codegraphcontext-rust 0.9.19__tar.gz → 0.9.21__tar.gz

{codegraphcontext_rust-0.9.19 → codegraphcontext_rust-0.9.21}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: codegraphcontext-rust
-Version: 0.9.19
+Version: 0.9.21
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11

{codegraphcontext_rust-0.9.19 → codegraphcontext_rust-0.9.21}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "codegraphcontext-rust"
-version = "0.9.19"
+version = "0.9.21"
 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-0.9.19 → codegraphcontext_rust-0.9.21}/rust/cgc-core/src/resolution/calls.rs

@@ -351,6 +351,22 @@ pub struct CallGroups {
     pub cls_to_cls: Vec<ResolvedCall>,
     pub file_to_fn: Vec<ResolvedCall>,
     pub file_to_cls: Vec<ResolvedCall>,
+    /// Calls that produced no edge, with the reason (Python builtins excluded).
+    pub unresolved: Vec<UnresolvedCall>,
+}
+
+/// A call the resolver dropped, kept for diagnostics (upstream: `record_skip`).
+#[derive(Debug, Clone)]
+pub struct UnresolvedCall {
+    pub caller_file_path: String,
+    pub caller_name: Option<String>,
+    pub line_number: usize,
+    pub called_name: String,
+    pub full_call_name: String,
+    /// Receiver type the extractor inferred, if any
+    pub receiver_type: Option<String>,
+    /// external_receiver_type | external_chain_type | receiver_namesake | skip_external
+    pub reason: &'static str,
 }
 
 /// Locate the file declaring type `ty`. Returns (file, exact): exact when
@@ -375,8 +391,9 @@ fn locate_type(
     Some((paths[0].clone(), false))
 }
 
-/// Resolve a single function call to its target.
-pub fn resolve_function_call(
+/// Resolve a single function call to its target; `Err(reason)` when the
+/// call produces no edge.
+pub fn resolve_call(
     call: &CallInput,
     caller_file_path: &str,
     local_names: &HashSet<String>,
@@ -384,7 +401,7 @@ pub fn resolve_function_call(
     imports_map: &HashMap<String, Vec<String>>,
     types: &TypeIndex,
     skip_external: bool,
-) -> Option<ResolvedCall> {
+) -> Result<ResolvedCall, &'static str> {
     let called_name = &call.name;
     let full_call = &call.full_name;
 
@@ -393,7 +410,7 @@ pub fn resolve_function_call(
     // from another language — is a regular method call.
     let is_python = caller_file_path.ends_with(".py") || caller_file_path.ends_with(".pyi");
     if is_python && !full_call.contains('.') && PYTHON_BUILTINS.contains(&called_name.as_str()) {
-        return None;
+        return Err("python_builtin");
     }
 
     let base_obj = if full_call.contains('.') {
@@ -514,7 +531,7 @@ pub fn resolve_function_call(
                     resolved = Some((found.map(|(_, f, _)| f).unwrap_or(file), 4));
                 }
                 // e.g. repo.findAll().stream(): target is outside the graph
-                ChainType::External => return None,
+                ChainType::External => return Err("external_chain_type"),
                 ChainType::Unknown => {}
             }
         }
@@ -581,7 +598,7 @@ pub fn resolve_function_call(
             // The receiver's type is known but lives outside the repo: the
             // target can't be in the graph (upstream: receiver_resolution_failed).
             if external_receiver_type {
-                return None;
+                return Err("external_receiver_type");
             }
             // Pointing it at the caller's file would bind it to a same-named
             // local function (false recursion), so drop it in that case —
@@ -589,7 +606,7 @@ pub fn resolve_function_call(
             if local_names.contains(called_name.as_str())
                 || call.context_name.as_deref() == Some(called_name.as_str())
             {
-                return None;
+                return Err("receiver_namesake");
             }
             resolved = Some((own(), 9));
         } else if !qualified_other && local_names.contains(called_name.as_str()) {
@@ -606,7 +623,7 @@ pub fn resolve_function_call(
                 })
                 .collect();
             if candidates.is_empty() && qualified_other && local_names.contains(called_name.as_str()) {
-                return None;
+                return Err("receiver_namesake");
             }
             // Try matching via local imports
             for p in &candidates {
@@ -628,7 +645,7 @@ pub fn resolve_function_call(
     }
 
