Xcelerator problem with more than 255 columns

[jhauswir]

Hello,

First of all thank you for this forum and the community behind it. It saved my life a couple of times! So here is my problem.

I'm currently facing a strange behavior with Xcelerator 10.2.1 on Excel 2010. I'm trying to create a spreadsheet with more than 255 columns and I see that the 255th is copied in the 256th. This is very strange because in the "Cube viewer" from Architect the values are correct for all the columns. As far as I know the 256 columns is the xls limitation.

I tried to save the file in xlsx but nothing changed. I tried to google the problem but I didn't find anything, maybe because most of the people didn't have so many columns... Maybe it has something to deal with the way TM1 store the data indexes, but I'm quite not sure about that...

Did anyone had this problem. Or did you have any idea?

Thank you very much.

Julien

[dan.kelleher]

Hi Julien,

Is this an Active Form you are creating? If so, it's probably something to do with the VBA behind the Active Form being originally developed in an older version of Excel. What is the purpose of the report/ entry sheet? 255 columns is a lot. Is there no way to transpose your spreadsheet onto rows, separate onto 2 separate sheets?

Thanks,

Dan

[jhauswir]

Hi Dan,

thanks for your reply. No in fact it's an ISB. The business wanted to have those data this way, I know it's a lot of columns but it's a very big spreadsheet, I tell our business to do that in a different way, but obviously the behavior of the application is not the one expected. I don't know if anyone had the same problem?

Regards

Julien

[mvaspal]

hi
If it is an ISB, I remember a post here on the forum where the issue was that the ISB has the old Excel's row and column limitation. Not sure it has been resolved since then.
Try to search for this in the forum, you should find it.

[lotsaram]

This limitation seems to have been solved (why I don't know as is seems quite pointless, god only knows there are more important things that the IBM developers could have patched up in Perspectives). I think the problem might just be that the Excel file was ORIGINALLY in .xls format and this is somehow "internally remembered" somewhere.

Using CX 10.1 and Excel 2010 with a brand new file in .xlsx format I just tested both an ISB and a slice with > 1500 columns. Both worked without issue.

Not that I would ever recommend designing or using a workbook with even 10% as many columns. I sure there are more sensible ways to solve whatever the business problem is, but it does appear to work from a pure technology perspective.
Using the same software combination ISB and slice of > 65536 rows seems to work as well, also NOT RECOMMENDED. (note, when I say "work" you can take that quite loosely. An ISB with 80K rows was actually quite reasonable and took about 1.5 minutes to return the data. A slice containing the same array of cells on the other hand required about 20 minutes. Again, NOT RECOMMENDED.)

[jhauswir]

Hello,

thanks for all your answers. I think a have to teach a little bit more our business about the best practices. Anyway, thanks for your answer.

[jim wood]

This limitation seems to have been solved (why I don't know as is seems quite pointless, god only knows there are more important things that the IBM developers could have patched up in Perspectives).

While I'm sure that IBM prioritise all bugs that are reported to them, you would still hope they would attempt to fix them all at some point. After all, if this has been posted on here twice that would at least show you that people are using it? Also keep in mind it depends on who reports it not just where it lies as product priority. I'm sure if you'd have reported it, it would have been fixed sooner?