Improve query coverage for Go, Rust, Java, C++, C, Solidity, Ruby, Lua, Bash, Zig

Expands tree-sitter symbol extraction across 10 languages with new patterns for
structs/interfaces, fields, constants, constructors, namespaces, templates, and
more. Highlights per validation repo:

- Go (gin):           +52% symbols, now distinguishes struct/interface/field/const
- C++ (nlohmann/json): +62% symbols, adds namespaces, templates, aliases, concepts
- Zig (zig compiler): +35% symbols, adds fields, consts, test declarations
- Java (guava):       +33% symbols, adds fields, constants, constructors, records
- Solidity (OZ):      +15% symbols, adds modifiers, state vars, errors, UDVTs
- Rust (axum):        +3%, adds const_item/static_item

Dedup changes in extract.rs to support overlapping generic+specific patterns:
- `definition.struct` and `definition.trait` added as default capture kinds
- Same byte-range dedup now lets later (more specific) patterns win
- Added containment-based dedup for same-name/same-kind symbols to eliminate
  C++ template_declaration wrapper duplicates
- Strip surrounding quotes from string-literal names (Zig test names)

Index version bumped 7→8. 15 new tests covering all new patterns.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: ind-igo

src/index.rs

@@ -9,7 +9,7 @@ use std::time::{Duration, SystemTime, UNIX_EPOCH};
 
 use crate::language::{detect_language, parse_and_extract, primary_extension, LangError};
 
-pub const INDEX_VERSION: u32 = 7;
+pub const INDEX_VERSION: u32 = 8;
 
 /// Compute the cache path for a given project root.
 /// Returns `~/.cache/cx/indexes/<hash>.db` where hash is derived from the canonical path.

src/language/extract.rs

@@ -107,7 +107,14 @@ pub(super) fn extract_symbols(
         };
 
         let name = match name_n.utf8_text(source) {
-            Ok(s) => s.to_string(),
+            Ok(s) => {
+                // Strip surrounding quotes from string-literal names (e.g. Zig test names)
+                if s.starts_with('"') && s.ends_with('"') && s.len() >= 2 {
+                    s[1..s.len()-1].to_string()
+                } else {
+                    s.to_string()
+                }
+            }
             Err(_) => continue,
         };
 
@@ -158,6 +165,8 @@ fn resolve_kind(config: &LanguageConfig, capture_name: &str, node: &Node) -> Opt
         "definition.enum" => Some(SymbolKind::Enum),
         "definition.module" => Some(SymbolKind::Module),
         "definition.constant" => Some(SymbolKind::Const),
+        "definition.struct" => Some(SymbolKind::Struct),
+        "definition.trait" => Some(SymbolKind::Trait),
         "definition.event" => Some(SymbolKind::Event),
         "definition.field" => Some(SymbolKind::Field),
         "definition.macro" => Some(SymbolKind::Fn),
@@ -226,15 +235,26 @@ fn deduplicate(symbols: Vec<Symbol>) -> Vec<Symbol> {
 
     for sym in symbols {
         if let Some(&idx) = seen.get(&sym.byte_range) {
-            if deduped[idx].kind == sym.kind {
-                // Later pattern wins for same byte range + kind (more specific name capture)
-                deduped[idx] = sym;
-            }
+            // Later pattern wins for same byte range (more specific capture)
+            deduped[idx] = sym;
         } else {
             seen.insert(sym.byte_range, deduped.len());
             deduped.push(sym);
         }
     }
 
