yeast: Support separate output node types in extractor generator

Language and LanguageSpec gain optional output_node_types field.
When set, the generator produces dbscheme/QL from the output types
and the extractor validates TRAP against them.

All existing extractors pass None (no behavior change).
Ruby extract() calls gain vec![] for the new rules parameter.

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>

Author: tausbn

ql/extractor/src/generator.rs

@@ -21,18 +21,22 @@ pub fn run(options: Options) -> std::io::Result<()> {
         Language {
             name: "QL".to_owned(),
             node_types: tree_sitter_ql::NODE_TYPES,
+            desugar: None,
         },
         Language {
             name: "Dbscheme".to_owned(),
             node_types: tree_sitter_ql_dbscheme::NODE_TYPES,
+            desugar: None,
         },
         Language {
             name: "Blame".to_owned(),
             node_types: tree_sitter_blame::NODE_TYPES,
+            desugar: None,
         },
         Language {
             name: "JSON".to_owned(),
             node_types: tree_sitter_json::NODE_TYPES,
+            desugar: None,
         },
     ];
 

ruby/extractor/src/generator.rs

@@ -21,10 +21,12 @@ pub fn run(options: Options) -> std::io::Result<()> {
         Language {
             name: "Ruby".to_owned(),
             node_types: tree_sitter_ruby::NODE_TYPES,
+            desugar: None,
         },
         Language {
             name: "Erb".to_owned(),
             node_types: tree_sitter_embedded_template::NODE_TYPES,
+            desugar: None,
         },
     ];
 

shared/tree-sitter-extractor/src/generator/language.rs

@@ -1,4 +1,9 @@
 pub struct Language {
     pub name: String,
     pub node_types: &'static str,
+    /// Optional yeast desugaring configuration. When set with an
+    /// `output_node_types_yaml`, the generator uses that YAML for the
+    /// dbscheme/QL library instead of `node_types`. The `rules` field is
+    /// unused at code-generation time; only the schema matters.
+    pub desugar: Option<yeast::DesugaringConfig>,
 }

shared/tree-sitter-extractor/src/generator/mod.rs

@@ -6,6 +6,7 @@ use std::io::Write;
 use std::path::PathBuf;
 
 use crate::node_types;
+use yeast;
 
 pub mod dbscheme;
 pub mod language;
@@ -68,7 +69,20 @@ pub fn generate(
         let token_name = format!("{}_token", &prefix);
         let tokeninfo_name = format!("{}_tokeninfo", &prefix);
         let reserved_word_name = format!("{}_reserved_word", &prefix);
-        let nodes = node_types::read_node_types_str(&prefix, language.node_types)?;
+        let effective_node_types: String = match language
+            .desugar
+            .as_ref()
+            .and_then(|c| c.output_node_types_yaml)
+        {
+            Some(yaml) => yeast::node_types_yaml::convert(yaml).map_err(|e| {
+                std::io::Error::other(format!(
+                    "Failed to convert YAML node-types to JSON for {}: {e}",
+                    language.name
+                ))
+            })?,
+            None => language.node_types.to_string(),
+        };
+        let nodes = node_types::read_node_types_str(&prefix, &effective_node_types)?;
         let (dbscheme_entries, mut ast_node_members, token_kinds) = convert_nodes(&nodes);
         ast_node_members.insert(&token_name);
         writeln!(&mut dbscheme_writer, "/*- {} dbscheme -*/", language.name)?;