File Content Conversion (receiver) and special characters

[Former Member]

Hi all,

I have a scenario that has a file receiver channel with content conversion. The record structure in the flat file is field-width delimited (hence no field separator) and the parameter 'fieldLengthTooShortHandling' has the value 'Cut' because the receiving system needs only specific widths for the fields. Hence if the field value exceeds the length permitted, the extra characters are clipped.

I observed that some characters are not handled properly while creating the text file. For example, one of the fields contained a "minus" character (not the hyphen). The flat file was created successfully. I opened the file in notepad and found that the "minus" character has appeared correctly and the column count in that record was as expected. However when the same file was opened in Textpad, the minus character was displayed as â | | ('a' with caret, bar, bar) So, all the fields after this field were shifted ahead by 2 characters and hence the total column count of the record had gone beyond the actual one.

All this started due to the error reported by the receiver system which processes the flat file. Due to shift of characters in the flat file, the processing failed. Moreover that system cannot process the special characters (like minus or non-Latin accented characters etc.) So although there is no issue in XI interface as such, I just want to know if anyone has more informtion on why the characters are displayed differently as mentioned above.

Regards,

Shankar

[stefan_grube · Accepted solution]

Find out which character encoding the receiver system expects and put the same to the receiver channel.

By default the file adapter would create a file with UTF-8, which might not be useable for the receiver.

Stefan

[Former Member]

Define data type like

order_recordset

order_row 1..unbound

f1

f2

All are same except communication channel configuration,

Message Protcol : File content select, then below you got additional parameters.

there you fill

Document name : Your sender message type.

Document namespace : Give your scenario namespace

Recordset name : order_recordset ( mentioned in the data type)

Recordset structure : order_row, *

Name Value

order_recordset.fieldSeparator : 'nl'

order_row.fieldSeparator : ,

order_row.endSeparator : 'nl'

based on your text file you fill the above parameter values.