Dear colleagues,
I hope this message finds you well.
Recently, I have been experimenting with xTerior in order to automate some of my work, and I have a few questions that may help achieve a clearer understanding of how this can be done optimally.
Here is the scenario: I'm working with some sort of deformation of the usual differential form setup. Essentially, I need to define a k-form whose exterior derivative is obtained by taking the usual derivative of its coefficient and then adding a d\Xi fator (which can be expressed as a linear combination of the usual dx-basis).
My approach so far goes as follows: to define the mentioned k-form structure, I started by expressing its coefficient as a combination of 0-forms and then combined these 0-forms with 1-forms dX that I previously defined to mimic the usual basis structure. The exterior derivative is implemented as the following: first, I compute the usual derivative of the k-form coefficient, and then add a d\Xi-factor to build the new (k+1)-form basis. The next step is to implement a Rule to rewrite d\Xi in terms of the usual dX.
This leads to my main questions:
1) First of all, is there a more efficient way to implement the above described scenario?
2) After performing the exterior derivative (as you can see in the attached notebook), is there a way to collect and organize all the dX ^ ... ^dX array terms, so that the resulting object can be expressed as a coefficient contracted with the usual dX-basis?
For clarity and to enrich the discussion, I have attached the mentioned notebook below.
Best regards,
João Matzenbacher.