Parsing fragments

[andrewbaxter]

Is your feature request related to a problem? Please describe.
I sometimes need to parse fragments of files that are complete subtrees.

Describe the solution you'd like
ParserParsers for all rules?

Describe alternatives you've considered
I can add fake text around the fragment to turn it into a valid full document, but this messes with line numbers and if the user supplied a non-fragment it could end up making the error messages confusing (referring to input that the user didn't originally supply).

Additional context
Maybe this is already possible... I dug around and didn't see an obvious way to do it. The only ParserParser generated seems to be for the root of the grammar and I didn't see a simple way to set up the state for other rules.

[igordejanovic]

I’ve thought about this in the past. Although I haven’t had an urgent need for it, I’ve always thought it would be a nice feature for debugging/testing purposes to parse just a sub-language.

For some parsing approaches (e.g., recursive-descent/combinators), it is straightforward, but for LR parsing, it’s not as simple. However, I think it could be doable.

The biggest challenge, I assume, is how to determine which state from the LR DFA is the correct one to put the parser in and how to extend the API to specify that state or non-terminal. Additionally, it’s important to determine when parsing is complete and the entire sub-tree has been constructed.

I need some time to think about it.