Skip to Content
avatar image
Former Member

File content is 'jumbled' or mixed up


We have a program which creates text files.

The content is output of table AENR and the related characteristic values of a change number. The file has over half a million records.

The issue is:

  • The output gets mixed up in the file sometimes.
  • The value in one cell appears in a different column or in a different row.
  • The issue occurs intermittently and there is no issue with the final output table when we do a debug and the session creates the file with perfect output.

Has anybody here experience similar issue?



Add comment
10|10000 characters needed characters exceeded

  • Get RSS Feed

2 Answers

  • May 24, 2017 at 07:41 PM

    Sorry, but your question is missing the most important information: how does your custom program work? What you say about the debug makes me think that you write in this file in parallel sessions. This won't work : every write to the dataset will do it "randomly", the operating system or the file system will receive blocks of data to write, and one block may arrive at the middle of another block. You can't write parallely. The only way to do it is to write to separate files (one per ABAP thread), and at the end you merge the files into one (of course, performance decreases a little bit, but there's no choice).

    Add comment
    10|10000 characters needed characters exceeded

  • Jul 24, 2017 at 07:02 AM

    Also - how are you looking at the resultant file ?

    If it's a tab delimited file then the position of the next 'column' can be dependant on the length of the data in the previous rows visually, but in the file it is in the right place.

    Add comment
    10|10000 characters needed characters exceeded