     if skip_external && is_unresolved_external {
-        return None;
+        return Err("skip_external");
     }
 
     let (resolved_path, tier) = resolved.unwrap_or_else(|| (own(), 9));
@@ -638,7 +655,7 @@ pub fn resolve_function_call(
     if let (Some(name), Some(_), Some(line)) =
         (&call.context_name, &call.context_type, call.context_line)
     {
-        Some(ResolvedCall {
+        Ok(ResolvedCall {
             call_type: "function".to_string(),
             caller_name: Some(name.clone()),
             caller_file_path: caller_file_path.to_string(),
@@ -654,7 +671,7 @@ pub fn resolve_function_call(
             dispatch_class: dispatch_class.clone(),
         })
     } else {
-        Some(ResolvedCall {
+        Ok(ResolvedCall {
             call_type: "file".to_string(),
             caller_name: None,
             caller_file_path: caller_file_path.to_string(),
@@ -672,6 +689,19 @@ pub fn resolve_function_call(
     }
 }
 
+/// Resolve a single function call to its target (None when dropped).
+pub fn resolve_function_call(
+    call: &CallInput,
+    caller_file_path: &str,
+    local_names: &HashSet<String>,
+    local_imports: &HashMap<String, String>,
+    imports_map: &HashMap<String, Vec<String>>,
+    types: &TypeIndex,
+    skip_external: bool,
+) -> Option<ResolvedCall> {
+    resolve_call(call, caller_file_path, local_names, local_imports, imports_map, types, skip_external).ok()
+}
+
 /// Language extension map for filtering imports_map by language.
 fn lang_extensions(lang: &str) -> Option<&[&str]> {
     match lang {
@@ -763,7 +793,7 @@ pub fn build_function_call_groups(
         };
 
         for call in &file_data.calls {
-            let first = match resolve_function_call(
+            let first = match resolve_call(
                 call,
                 caller_file_path,
                 &local_names_owned,
@@ -772,8 +802,20 @@ pub fn build_function_call_groups(
                 &types,
                 skip_external,
             ) {
-                Some(r) => r,
-                None => continue,
+                Ok(r) => r,
+                Err("python_builtin") => continue,
+                Err(reason) => {
+                    groups.unresolved.push(UnresolvedCall {
+                        caller_file_path: caller_file_path.to_string(),
+                        caller_name: call.context_name.clone(),
+                        line_number: call.line_number,
+                        called_name: call.name.clone(),
+                        full_call_name: call.full_name.clone(),
+                        receiver_type: call.inferred_obj_type.clone(),
+                        reason,
+                    });
+                    continue;
+                }
             };
 
             // Virtual dispatch (upstream: Java interface fan-out): a call
@@ -1313,4 +1355,18 @@ mod tests {
         let r = resolve_function_call(&call, "/r/Log.ts", &HashSet::new(), &HashMap::new(), &HashMap::new(), &TypeIndex::default(), false);
         assert!(r.is_none());
     }
+
+    #[test]
+    fn test_unresolved_calls_are_recorded_with_reason() {
+        let mut f = file("/r/A.java", &[("A", &["run", "add"], &[])], vec![]);
+        let mut ext = receiver_call("add", "items.add", Some("List"));
+        ext.context_name = Some("run".into());
+        let mut builtin = receiver_call("len", "len", None);
+        builtin.context_name = Some("run".into());
+        f.calls = vec![ext, builtin];
+        let groups = build_function_call_groups(&[f], &HashMap::new(), &HashMap::new(), false);
+        assert_eq!(groups.unresolved.len(), 1); // the builtin isn't recorded (and A.java isn't Python anyway)
+        let u = &groups.unresolved[0];
+        assert_eq!((u.called_name.as_str(), u.reason, u.receiver_type.as_deref()), ("add", "external_receiver_type", Some("List")));
+    }
 }

{codegraphcontext_rust-0.9.19 → codegraphcontext_rust-0.9.21}/rust/cgc-python/src/lib.rs

