Problem with File

[Former Member]

Hello Experts,

In our SAP system a file is uploaded to the FTP server through some program.The structure of the file is

MATNR & Net Weight

8-108-447-283|0,521

8-104-930-756|4,489

8-108-222-232|0,022

8-104-546-656|0,042

8-104-546-657|0,051

8-104-937-094|5,417

8-104-919-598|0,308

8-104-146-444|0,195

But when is it uploaded to the Apllication server.It displays like this:

8-108-447-283 0,521#

8-104-930-756 4,489#

8-108-222-232 0,022#

8-104-546-656 0,042#

8-104-546-657 0,051#

8-104-937-094 5,417#

8-104-919-598 0,308#

8-104-146-444 0,195#

8-104-145-877 5,316#

This # symbol is added at the end of every line.

My task is to update the Netweight of the Material thro' BAPI and it is working fine.

Because of this # I am getting dump when I am translating the Char format of weight to Quantity format.

If I try to replace the # symbol by space, The hash symbol is not reconginsed.

Please tell me a solution to avoid the hash symbol when uploading into my report.

Thanks in Advance.

Vasanth

[Former Member]

Dear Sridhar

You statement works if our SAP system and the system sending data both are of same type(UNICODE/NON-UNICODE) when itz different like ours UNICODE and the source NON-UNICODE your statement will not work.

Above is just for your information.

Kind Regards

Eswar

[Former Member]

Hi Vasanth,

As per me I have taken the structure that you have specified and done a little bit of programming so rather that passing the sturcture that you have specified specify the structure as

8-108-447-283 0,521

8-104-930-756 4,489

8-108-222-232 0,022

8-104-546-656 0,042

8-104-546-657 0,051

8-104-937-094 5,417

8-104-919-598 0,308

8-104-146-444 0,195

here the funda is to pass the material as 18 characters. though it includes only 13 characters that you have passed in the flat file.

Cheers,

Pedhadodi Sudhakar

[Former Member]

Hi Vasanth

I guess the problem is the codepage mismatch between the source system and your system.

If you are working on Unicode System and the source system is NON-UNICODE, open the dataset with addition NON-UNICODE.

Unicode enabled system will show CR & LF characters as # when you try to view them in debugging.

Kind Regards

Eswar

Note: Reward for helpful answers

[sridharreddy_kondam]

Hi Vasanth,

Does the file from the application server has fixed length with 21 chars per line(including Hash)?

Have u tried with this

REPLACE FIRST OCCURRENCE OF '#' IN NetWeight WITH ''.

Regards,

Sridhar

[Former Member]

Hi Sridhar,

I have tried like this.

<b>SPLIT L_F_LINE AT '|' INTO G_T_DATA-MATNR G_T_DATA-BRGEW.

REPLACE ALL OCCURRENCES OF '#' IN G_T_DATA-BRGEW WITH SPACE.</b>

But the problem is the # symbol is not recongised by the system.

Could u please tell me the solution for this.

Thanks

Vasanth

[Former Member]

HI

'#' will not be recognized when you try to replace as it shows CR and LF as '#' characters in debugging whereas they are different characters. This is due to character type conversion in UNICODE enabled systems.

Kind Regards

Eswar

[sridharreddy_kondam]

Hi Vasanth,

I have tried this and its working ...

data : G_T_DATA-MATNR type matnr,

G_T_DATA-BRGEW(6) type c.

data s1 type string.

s1 = 'abcdefgh | 23456#'.

SPLIT S1 AT '|' INTO G_T_DATA-MATNR G_T_DATA-BRGEW.

*REPLACE ALL OCCURRENCES OF '#' IN G_T_DATA-BRGEW WITH SPACE.

REPLACE FIRST OCCURRENCE OF '#' IN G_T_DATA-BRGEW WITH ''.

write G_T_DATA-BRGEW.

Regards,

Sridhar

[Former Member]

Hello Eswar,

I am using this code to open the file from Application server

<b> CLEAR: G_T_DATA,L_F_LINE.

REFRESH: G_T_DATA.

OPEN DATASET P_FILE FOR INPUT IN TEXT MODE ENCODING NON-UNICODE.

*DEFAULT.

IF SY-SUBRC EQ 0.

DO.

READ DATASET P_FILE INTO L_F_LINE.

IF SY-SUBRC = 0.

• rpd2kor 09.08.06 -sm

G_T_DATA-MATNR = L_F_LINE(13).

WRITE L_F_LINE+14 TO G_T_DATA-BRGEW.

SHIFT G_T_DATA-BRGEW RIGHT DELETING TRAILING SPACE.

• SPLIT L_F_LINE AT '|' INTO G_T_DATA-MATNR G_T_DATA-BRGEW.

• REPLACE ALL OCCURRENCES OF '#' IN G_T_DATA-BRGEW WITH SPACE.

• rpd2kor 09.08.06 -em

CALL FUNCTION 'CONVERSION_EXIT_MATN1_INPUT'

EXPORTING

INPUT = G_T_DATA-MATNR

IMPORTING

OUTPUT = G_T_DATA-MATNR.

APPEND G_T_DATA.

CLEAR: G_T_DATA,L_F_LINE.</b>

Now what happends is when I am reading the line from <b>P_FILE the L_F_LINE = 8-108-447-283|0,521#</b>

What woulb be the probable error of the issue.

Thanks

Vasanth

[Former Member]

Hello Sridhar,

As u said Its working for me if the I pass the string as

s1 = 'abcdefgh | 23456#'.

Since here s1 is having the hash symbol. But the # is not actually available when the file is imported to FTP server from CAD system. When it is added to the FTP server of the SAP system, the # symbol is automatically added at the end of the each line.

Thanks

Vasanth

[sridharreddy_kondam]

Hi Vasanth,

Try using this statement...

<b>OPEN DATASET P_FILE FOR INPUT IN TEXT MODE ENCODING DEFAULT.</b>

DO.

READ DATASET P_FILE INTO v_string.

-->now use ur code either with REPLACE or the Function module

IF sy-subrc <> 0.

EXIT.

ENDIF.

Regards ,

Sridhar

[Former Member]

Hi Vasanth

Can you try adding "IGNORING CONVERSION ERRORS". Also check if source system sends data in NON-UNICODE or UTF-8/16 format.

Kind Regards

Eswar

[Former Member]

Dear Sridhar

You statement works if our SAP system and the system sending data both are of same type(UNICODE/NON-UNICODE) when itz different like ours UNICODE and the source NON-UNICODE your statement will not work.

Above is just for your information.

Kind Regards

Eswar