I have a problem reading from excel file source in DI. 3.1 version. The MAX string length for strings automatically is assigned to 255 characters. This means strings greater than 255 characters cannot be read into DI excel file format in DI. The string is truncated.
Is there a fix for this? How about 3.2? Is this corrected?
I’ve just tested this with Data Services 12.2.2 (XI 3.2 with Service Pack 2 applied).
Using Toad, I took one of our long text description tables (which contains converted CLOB to VARCHAR descriptions), I selected all item descriptions longer than 255 characters and exported the end result (incl. some key columns and varchar attribute columns) to an Excel sheet. (XLS in Excel 2000/2003 format).
I then created a new Excel File Format based on this Excel file in our DS Development area. When retrieving the metadata, DS actually listed this column as VARCHAR with lenght 0 and defaulted all other character columns to VARCHAR 255.
I changed the definition of the description column from VARCHAR (0) in the Excel File Format definition to VARHCAR 4000 which is the max Varchar length in Oracle 10g.
I then created a data flow, using this Excel File Format as source, with a Query transform and a new template table in the middle.
(Excel -> QRY -> Template Table).
There were no errors or warnings running this job and all 5076 rows from the Excel sheet were correctly imported.
I then ran this SQL to group and count records by description length to validate that none of the content had been cut-off at 255 characters:
select length(step_desc_char), count(1) from test_excel
group by length(step_desc_char)
order by 1
Recently we had the same problem, but I think I found the clue for this. In the Technical Manual I you can read that if the data integrator can not determinate the type of coloumn it will be set to varchar255. And the another piece of information is that the DI detect the type of the date only on the first 100, or I do not remember exactly on how much column…
So you have to revision the format of the excel, and do not let it solely to the BODI.
Does anyone know if there is a way around this issue?
Changing the formatting in Excel for the column or entire sheet to “Text” doesn’t resolve the issues. I think that’s what Medan was suggesting.
The column type is defined in the BODS XLS template object as VARCHAR 1000.
Yes. The spreadsheet file template has been manually created with VARCHAR columns of 1000 and regardless of this configuration it still crops all text strings being loaded to 255.
[list]
[:87a02f8401]SAP Data Servers version 14.0 (also called 4.0 by some)
[:87a02f8401]Windows Server 2008 R2
[*:87a02f8401]Client and servers are both on the Windows platform.
[/list]
I also tried experimenting with the same file saved as:
[list]
[:87a02f8401]97-2003 xls
[:87a02f8401]2010 xlsx
[/list]
…but it happens with both.
Could you make a test file available, so I can have a look and see if I can nor cannot replicate the problem with our own DS environment?
DS 4.0 has been out of mainstream support for a few years now. The most current version is DS 4.2 SP08 so you’re 2 minor product releases (4.1.0 and 4.2.0) and 6 Service Packs behind. (DS 4.2 SP02 was the GA release version).
It could well be that this issue was resolved somewhere in between these many releases? If you have a file available then I can see if the problem actually exists in DS 4.2 SP08. If I can’t replicate the problem, an upgrade may be overdue.
Note that if your company has been paying continuous maintenance fees to SAP, you are always eligible to download & upgrade to the latest version of DS.
I’m now seeing this issue on DS4.2sp8.1 and sp9.1. I’ve tried all the suggestions on this thread to no avail. So I’ve raised a ticket with SAP to see if they can shed any light on this!
I am having the same issue where it truncates the data to 255 characters when using Excel source no matter what I set the column to in the FF. The only time it runs correctly is if I put the biggest row in the first row of data.