MegaIng

This can't really be avoided. Contextual lexer doesn't quite do what you want, it's still limited by the LALR parser. If you want to do this kind of stuff, either...

@gideongrinberg It kinda is. We have a mirror that should match exactly what the actual parser accepts: [lark.lark](https://github.com/lark-parser/lark/blob/master/lark/grammars/lark.lark ), but that is not what is being used internally. That is...

@erezsh That actually sounds like a good idea. I would suggest that the argument takes a `Lark` instance (e.g. Something that has a `.parse(str) -> Tree` method). This would allow...

@erezsh This is an interesting use case for everything we are currently talking about.

The problem is `what is data?`, e.g. which element of the grammar is it?

And we will be even more responsive :-P. > The "data" is the top-level terminal in the grammar. Although erezsh did correctly spot the misunderstanding, this might not work since...

The existing interactive parser can be used to create a small wrapper that does this. ```python from queue import Queue from lark import Discard, Lark json_grammar = r""" ?start: "["...

@ThatXliner No, `_rules` should not be at grammar level, otherwise stuff like below has terrible performance ``` start: _options _options _options: a | b | c | d | e...

Why does it make it unreproducible? (And the website you linked is not that relevant for this situation from what I see)

And also considering that sphinx as a [warning about deactivating it](https://www.sphinx-doc.org/en/1.2/config.html#confval-html_copy_source) I am not sure that this by itself is a good enough reason to change it.