Delimited Text ASCII Error

[Former Member]

Hello,

I am trying to import a vendors list that I extracted from table LFA1 of an R/3 system. I saved it as unconverted list in a file called vendors.txt.

I like to use it to populate the vendors lookup table of my MDM repository. But when trying to open the file with the MDM Import Manager and having delimited text selected I run into the following error:

Error Connecting to source

Logon Error: xFreedBlockTextStreamer: Found non-conforming characters in input stream . Line 9 .

When looking at line 9 of the file I can see that there are some characters like ä,ö or å.

Did somebody experience this type of problem? Is there any way to overcome this?

Thanks and Warmest Regards,

Ingo

[Former Member · Accepted solution]

Ingo,

Try to open your file in notepad and save it in UTF-8 encoding

BR,

Vadim

[Former Member · Accepted solution]

Hi Vadim,

This solved the problem, thanks!

Warmest Regards,

Ingo

[Former Member]

Hi Ingo,

I received the same logon error as you did,and I too have the same kind of non Ascii characters (âjâbâRü[âV ) in my text file as you have in your file.

When I analysed my Text file correctly,I found that my source text file was not properly delimited.

A complete record was under one field instead of being splitted into several fields.

But this problem i was able to analyse only when i opened my .txt file as .xls.

So Just try this:

1) Open your .txt file as .xls and the go through every field to see if the values are properly distributed.

2)You can always maintain the leading Zeroes in Excel by making those field type as type TEXT instead of General.

3) Even if your Rep is not multilingual.still this kind of data can be imported into MDM without any difficulty if you import it using an Excel Mapping.

I know you do not prefer to use Excel but your problem can be solved using it and besides the only problem that you are facing about the leading zeroes being lost can also be handled easily.

Hope it helps.

Thanks & Regards

Simona

[Former Member]

Hi Ingo,

Explanation: You are facing this problem because the text file contains special characters.These are actually multilingual characters,your source had some data which were in some other language other than 'English' thats why MDM was not able to import this data.

Solution:

Check to see if your Repository is Multilingual.

Make the Field which is containing this data as multilingual as well.

If it still does not work then check your text filr properly.the source may not be correctly Delimited.Try then to convert this .txt file to .xls and try importing it will work fine.

Hope it helped

Reward Points if found useful.

Thanks & Regards

Simona

[Former Member]

Hi Simona,

Thanks for your help!

Let's check your solution:

The repository we are using is multilingual. Some fields in the repository are multilingual as well. But for a lookup field like vendor you would not want it to be multilingual as you are saving only the name in its original language, not the translations of it. Anyway the error I described occurs before any mapping can be done, so this shouldn't be the source of the error.

I checked my file and it's correctly delimited. I tried it again using a very simple example file, and it gets imported as long as there are no non-english characters like ä,ö.. But if I add some of those the error shows up.

The conclusion is that the MDM Import Manager does not accept non-english characters from text files.

I know this works when importing from Excel-files. Anyway R/3 does not write the leading zeros for the vendor code into the Excel, it writes 2021308 instead of 0002021308. But with text files it does. Because of this I got into the situation of importing text files.

Thanks & Best Regards,

Ingo