@@ -88,12 +88,13 @@ fn parse_and_prescan(
 
 /// Convert Python all_file_data + imports_map into Rust types, run resolution, return results.
 #[pyfunction]
-#[pyo3(signature = (all_file_data, imports_map, skip_external=false))]
+#[pyo3(signature = (all_file_data, imports_map, skip_external=false, with_unresolved=false))]
 fn resolve_call_groups(
     py: Python<'_>,
     all_file_data: &Bound<'_, PyList>,
     imports_map: &Bound<'_, PyDict>,
     skip_external: bool,
+    with_unresolved: bool,
 ) -> PyResult<PyObject> {
     // Convert imports_map: Python dict -> Rust HashMap
     let rust_imports_map = py_dict_to_imports_map(imports_map)?;
@@ -120,8 +121,25 @@ fn resolve_call_groups(
         build_function_call_groups(&rust_files, &rust_imports_map, &file_class_lookup, skip_external)
     });
 
-    // Convert back to Python: 6-tuple of lists of dicts
-    call_groups_to_py(py, &groups)
+    // Convert back to Python: 6-tuple of lists of dicts; with_unresolved
+    // returns (6-tuple, [unresolved call dicts]) instead.
+    let call_groups = call_groups_to_py(py, &groups)?;
+    if !with_unresolved {
+        return Ok(call_groups);
+    }
+    let unresolved = PyList::empty(py);
+    for u in &groups.unresolved {
+        let d = PyDict::new(py);
+        d.set_item("caller_file_path", &u.caller_file_path)?;
+        d.set_item("caller_name", &u.caller_name)?;
+        d.set_item("line_number", u.line_number)?;
+        d.set_item("called_name", &u.called_name)?;
+        d.set_item("full_call_name", &u.full_call_name)?;
+        d.set_item("receiver_type", &u.receiver_type)?;
+        d.set_item("reason", u.reason)?;
+        unresolved.append(d)?;
+    }
+    Ok(PyTuple::new(py, [call_groups, unresolved.into_any().unbind()])?.into_any().unbind())
 }
 
 /// Resolve inheritance links.

codegraphcontext_rust-0.9.21/src/codegraphcontext/tools/indexing/angular_routes.py

