cat normalfilename | $delim_cmd > delimatedfile
and after running this
i am getting the error

awk: cmd. line:1: '{
awk: cmd. line:1: ^ invalid char ''' in expression

AnanthaP · #11 (**permalink**) 11-22-09, 23:08

col1 Decimal(3), cols 1,2,3 substr($0,1,3) in awk
col2 Decimal(2), cols 4,5 : substr($0,4,2) in awk
col3 String(1), and so on
col4_blank1* string(2),
col5_blank2 *string(10),
col6 string(3);

as another post said, it helps if you try it yourself first.

End

Internet Services

Web Development

Digital Marketing

Consumer Tech