Variable data sets and varying font styles

[JacksonDavis]

I'm making a load of thumbnails for some YouTube videos which has the episode number in the top right corner with a little superscript hash before it. I'm trying to import the data set variables using a txt file and export accordingly, however as expected, everything displays as superscript instead of keeping their separate states.

Is there a way to have only have the first character display in super script and the ones after that in normal text when using data sets?

I would use a separate layer for the #, however, because the text is aligned to the right and the numbers run into the double figures, the fluctuation in values causes the text to shift into it.

Wondering if anyone has a solution for this, or any ideas about how I could go about this!

Thanks guys, wanting to save some time with this as I don't fancy changing the values manually, but might have to at this point!

[JJMack]

I have nevet use Data Driven graphics.

You may need to have the replacement text switch back to normal text.  Or he hash character may be acting a control character in the text replacement text  insert a escape control character brfore the hash so the hash is use as a text hash not a control character if that does not work try using two hash characters  instead of one. 

It may be easiest to just make the superscript hash its own layer and not even replace it  just replace the text number the is positioned following the Superscript hash.  

[Silly-V]

I wonder if you'll be able to use action processing to move each superscript item into place somehow?

[r-bin]

If I understand correctly.

You can use two text layers with different sizes for one variable. In the first layer with the help of a mask, show only the first character, in the second, show everything starting with the second character.

[Jarda Bereza]

In batch processing you can record rectangular selection and then align layer to that selection e.g. left top