Usage of the push-scoped-node type

[oppiliappan]

Thanks for exposing this as a crate! I had a quick question about using the DSL, in an example like this:

(identifier) @id {
  node new_exported_scope_node
  attr (new_exported_scope_node) is_exported
  node new_push_scoped_symbol_node
  attr (new_push_scoped_symbol_node)
    type = "push_scoped_symbol",
    symbol = (source-text @id),
    scope = new_exported_scope_node   ; how does one refer to a "parent" node here
}

the new_push_scoped_symbol_node has to originate from the capture, or be created within the same stanza, if I understand the capabilities of the DSL. In such a scenario, how does one refer to, say, the "parent" node (to set the scope attribute to parent, for example)? I may be misunderstanding the usage of scoped nodes, so perhaps such a necessity never arises, please do correct me if so.

Thanks!

[hendrikvanantwerpen]

Hi @nerdypepper! You are correct, it is actually very common to make a connection between a child node and its parent. The idiomatic way to do that is not from the child but from the parent node.

For example, see how the .lexical_scope variable is set for all statements in a block as here:

statement_block
  (_)@stmt
)@block {
  ; propagate lexical scope
  edge @stmt.lexical_scope -> @block.lexical_scope

[oppiliappan]

Ah I thought as much, thanks for clarifying, and thanks for the link!