-    deduped
+    // Remove symbols whose range is strictly contained within another symbol
+    // of the same name (e.g. class_specifier inside template_declaration).
+    let ranges: Vec<_> = deduped.iter().map(|s| (s.name.as_str(), s.byte_range)).collect();
+    let mut keep = vec![true; deduped.len()];
+    for (i, (name_i, (start_i, end_i))) in ranges.iter().enumerate() {
+        for (j, (name_j, (start_j, end_j))) in ranges.iter().enumerate() {
+            if i != j && name_i == name_j && deduped[i].kind == deduped[j].kind && start_j <= start_i && end_i <= end_j && (start_j, end_j) != (start_i, end_i) {
+                keep[i] = false;
+                break;
+            }
+        }
+    }
+
+    deduped.into_iter().zip(keep).filter(|(_, k)| *k).map(|(s, _)| s).collect()
 }

src/language/mod.rs

@@ -118,7 +118,7 @@ static LANGUAGES: &[LanguageConfig] = &[
         sig_body_child: None,
         sig_delimiter: Some(b'{'),
         kind_overrides: &[],
-        ref_node_types: &["identifier"],
+        ref_node_types: &["identifier", "type_identifier"],
     },
     LanguageConfig {
         name: "ruby",

src/language/queries/bash.rs

@@ -1,4 +1,12 @@
 pub const QUERY: &str = r#"
+; --- Functions ---
+
 (function_definition
   name: (word) @name) @definition.function
+
+; --- Top-level variable assignments ---
+
+(program
+  (variable_assignment
+    name: (variable_name) @name) @definition.constant)
 "#;

src/language/queries/c.rs

@@ -1,4 +1,6 @@
 pub const QUERY: &str = r#"
+; --- Functions ---
+
 (function_definition
   declarator: (function_declarator
     declarator: (identifier) @name)) @definition.function
@@ -8,13 +10,34 @@ pub const QUERY: &str = r#"
     declarator: (function_declarator
       declarator: (identifier) @name))) @definition.function
 
+; --- Structs & unions ---
+
 (struct_specifier
   name: (type_identifier) @name
   body: (_)) @definition.class
 
+(union_specifier
+  name: (type_identifier) @name
+  body: (_)) @definition.class
+
+; --- Enums ---
+
 (enum_specifier
   name: (type_identifier) @name) @definition.enum
 
+; --- Type definitions ---
+
 (type_definition
   declarator: (type_identifier) @name) @definition.type
+
+; --- Function prototypes ---
+
+(declaration
+  declarator: (function_declarator
+    declarator: (identifier) @name)) @definition.function
+
+(declaration
+  declarator: (pointer_declarator
+    declarator: (function_declarator
+      declarator: (identifier) @name))) @definition.function
 "#;

src/language/queries/cpp.rs

@@ -1,4 +1,6 @@
 pub const QUERY: &str = r#"
+; --- Functions (also covers in-class constructors) ---
+
 (function_definition
   declarator: (function_declarator
     declarator: (identifier) @name)) @definition.function
@@ -17,16 +19,70 @@ pub const QUERY: &str = r#"
     declarator: (qualified_identifier
       name: (identifier) @name))) @definition.method
 
+; --- Destructors ---
+
+(function_definition
+  declarator: (function_declarator
+    declarator: (destructor_name
+      (identifier) @name))) @definition.function
+
+(function_definition
+  declarator: (function_declarator
+    declarator: (qualified_identifier
+      name: (destructor_name
+        (identifier) @name)))) @definition.function
+
+; --- Template functions ---
+
+(template_declaration
+  (function_definition
+    declarator: (function_declarator
+      declarator: (identifier) @name))) @definition.function
+
+(template_declaration
+  (function_definition
+    declarator: (function_declarator
+      declarator: (qualified_identifier
+        name: (identifier) @name)))) @definition.method
+
+; --- Classes & structs ---
+
 (struct_specifier
   name: (type_identifier) @name
   body: (_)) @definition.class
 
 (class_specifier
   name: (type_identifier) @name) @definition.class
 
+(template_declaration
+  (class_specifier
+    name: (type_identifier) @name)) @definition.class
+
+(template_declaration
+  (struct_specifier
+    name: (type_identifier) @name
+    body: (_))) @definition.class
+
+; --- Enums ---
+
 (enum_specifier
   name: (type_identifier) @name) @definition.enum
 
