So, according to you, {detri} is a relation between a timestamp / time address, and an event?
This would be an useful predicate indeed, but that seems to be very different from the relation described in the Gimste.
If {detri} has the meaning you're suggesting, then another predicate is necessary for relating a date structure to an event using a specific calendar system, or to convert a date structure into a timestamp / pure number address.
You seem to invest {li} with a great power, greater than what I'd have expected.
It seems that all we know is that {li} is provided with a numeric _expression_, that it must returns a pure number, but that the function/algorithm for interpreting the numeric _expression_ is entirely left to the context (and there's no way to explicitly provide one using {li}). So much more than merely the numeric base is left to the context, it seems.
If so, it becomes difficult to explicitly express which calendar system to use for evaluating the li-_expression_; it seems to me that the speaker would need to use an appropriate predicate for precisely stating how to interpret the _expression_, and avoid using {li}.