pkra
commented
3 years ago I'm not sure how I feel about it. Authors can of course fix this themselves by randomly modifying duplicate equations (e.g., adding IDs - duh). They could also post-process to make IDs unique. So I don't know how far I'd go. Just a counter or full custom functions? It's a rabbit hole with limited gain, I think.
A common scenario is to go through a page of content and apply sre-to-tree to every equation. In this case, there can easily be identical equations, leading to identical ids. While this doesn't cause any real issues, it's invalid.
Since we cannot guess this here, authors might want to to customize the ids.