codegraphcontext-rust 0.9.12__tar.gz → 0.9.14__tar.gz

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codegraphcontext-rust
-Version: 0.9.12
+Version: 0.9.14
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "codegraphcontext-rust"
-version = "0.9.12"
+version = "0.9.14"
 description = "Rust-accelerated code indexing engine for CodeGraphContext. 19 languages, 5-18x faster parsing via tree-sitter + rayon."
 authors = [
     { name = "Shashank Shekhar Singh", email = "shashankshekharsingh1205@gmail.com" },

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/rust/cgc-core/src/parser.rs

@@ -73,6 +73,7 @@ pub fn parse_file(
 pub fn parse_files_parallel(
     file_specs: &[(String, String, bool)], // (path, lang, is_dependency)
     num_threads: Option<usize>,
+    index_source: bool,
 ) -> Vec<ParseResult> {
     use rayon::prelude::*;
 
@@ -84,7 +85,7 @@ pub fn parse_files_parallel(
     pool.install(|| {
         file_specs
             .par_iter()
-            .map(|(path, lang, is_dep)| parse_file(path, lang, *is_dep, false))
+            .map(|(path, lang, is_dep)| parse_file(path, lang, *is_dep, index_source))
             .collect()
     })
 }
@@ -94,6 +95,7 @@ pub fn parse_files_parallel(
 pub fn parse_and_prescan_parallel(
     file_specs: &[(String, String, bool)], // (path, lang, is_dependency)
     num_threads: Option<usize>,
+    index_source: bool,
 ) -> (Vec<ParseResult>, HashMap<String, Vec<String>>) {
     use rayon::prelude::*;
 
@@ -107,7 +109,7 @@ pub fn parse_and_prescan_parallel(
         file_specs
             .par_iter()
             .map(|(path, lang, is_dep)| {
-                let result = parse_file(path, lang, *is_dep, false);
+                let result = parse_file(path, lang, *is_dep, index_source);
 
                 // Extract pre-scan names from parsed data
                 let resolved = fs::canonicalize(Path::new(path))

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/rust/cgc-python/src/lib.rs

@@ -27,13 +27,14 @@ fn parse_file(
 }
 
 #[pyfunction]
-#[pyo3(signature = (file_specs, num_threads=None))]
+#[pyo3(signature = (file_specs, num_threads=None, index_source=false))]
 fn parse_files_parallel(
     py: Python<'_>,
     file_specs: Vec<(String, String, bool)>,
     num_threads: Option<usize>,
+    index_source: bool,
 ) -> PyResult<Vec<PyObject>> {
-    let results = py.allow_threads(|| parser::parse_files_parallel(&file_specs, num_threads));
+    let results = py.allow_threads(|| parser::parse_files_parallel(&file_specs, num_threads, index_source));
     results
         .into_iter()
         .map(|r| conversions::parse_result_to_py(py, r))
@@ -54,14 +55,15 @@ fn pre_scan_for_imports(py: Python<'_>, file_specs: Vec<(String, String)>) -> Py
 /// Parse files in parallel AND build imports_map in one pass.
 /// Returns (list_of_file_data_dicts, imports_map_dict).
 #[pyfunction]
-#[pyo3(signature = (file_specs, num_threads=None))]
+#[pyo3(signature = (file_specs, num_threads=None, index_source=false))]
 fn parse_and_prescan(
     py: Python<'_>,
     file_specs: Vec<(String, String, bool)>,
     num_threads: Option<usize>,
+    index_source: bool,
 ) -> PyResult<PyObject> {
     let (results, imports_map) = py.allow_threads(|| {
-        parser::parse_and_prescan_parallel(&file_specs, num_threads)
+        parser::parse_and_prescan_parallel(&file_specs, num_threads, index_source)
     });
 
     let file_data_list: Vec<PyObject> = results

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/src/codegraphcontext/tools/indexing/engine.py

@@ -43,12 +43,13 @@ def parse_file(path, lang, is_dependency=False, index_source=False,
     return parser_wrapper.parse(path, is_dependency, index_source=index_source, **kwargs)
 
 
-def parse_files_parallel(file_specs, num_threads=None):
+def parse_files_parallel(file_specs, num_threads=None, index_source=False):
     """Parse multiple files in parallel using Rust.
 