@@ -0,0 +1,396 @@
+"""Angular Router: route tables, lazy-loaded modules and navigation calls.
+
+Routes come from `Routes` arrays passed to `RouterModule.forRoot/forChild`
+or `provideRouter`, or exported for `loadChildren`. Full paths are built by
+following `loadChildren: () => import('./x/x.module').then(m => m.XModule)`
+from the root routes into the child module's routing file, so a page gets
+`/home/tabs/tab1` rather than just `tab1`.
+
+Navigations are `router.navigate([...])` / `router.navigateByUrl('...')`
+calls, attributed to their enclosing function and matched to a route.
+"""
+
+from __future__ import annotations
+
+import os
+import re
+from typing import Any, Dict, List, Optional, Tuple
+
+_ROUTE_ARRAY_DECL = re.compile(
+    r"\b(?:export\s+)?(?:const|let|var)\s+(\w+)\s*(?::\s*(Routes|Route\s*\[\s*\]))?\s*=\s*\["
+)
+_ROUTER_CALL = re.compile(r"\b(?:RouterModule\.(forRoot|forChild)|(provideRouter))\s*\(")
+_IMPORT_SPEC = re.compile(r"import\(\s*['\"]([^'\"]+)['\"]\s*\)")
+_THEN_SYMBOL = re.compile(r"\.then\(\s*\(?\s*(\w+)\s*\)?\s*=>\s*\1\.(\w+)")
+_NAVIGATE = re.compile(r"\b(?:\w+\.)*\w*[Rr]outer\s*\.\s*(navigate|navigateByUrl)\s*\(")
+_TS_EXTS = (".ts", ".tsx", ".js", ".mjs")
+
+
+# ── Small JS literal parser ──────────────────────────────────────────
+
+def _strip_comments(src: str) -> str:
+    """Blank out // and /* */ comments (string-aware, keeps offsets)."""
+    out, i, n = [], 0, len(src)
+    while i < n:
+        c = src[i]
+        if c in "'\"`":
+            j = i + 1
+            while j < n and src[j] != c:
+                j += 2 if src[j] == "\\" else 1
+            out.append(src[i:j + 1]); i = j + 1
+        elif src.startswith("//", i):
+            j = src.find("\n", i); j = n if j == -1 else j
+            out.append(" " * (j - i)); i = j
+        elif src.startswith("/*", i):
+            j = src.find("*/", i + 2); j = n if j == -1 else j + 2
+            out.append(re.sub(r"[^\n]", " ", src[i:j])); i = j
+        else:
+            out.append(c); i += 1
+    return "".join(out)
+
+
+def _match_close(code: str, i: int) -> int:
+    """Index just past the bracket matching code[i] ('[', '{' or '(')."""
+    depth, n = 0, len(code)
+    while i < n:
+        c = code[i]
+        if c in "'\"`":
+            j = i + 1
+            while j < n and code[j] != c:
+                j += 2 if code[j] == "\\" else 1
+            i = j + 1
+            continue
+        if c in "[{(":
+            depth += 1
+        elif c in "]})":
+            depth -= 1
+            if depth == 0:
+                return i + 1
+        i += 1
+    return n
+
+
+def _split_top(code: str, start: int, end: int) -> List[Tuple[int, int]]:
+    """Comma-separated top-level segments of code[start:end]."""
+    segs, depth, seg_start, i = [], 0, start, start
+    while i < end:
+        c = code[i]
+        if c in "'\"`":
+            j = i + 1
+            while j < end and code[j] != c:
+                j += 2 if code[j] == "\\" else 1
+            i = j + 1
+            continue
+        if c in "[{(":
+            depth += 1
+        elif c in "]})":
+            depth -= 1
+        elif c == "," and depth == 0:
+            segs.append((seg_start, i)); seg_start = i + 1
+        i += 1
+    segs.append((seg_start, end))
+    return [(a, b) for a, b in segs if code[a:b].strip()]
+
+
+def _string_value(text: str) -> Optional[str]:
+    m = re.fullmatch(r"\s*(['\"`])(.*?)\1\s*", text, re.S)
+    return m.group(2) if m else None
+
+
+def _parse_object(code: str, start: int, end: int) -> Dict[str, Tuple[str, int]]:
+    """Top-level `key: value` pairs of an object literal code[start:end] ({...})."""
+    props = {}
+    for a, b in _split_top(code, start + 1, end - 1):
+        m = re.match(r"\s*['\"]?(\w+)['\"]?\s*:", code[a:b])
+        if m:
+            vstart = a + m.end()
+            props[m.group(1)] = (code[vstart:b].strip(), vstart + (len(code[vstart:b]) - len(code[vstart:b].lstrip())))
+    return props
+
+
+def _parse_routes(code: str, start: int, end: int) -> List[Dict[str, Any]]:
+    """Parse a routes array code[start:end] ([...]) into route dicts."""
+    routes = []
+    for a, b in _split_top(code, start + 1, end - 1):
+        seg = code[a:b].strip()
+        off = a + (len(code[a:b]) - len(code[a:b].lstrip()))
+        if seg.startswith("..."):  # spread of another routes array
+            routes.append({"ref": seg[3:].strip(), "pos": off})
+            continue
+        if re.fullmatch(r"\w+", seg):
+            routes.append({"ref": seg, "pos": off})
+            continue
+        if not seg.startswith("{"):
+            continue
+        props = _parse_object(code, off, _match_close(code, off))
+        route: Dict[str, Any] = {"pos": off, "path": None}
+        if "path" in props:
+            route["path"] = _string_value(props["path"][0])
+        if "component" in props:
+            route["component"] = props["component"][0].strip()
+        if "redirectTo" in props:
+            route["redirect"] = _string_value(props["redirectTo"][0]) or props["redirectTo"][0]
+        for key in ("loadChildren", "loadComponent"):
+            if key in props:
+                val = props[key][0]
+                spec = _IMPORT_SPEC.search(val)
+                sym = _THEN_SYMBOL.search(val)
+                legacy = _string_value(val)  # 'path/to/x.module#XModule'
+                if spec:
+                    route[key] = (spec.group(1), sym.group(2) if sym else None)
