Closed tirix closed 1 year ago
There was no feedback to this PR, but it looks like we are moving towards an architecture where $j
commands are handled separately in each "pass".
As a result, command check results are available in the corresponding pass, (e.g. provable
typecode in the Grammar pass, and primitive
commands in the definition pass) so there is no need to store them in a central place.
So I will close this PR if there is no objection.
A proposal to move
$j
command interpretation to a lazily evaluatedCommands
struct, as discussed in #103. Currently only supports thesyntax
command, but more could be added.Grammar
can be later refitted to use this struct.