+; --- Type definitions & aliases ---
+
 (type_definition
   declarator: (type_identifier) @name) @definition.type
+
+(alias_declaration
+  name: (type_identifier) @name) @definition.type
+
+; --- Namespaces ---
+
+(namespace_definition
+  name: (namespace_identifier) @name) @definition.module
+
+; --- Concepts (C++20) ---
+
+(concept_definition
+  name: (identifier) @name) @definition.type
 "#;

src/language/queries/go.rs

@@ -1,10 +1,44 @@
 pub const QUERY: &str = r#"
+; --- Functions ---
+
 (function_declaration
   name: (identifier) @name) @definition.function
 
 (method_declaration
   name: (field_identifier) @name) @definition.method
 
+; --- Structs (must come before generic type_spec) ---
+
+(type_spec
+  name: (type_identifier) @name
+  type: (struct_type)) @definition.struct
+
+; --- Interfaces ---
+
+(type_spec
+  name: (type_identifier) @name
+  type: (interface_type)) @definition.interface
+
+; --- Other named types (catch-all for type aliases, func types, etc.) ---
+
 (type_spec
   name: (type_identifier) @name) @definition.type
+
+(type_alias
+  name: (type_identifier) @name) @definition.type
+
+; --- Constants ---
+
+(const_spec
+  name: (identifier) @name) @definition.constant
+
+; --- Interface method specs ---
+
+(method_elem
+  name: (field_identifier) @name) @definition.method
+
+; --- Struct fields ---
+
+(field_declaration
+  name: (field_identifier) @name) @definition.field
 "#;

src/language/queries/java.rs

@@ -1,13 +1,39 @@
 pub const QUERY: &str = r#"
+; --- Type declarations ---
+
 (class_declaration
   name: (identifier) @name) @definition.class
 
-(method_declaration
-  name: (identifier) @name) @definition.method
+(record_declaration
+  name: (identifier) @name) @definition.class
 
 (interface_declaration
   name: (identifier) @name) @definition.interface
 
+(annotation_type_declaration
+  name: (identifier) @name) @definition.interface
+
 (enum_declaration
   name: (identifier) @name) @definition.enum
+
+; --- Methods & constructors ---
+
+(method_declaration
+  name: (identifier) @name) @definition.method
+
+(constructor_declaration
+  name: (identifier) @name) @definition.method
+
+; --- Fields & constants ---
+
+(field_declaration
+  declarator: (variable_declarator
+    name: (identifier) @name)) @definition.field
+
+(constant_declaration
+  declarator: (variable_declarator
+    name: (identifier) @name)) @definition.constant
+
+(enum_constant
+  name: (identifier) @name) @definition.constant
 "#;

src/language/queries/lua.rs

@@ -1,4 +1,6 @@
 pub const QUERY: &str = r#"
+; --- Functions ---
+
 (function_declaration
   name: (identifier) @name) @definition.function
 
@@ -9,4 +11,12 @@ pub const QUERY: &str = r#"
 (function_declaration
   name: (method_index_expression
     method: (identifier) @name)) @definition.method
+
+; --- Top-level local variables (module tables, constants) ---
+
+(chunk
+  (variable_declaration
+    (assignment_statement
+      (variable_list
+        name: (identifier) @name))) @definition.constant)
 "#;

src/language/queries/ruby.rs

@@ -1,13 +1,29 @@
 pub const QUERY: &str = r#"
+; --- Classes & modules ---
+
+(class
+  name: (constant) @name) @definition.class
+
+(module
+  name: (constant) @name) @definition.module
+
+; --- Methods ---
+
 (method
   name: (_) @name) @definition.method
 
 (singleton_method
   name: (_) @name) @definition.method
 
-(class
-  name: (constant) @name) @definition.class
+(alias
+  name: (_) @name) @definition.method
 
-(module
-  name: (constant) @name) @definition.module
+; --- Constants ---
+
+(assignment
+  left: (constant) @name) @definition.constant
+
+(assignment
+  left: (scope_resolution
+    name: (constant) @name)) @definition.constant
 "#;