+                elif legacy and "#" in legacy:
+                    p, s = legacy.split("#", 1)
+                    route[key] = (p, s)
+        if "children" in props and props["children"][0].startswith("["):
+            cstart = props["children"][1]
+            route["children"] = _parse_routes(code, cstart, _match_close(code, cstart))
+        routes.append(route)
+    return routes
+
+
+# ── Per-file route tables ────────────────────────────────────────────
+
+def _line_of(code: str, pos: int) -> int:
+    return code.count("\n", 0, pos) + 1
+
+
+def _file_routes(code: str) -> Dict[str, Any]:
+    """Named route arrays and router registrations found in one file."""
+    arrays: Dict[str, Dict[str, Any]] = {}
+    for m in _ROUTE_ARRAY_DECL.finditer(code):
+        start = m.end() - 1
+        arrays[m.group(1)] = {
+            "routes": _parse_routes(code, start, _match_close(code, start)),
+            "typed": bool(m.group(2)),
+        }
+    registrations = []  # (mode, array name or None, inline routes or None)
+    for m in _ROUTER_CALL.finditer(code):
+        mode = "root" if (m.group(1) == "forRoot" or m.group(2)) else "child"
+        args_start = m.end()
+        while args_start < len(code) and code[args_start].isspace():
+            args_start += 1
+        if args_start < len(code) and code[args_start] == "[":
+            registrations.append((mode, None, _parse_routes(code, args_start, _match_close(code, args_start))))
+        else:
+            ident = re.match(r"(\w+)", code[args_start:])
+            if ident:
+                registrations.append((mode, ident.group(1), None))
+    return {"arrays": arrays, "registrations": registrations}
+
+
+def _resolve_import(from_rel: str, spec: str, known: set) -> Optional[str]:
+    if not spec.startswith("."):
+        return None
+    base = os.path.normpath(os.path.join(os.path.dirname(from_rel), spec))
+    for cand in [base] + [base + e for e in _TS_EXTS] + [os.path.join(base, "index" + e) for e in _TS_EXTS]:
+        if cand in known:
+            return cand
+    return None
+
+
+def _join(prefix: str, path: Optional[str]) -> str:
+    # `path: \`${PREFIX}/x\`` → dynamic segment
+    path = re.sub(r"\$\{[^}]*\}", ":param", path or "")
+    parts = [p for p in (prefix.strip("/"), path.strip("/")) if p]
+    return "/" + "/".join(parts)
+
+
+def extract_angular_routes(
+    parsed_results: List[Dict[str, Any]],
+    repo_path: str,
+) -> List[Dict[str, Any]]:
+    """Angular routes with full paths, as route dicts (method PAGE/REDIRECT)."""
+    files: Dict[str, Dict[str, Any]] = {}
+    codes: Dict[str, str] = {}
+    for fd in parsed_results:
+        path = fd.get("path", "")
+        if fd.get("lang") not in ("typescript", "tsx", "javascript") or not os.path.exists(path):
+            continue
+        try:
+            src = open(path, encoding="utf-8", errors="replace").read()
+        except OSError:
+            continue
+        if "Route" not in src and "provideRouter" not in src:
+            continue
+        rel = os.path.relpath(path, repo_path)
+        code = _strip_comments(src)
+        info = _file_routes(code)
+        if info["arrays"] or info["registrations"]:
+            files[rel] = info
+            codes[rel] = code
+    if not files:
+        return []
+    known = {os.path.relpath(fd.get("path", ""), repo_path) for fd in parsed_results}
+
+    def array_by_name(rel: str, name: str):
+        arr = files.get(rel, {}).get("arrays", {}).get(name)
+        return arr["routes"] if arr else None
+
+    def child_tables(target: str, symbol: Optional[str]) -> List[Tuple[str, List]]:
+        """Route tables a lazy `loadChildren` target contributes."""
+        out = []
+        if target in files:
+            if symbol and array_by_name(target, symbol) is not None:
+                return [(target, array_by_name(target, symbol))]
+            out += [(target, routes) for _, routes in _registered(target)]
+        if not out:  # the module imports a sibling *-routing.module
+            d = os.path.dirname(target)
+            for rel in files:
+                if os.path.dirname(rel) == d and rel != target and ("routing" in rel or "routes" in rel):
+                    out += [(rel, routes) for mode, routes in _registered(rel) if mode == "child"]
+        return out
+
+    def _registered(rel: str) -> List[Tuple[str, List]]:
+        info = files[rel]
+        out = []
+        for mode, name, inline in info["registrations"]:
+            routes = inline if inline is not None else array_by_name(rel, name) if name else None
+            if routes is not None:
+                out.append((mode, routes))
+        return out
+
+    results: List[Dict[str, Any]] = []
+    emitted = set()
+    reached = set()  # ids of route lists visited from a root
+
+    def walk(rel: str, routes: List, prefix: str, depth: int):
+        if depth > 12:
+            return
+        reached.add(id(routes))
+        for r in routes:
+            if "ref" in r:  # ...OTHER_ROUTES / OTHER_ROUTES inside the array
+                ref = array_by_name(rel, r["ref"])
+                if ref is not None and id(ref) not in reached:
+                    walk(rel, ref, prefix, depth + 1)
+                continue
+            full = _join(prefix, r.get("path"))
+            line = _line_of(codes[rel], r["pos"])
+            handler, method = None, "PAGE"
+            if r.get("component"):
