[RFC] Keeping more higher level structures for efficient data processing?

[Markus Elfring]

Hello,

Well-known data processing approaches can depend on various input files.
The input files are usually transformed into higher level structures.
This data transformation needs some resources and efforts.
https://en.wikipedia.org/wiki/Parsing

You might be looking for ways to avoid the repetition of such resources and efforts.
I imagine that an information system can be constructed which would manage
corresponding higher level structures for a while.
How will circumstances evolve for further software evolution in such directions?

Regards,
Markus