     Args:
         file_specs: list of (path_str, lang, is_dependency) tuples
         num_threads: optional thread count (0 = auto)
+        index_source: if True, extract docstrings and source bodies
 
     Returns:
         list of file_data dicts
@@ -59,7 +60,7 @@ def parse_files_parallel(file_specs, num_threads=None):
     if not RUST_AVAILABLE:
         raise RuntimeError("Rust engine required for parallel parsing")
 
-    return _rust_parse_files_parallel(file_specs, num_threads)
+    return _rust_parse_files_parallel(file_specs, num_threads, index_source)
 
 
 def pre_scan_for_imports(file_specs):

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/src/codegraphcontext/tools/indexing/pipeline.py

@@ -84,7 +84,7 @@ async def run_tree_sitter_index_async(
             info_logger(f"Rust combined parse+prescan for {len(rust_files)} files...")
             t_parse = time.time()
             specs = [(str(f), lang, is_dependency) for f, lang in rust_files]
-            rust_results, imports_map = _rust_parse_and_prescan(specs)
+            rust_results, imports_map = _rust_parse_and_prescan(specs, index_source=True)
             info_logger(f"Rust parse+prescan done in {time.time() - t_parse:.1f}s "
                         f"({len(imports_map)} symbols)")
 

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/src/codegraphcontext/tools/languages/python.py

@@ -264,9 +264,10 @@ class PythonTreeSitterParser:
                     "is_dependency": False,
                 }
 
+                # Always extract docstring (lightweight, no index_source needed)
+                func_data["docstring"] = self._get_docstring(body_node)
                 if self.index_source:
                     func_data["source"] = self._get_node_text(func_node)
-                    func_data["docstring"] = self._get_docstring(body_node)
 
                 functions.append(func_data)
         return functions
@@ -302,9 +303,10 @@ class PythonTreeSitterParser:
                     "lang": self.language_name,
                     "is_dependency": False,
                 }
+                # Always extract docstring (lightweight, no index_source needed)
+                class_data["docstring"] = self._get_docstring(body_node)
                 if self.index_source:
                     class_data["source"] = self._get_node_text(class_node)
-                    class_data["docstring"] = self._get_docstring(body_node)
 
                 classes.append(class_data)
         return classes

{codegraphcontext_rust-0.9.12 → codegraphcontext_rust-0.9.14}/src/codegraphcontext/tools/languages/typescript.py

@@ -141,6 +141,36 @@ class TypescriptTreeSitterParser:
     def _get_docstring(self, body_node):
         return None
 
+    def _get_jsdoc(self, node) -> str | None:
+        """Extract JSDoc /** ... */ comment immediately preceding node."""
+        parent = node.parent
+        if not parent:
+            return None
+        siblings = list(parent.children)
+        try:
+            idx = siblings.index(node)
+        except ValueError:
+            return None
+        for i in range(idx - 1, -1, -1):
+            sib = siblings[i]
+            if sib.type == "comment":
+                raw = self._get_node_text(sib)
+                if raw.startswith("/**"):
+                    lines = []
+                    for line in raw.splitlines():
+                        line = line.strip().lstrip("/*").strip()
+                        if line.startswith("@"):
+                            break
+                        if line:
+                            lines.append(line)
+                    result = " ".join(lines).strip()
+                    return result[:500] if result else None
+                elif raw.startswith("//"):
+                    return raw.lstrip("/").strip()[:500]
+            elif sib.type not in ("\n", "newline"):
+                break
+        return None
+
     def parse(self, path: Path, is_dependency: bool = False, index_source: bool = False) -> Dict:
         self.index_source = index_source
         with open(path, "r", encoding="utf-8") as f:
@@ -233,7 +263,6 @@ class TypescriptTreeSitterParser:
                 args = [self._get_node_text(data['single_param'])]
             context, context_type, _ = self._get_parent_context(func_node)
             class_context = context if context_type == 'class_declaration' else None
-            docstring = None
             func_data = {
                 "name": name,
                 "line_number": func_node.start_point[0] + 1,
@@ -246,11 +275,12 @@ class TypescriptTreeSitterParser:
                 "decorators": [],
                 "lang": self.language_name,
                 "is_dependency": False,
+                # Always extract JSDoc (lightweight, no index_source needed)
+                "docstring": self._get_jsdoc(func_node),
             }
 
             if self.index_source:
                 func_data["source"] = self._get_node_text(func_node)
-                func_data["docstring"] = docstring
             functions.append(func_data)
         return functions
 
@@ -317,9 +347,10 @@ class TypescriptTreeSitterParser:
                     "lang": self.language_name,
                     "is_dependency": False,
                 }
+                # Always extract JSDoc (lightweight, no index_source needed)
+                class_data["docstring"] = self._get_jsdoc(class_node)
                 if self.index_source:
                     class_data["source"] = self._get_node_text(class_node)
-                    class_data["docstring"] = self._get_docstring(class_node)
                 classes.append(class_data)
         return classes