+                handler = r["component"]
+            elif r.get("loadComponent"):
+                handler = r["loadComponent"][1] or r["loadComponent"][0]
+            elif r.get("loadChildren"):
+                handler = r["loadChildren"][1] or r["loadChildren"][0]
+            elif r.get("redirect") is not None:
+                to = str(r["redirect"])
+                handler = "→ " + (to if to.startswith("/") else _join(prefix, to))
+                method = "REDIRECT"
+            if handler and (method, full, rel, line) not in emitted:
+                emitted.add((method, full, rel, line))
+                results.append({"method": method, "path": full, "handler": handler,
+                                "file": rel, "line": line, "framework": "angular"})
+            if r.get("children"):
+                walk(rel, r["children"], full, depth + 1)
+            if r.get("loadChildren"):
+                spec, symbol = r["loadChildren"]
+                target = _resolve_import(rel, spec, known)
+                if target:
+                    for trel, troutes in child_tables(target, symbol):
+                        if id(troutes) not in reached:
+                            walk(trel, troutes, full, depth + 1)
+
+    for rel in files:
+        for mode, routes in _registered(rel):
+            if mode == "root":
+                walk(rel, routes, "", 0)
+    # Child tables never reached from a root (unknown mount point): keep
+    # their local paths so the pages are still listed.
+    for rel in files:
+        for mode, routes in _registered(rel):
+            if id(routes) not in reached:
+                walk(rel, routes, "", 0)
+    return results
+
+
+# ── Navigations ──────────────────────────────────────────────────────
+
+def _nav_target(code: str, args_start: int, kind: str) -> Optional[str]:
+    """Target path of a navigate([...]) / navigateByUrl('...') call.
+
+    Non-literal segments become `:param`; a path is absolute when its first
+    literal segment starts with '/'.
+    """
+    while args_start < len(code) and code[args_start].isspace():
+        args_start += 1
+    if kind == "navigate":
+        if args_start >= len(code) or code[args_start] != "[":
+            return None
+        end = _match_close(code, args_start)
+        parts, absolute = [], False
+        for idx, (a, b) in enumerate(_split_top(code, args_start + 1, end - 1)):
+            s = _string_value(code[a:b])
+            if s is None:
+                parts.append(":param")
+                continue
+            if idx == 0 and s.startswith("/"):
+                absolute = True
+            parts += [":param" if "${" in p else p for p in s.split("/") if p]
+        return ("/" if absolute else "") + "/".join(parts)
+    m = re.match(r"(['\"`])(.*?)\1", code[args_start:args_start + 500], re.S)
+    if not m:
+        return None
+    return re.sub(r"\$\{[^}]*\}", ":param", m.group(2).split("?")[0])
+
+
+def _match_route(target: str, routes: List[Dict[str, Any]]) -> Optional[Dict[str, Any]]:
+    """Best route for an absolute target path (literal segments must match)."""
+    if not target.startswith("/"):
+        return None
+    tseg = [s for s in target.strip("/").split("/") if s]
+    best, best_score = None, -1
+    for r in routes:
+        if r["method"] != "PAGE":
+            continue
+        rseg = [s for s in r["path"].strip("/").split("/") if s]
+        if len(rseg) != len(tseg):
+            continue
+        score = 0
+        for a, b in zip(rseg, tseg):
+            if a.startswith(":") or b == ":param":
+                continue
+            if a != b:
+                break
+            score += 1
+        else:
+            if score > best_score:
+                best, best_score = r, score
+    return best
+
+
+def extract_navigations(
+    parsed_results: List[Dict[str, Any]],
+    repo_path: str,
+    angular_routes: List[Dict[str, Any]],
+) -> List[Dict[str, Any]]:
+    """router.navigate / navigateByUrl calls with their enclosing function and route."""
+    out = []
+    for fd in parsed_results:
+        path = fd.get("path", "")
+        if fd.get("lang") not in ("typescript", "tsx", "javascript") or not os.path.exists(path):
+            continue
+        try:
+            src = open(path, encoding="utf-8", errors="replace").read()
+        except OSError:
+            continue
+        if "navigate" not in src:
+            continue
+        code = _strip_comments(src)
+        rel = os.path.relpath(path, repo_path)
+        funcs = [(f.get("line_number", 0), f.get("end_line", 0) or f.get("line_number", 0), f.get("name", ""))
+                 for f in fd.get("functions", [])]
+        for m in _NAVIGATE.finditer(code):
+            target = _nav_target(code, m.end(), m.group(1))
+            if target is None:
+                continue
+            line = _line_of(code, m.start())
+            enclosing = [f for f in funcs if f[0] <= line <= f[1]]
+            caller = max(enclosing, key=lambda f: f[0])[2] if enclosing else ""
+            route = _match_route(target, angular_routes)
+            out.append({
+                "caller_name": caller, "caller_path": rel, "line": line, "target": target,
+                "route_path": route["path"] if route else "",
+                "route_handler": route["handler"] if route else "",
+                "route_file": route["file"] if route else "",
+            })
+    return out

