Getting at fact_id

[jay9909]

Hi! I'm building a service that needs the fact_id attribute (the original id from the iXBRL element) for source-linking back to SEC filings. I can see it's captured during parsing in _extract_facts (instance.py:394) and stored on the Fact model (models.py:177), but it doesn't surface in the fact_dict built by get_facts() (facts.py:967-976), so it's not available through to_dataframe().

I'm currently iterating over xbrl._facts (xbrl.py:241) to access the underlying Fact objects directly and grab .fact_id from there. It works well in my testing, but obviously it's a code smell reaching into a private data structure. Just want to ask: are there any gotchas I should be aware of with this approach? Is the _facts dict / Fact model structure reasonably stable, or is it an area might change? Or maybe is there an available part of the API I've just overlooked entirely to get at it?

Thanks for the amazing library. It's incredible.

[dgunning]

@-

[jay9909]

Saw this implemented in commit 0785b87. Thank you!