topic Re: Variable Length Files and ABAP in Application Development and Automation Discussions

Variable Length Files and ABAP

Former Member — Wed, 10 May 2006 19:57:26 GMT

I have a comma delimited file that is of a variable length.

The first 5 fields are fixed, but then I have a counter field and then 2 fields that repeat for each counter.

So if the counter field was 3, I’d have 6 additional fields, but if the counter was 5, I’d have 10 additional fields.

The goal is to end up with 3 records (since the counter was 3) with the 5 fixed fields then the 2 fields respectively.

Such as if it looks like this:

Abc,abc,abc,abc,abc,003,1a,1b,2a,2b,3a,3c

DCE,abc,abc,abc,abc,005,1a,1b,2a,2b,3a,3c,4a,4b,5a,5b

I need it to look like this or at least be able to read it similar to this.

Abc,abc,abc,abc,abc,003,1a,1b

Abc,abc,abc,abc,abc,003,2a,2b,

Abc,abc,abc,abc,abc,003,3a,3c

DCE,abc,abc,abc,abc,005,1a,1b

DCE,abc,abc,abc,abc,005,2a,2b

DCE,abc,abc,abc,abc,005,3a,3c

DCE,abc,abc,abc,abc,005,4a,4b

DCE,abc,abc,abc,abc,005,5a,5b

Re: Variable Length Files and ABAP

RichHeilman — Wed, 10 May 2006 20:05:32 GMT

Pam, you can try something like this.



report zrich_0001.

data: record type string.
data: counter type i.
data: index type i.
data: iout type table of string with header line.
data: irec type table of string with header line.

record = 'DCE,abc,abc,abc,abc,005,1a,1b,2a,2b,3a,3c,4a,4b,5a,5b'.

split record at ',' into table irec.

index  = 6.
read table irec index index.
counter = irec.

do counter times.

  clear iout.

  loop at irec.
    check sy-tabix <= 6.
    if sy-tabix = 1.
      iout = irec.
    else.
      concatenate iout irec into iout separated by ','.
    endif.
  endloop.

  do 2 times.
    index  = index + 1.
    read table irec index index.
    concatenate iout irec into iout separated by ','.
  enddo.

  append iout.

enddo.


loop at iout.
  write:/ iout.
endloop.

Regards,

Rich Heilman

Re: Variable Length Files and ABAP

Former Member — Wed, 10 May 2006 21:49:13 GMT

Hi,

I guess rich has written more efficient program anyways this is what I could do.

data : begin of i_line occurs 0,

line(100),

end of i_line.

data : begin of i_format occurs 0,

txt1(5),

txt2(5),

txt3(5),

txt4(5),

txt5(5),

txt6(5),

txt7(5),

txt8(5),

end of i_format.

data : comma type c,

off TYPE i,

moff TYPE i,

mlen TYPE i,

count type i,

len(3),

n type i.

start-of-selection.

i_line-line = 'Abc,abc,abc,abc,abc,003,1a,1b,2a,2b,3a,3c'.

append i_line.

clear i_line.

i_line-line = 'DCE,abc,abc,abc,abc,005,1a,1b,2a,2b,3a,3c,4a,4b,5a,5b'.

append i_line.

clear i_line.

comma = ','.

loop at i_line.

off = 0.

count = 0.

moff = 0.

while count lt 5.

count = count + 1.

find comma in section OFFSET off OF

i_line-line

MATCH OFFSET moff

MATCH LENGTH mlen.

if sy-subrc = 0.

off = moff + mlen.

write : / 'moff',moff,

'mlen', mlen..

endif.

endwhile.

moff = moff + 1.

write : i_line-line+moff(3) to len no-zero.

write 😕 'The len is ', len.

i_format-txt1 = i_line+0(3).

i_format-txt2 = i_line+4(3).

i_format-txt3 = i_line+8(3).

i_format-txt4 = i_line+12(3).

i_format-txt5 = i_line+16(3).

i_format-txt6 = i_line+20(3).

n = 24.

do len times.

i_format-txt7 = i_line+n(2).

n = n + 3.

i_format-txt8 = i_line+n(2).

n = n + 3.

append i_format.

enddo.

endloop.

loop at i_format.

write 😕 i_format.

endloop.

end-of-selection.

Regards

Vick