{codegraphcontext_rust-0.9.19 → codegraphcontext_rust-0.9.21}/src/codegraphcontext/tools/indexing/cgc_wiki_cli.py

@@ -75,7 +75,7 @@ def cmd_index(args):
     specs = [(f, l, False) for f, l in files]
     results, imports_map = parse_and_prescan(specs)
     valid = [r for r in results if "error" not in r]
-    call_groups = resolve_call_groups(valid, imports_map, False)
+    call_groups, unresolved = resolve_call_groups(valid, imports_map, False, True)
     inheritance, _ = resolve_inheritance(valid, imports_map)
     t_parse = time.time() - t0
     print(f"\nParse + resolve: {t_parse:.1f}s")
@@ -86,7 +86,7 @@ def cmd_index(args):
     db_path = os.path.join(output_dir, "graph.duckdb")
     t0 = time.time()
     writer = DuckDBGraphWriter(db_path)
-    counts = writer.write_all(valid, repo_path, call_groups, inheritance)
+    counts = writer.write_all(valid, repo_path, call_groups, inheritance, unresolved=unresolved)
     t_write = time.time() - t0
 
     # Read back from DB (no duplicate extraction)

{codegraphcontext_rust-0.9.19 → codegraphcontext_rust-0.9.21}/src/codegraphcontext/tools/indexing/engine.py

@@ -81,11 +81,15 @@ def pre_scan_for_imports(file_specs):
     return _rust_pre_scan(file_specs)
 
 
-def resolve_call_groups(all_file_data, imports_map, skip_external=False):
-    """Resolve function calls into 6-category groups using Rust."""
+def resolve_call_groups(all_file_data, imports_map, skip_external=False, with_unresolved=False):
+    """Resolve function calls into 6-category groups using Rust.
+
+    With ``with_unresolved=True`` returns ``(groups, unresolved)`` where
+    ``unresolved`` lists the calls that produced no edge and why.
+    """
     if not RUST_AVAILABLE:
         raise RuntimeError("Rust engine required for resolve_call_groups")
-    return _rust_resolve_calls(all_file_data, imports_map, skip_external)
+    return _rust_resolve_calls(all_file_data, imports_map, skip_external, with_unresolved)
 
 
 def resolve_inheritance(all_file_data, imports_map):