topic Re: Read Dataset encoding problem on unicode programs in Application Development and Automation Discussions

Read Dataset encoding problem on unicode programs

kadir_ozgur — Sat, 12 Nov 2011 16:49:52 GMT

Dear all,

I have a file handling problem on ECC 6.04 upgrade from ECC 5.00. The 'Read Dataset' statements on 6.04 system are not working which are working properly on 5.00 system. I need to read text files. The packed numbers on these files are coming as #. When I check them on AL11 transaction I see them # again. When I debug the programs on non-unicode(5.00) system, I saw that the packed numbers are coming as # again but when I assign them to a variable which type is packed, they became 1.500 for example. I need to use character type variables on unicode(6.04) system, otherwise I get syntax error. Thats why I could not use a packed number to convert # character to real numbers. I have tried all Read Dataset additions(legacy, code pages, non-unicode, utf-8 etc.) but every time I got dump on assigning # variables to a character field. I think I need to get values from file as real numbers like 1.500. I tried the same programs with a unicode converted files but still not working. I need to see the file content as real numbers I think, or find a way to convert these # characters to a numbers wtihout getting dump.

All suggestions will be welcomed.

Thanks.

Re: Read Dataset encoding problem on unicode programs

Former Member — Sun, 13 Nov 2011 08:14:40 GMT

Can u tell whats the code you have written.

Also tell me one line of data in your file.

Re: Read Dataset encoding problem on unicode programs

kadir_ozgur — Sun, 13 Nov 2011 10:19:21 GMT

open dataset file for input in text mode encoding non-unicode message mess.

The code above is an example that I have used. I already used 'encoding utf-8, default' and 'legacy text mode' also.

K1056-15 and #####P# is an example line of my file.

K1056-15 is type afpo-aufnr

######P# is type mbew-lbkum.

I can get the afpo-aufnr type value correctly but mbew-lbkum type value come as ######P# and I can not assign this value to another structure or internal table field which has type mbew-lbkum. I got syntax error when I try that. But in non-unicode systems, when I assign value '######P#' to a field that has type mbew-lbkum, #####P# transforms to 1.500.

Thanks for your help Manjunath.

Re: Read Dataset encoding problem on unicode programs

Former Member — Mon, 14 Nov 2011 04:19:40 GMT

Sorry, i am not very sure what your issue is.

On your SAP Logon PAD , click on Change Item --> Go to Code Page tab

Check the Encoding details over there : try using 'DEFAULT - ANSI for UNICODE system'

see if this works