VBA code in Veda input files

[AKanudia]

Dear users,

Veda has been using functions from the local Excel installation to work with the files. This creates some vulnerability due to the increasingly wide variety of Excel installations. We are considering using an alternative, which is much faster and doesn't rely on the local Excel application. The only downside is that this library doesn't support VBA code. So, Veda will not be able to refresh files that have any VBA code.

The only case where this really makes a difference is when these files also have FILL tags, because Veda refreshes Excel files only after writing FILL tags (and for parameteric scenarios). If the VBA code does not interact with the output of FILL tags then it would be possible for users to reconfigure the data declarations and tolerate this limitation. But this approach will not work if the VBA code interacts with the output of FILL tags.

I have used VBA code in several cases but always for text processing. I do not have a single case where it interacts with FILL tag output.

We would like to know the kind of role VBA plays in Veda input files for other users. Please respond to this poll, so that we can take an informed decision. We will make a beta version available for testing very soon.

Thanks,
Amit

[Antti-L]

> If the VBA code does not interact with the output of FILL tags then it would be possible for users to reconfigure the data declarations and tolerate this limitation.

Could you elaborate on this?  How extensive would be the tasks that would be required for users to reconfigure the data declarations (which according to the above, would seemingly be needed for tolerating the limitation when VBA code does not interact with the output of FILL tags)?

[AKanudia]

The FILL tag and related tables would need to be moved to a file without VBA.

[Antti-L]

Ahh so, I see, thanks.
In other words, the whole Excel file becomes unreadable for the library, if it includes any VBA?
Would *.xlsm files with Fill tables (without VBA) still be readable for the library?

[AKanudia]

There is no problem reading files that have VBA. Basically, this library cannot *run* VBA. So, files with VBA cannot be recalculated with this library. Veda2.0 recalculates files only after writing in them, which is only for FILL tags (apart from parametric scenarios).

[Antti-L]

But I think you said that even VBA that is not used for calculating anything that needs to recalculated when processing FILL tables would not work, and any such file would no longer work, even if there is no VBA interaction?  Sorry for being slow understanding why that is so...

[AKanudia]

You are not slow This is a good point. The thing is that once we instruct this library to recalculate, it reports errors in the cells that call VBA code. We have written to the developers to see if they can leave the VBA cells alone while recalculating.

[AKanudia]

Beta version will be available later today

[Rahul]

Here is a Beta version for testing: Dropbox link (download)

[olexandr]

Personally I like the idea. I've just tested the beta version on TIM (https://github.com/esma-cgep/TIM/tree/7e...b090d8feb6). I have not noticed any processing speed improvements, but can confirm that this issue (https://github.com/kanors-emr/Veda2.0-In.../issues/22) is gone.

[Antti-L]

I've just tested a bit the latest version (3.0.2) and I can definitely see performance improvement for the Excel file reading.
Thus far this looks good, congrats! 

[AKanudia]

Thanks Antti

Not using the locally installed Excel libraries is also a huge improvement. Excel installation is no longer a prerequisite for Veda. We have removed Excel from Veda online servers.