feat: add ArkTS (.ets) language support with tree-sitter parser

internal/cbm/cbm.h

@@ -170,6 +170,7 @@ typedef enum {
     CBM_LANG_QML,       // Qt QML (Qt Modeling Language — declarative UI + embedded JS)
     CBM_LANG_CFSCRIPT,  // CFML script dialect (.cfc components — Lucee/ColdFusion)
     CBM_LANG_CFML,      // CFML tag dialect (.cfm templates — Lucee/ColdFusion)
+    CBM_LANG_ARKTS,     // ArkTS (HarmonyOS/OpenHarmony declarative UI language)
     CBM_LANG_COUNT
 } CBMLanguage;
 

internal/cbm/extract_defs.c

@@ -697,6 +697,30 @@ TSNode cbm_resolve_func_name(TSNode node, CBMLanguage lang) {
             }
         }
 
+        /* ArkTS: function_declaration / decorated_function_declaration
+         * have no `name` field; the function name is a plain `identifier` child.
+         * build_method has no name at all — synthesize one from the keyword. */
+        if (lang == CBM_LANG_ARKTS && (strcmp(kind, "function_declaration") == 0 ||
+                                       strcmp(kind, "decorated_function_declaration") == 0)) {
+            TSNode id = cbm_find_child_by_kind(node, "identifier");
+            if (!ts_node_is_null(id)) {
+                return id;
+            }
+        }
+        if (lang == CBM_LANG_ARKTS && strcmp(kind, "build_method") == 0) {
+            /* build() is always named "build" in ArkTS components. Return a
+             * synthetic name by finding the "build" keyword child. If no named
+             * child matches, fall through — extract_func_def will skip it. */
+            TSNode kw = cbm_find_child_by_kind(node, "build");
+            if (!ts_node_is_null(kw)) {
+                return kw;
+            }
+            /* tree-sitter-arkts may use "build" as an anonymous token.
+             * Try the first named child which could be the body. If no
+             * identifier exists, we still want to emit a def — handled in
+             * extract_func_def via a special case. */
+        }
+
         // PowerShell function_statement has no `name` field; the name is a
         // `function_name` child node (#35).
         if (lang == CBM_LANG_POWERSHELL && strcmp(kind, "function_statement") == 0) {
@@ -2607,7 +2631,8 @@ static const char *class_label_for_kind(const char *kind) {
         return "Enum";
     }
     if (strcmp(kind, "type_alias_declaration") == 0 || strcmp(kind, "type_item") == 0 ||
-        strcmp(kind, "type_alias") == 0 || strcmp(kind, "type_definition") == 0) {
+        strcmp(kind, "type_alias") == 0 || strcmp(kind, "type_definition") == 0 ||
+        strcmp(kind, "type_declaration") == 0) {
         return "Type";
     }
     return "Class";
@@ -3049,13 +3074,21 @@ static char *go_receiver_type_name(CBMArena *a, TSNode recv, const char *source)
 
 static void extract_func_def(CBMExtractCtx *ctx, TSNode node, const CBMLangSpec *spec) {
     CBMArena *a = ctx->arena;
+    const char *kind = ts_node_type(node);
 
     TSNode name_node = cbm_resolve_func_name(node, ctx->language);
-    if (ts_node_is_null(name_node)) {
+
+    /* ArkTS: build_method has no name child — synthesize "build". */
+    char *synth_name = NULL;
+    if (ts_node_is_null(name_node) && ctx->language == CBM_LANG_ARKTS &&
+        strcmp(kind, "build_method") == 0) {
+        synth_name = cbm_arena_strdup(a, "build");
+    }
+    if (ts_node_is_null(name_node) && !synth_name) {
         return;
     }
 
-    char *name = cbm_func_name_node_text(a, name_node, ctx->source);
+    char *name = synth_name ? synth_name : cbm_func_name_node_text(a, name_node, ctx->source);
     if (!name || !name[0] || strcmp(name, "function") == 0) {
         return;
     }
@@ -3482,6 +3515,12 @@ static void extract_class_def(CBMExtractCtx *ctx, TSNode node, const CBMLangSpec
         (ctx->language == CBM_LANG_SWIFT || ctx->language == CBM_LANG_KOTLIN)) {
         name_node = cbm_find_child_by_kind(node, "type_identifier");
     }
+    // ArkTS: class_declaration / component_declaration / interface_declaration /
+    // enum_declaration / type_declaration have no `name` field; the name is a
+    // plain `identifier` child.
+    if (ts_node_is_null(name_node) && ctx->language == CBM_LANG_ARKTS) {
+        name_node = cbm_find_child_by_kind(node, "identifier");
+    }
     // Protobuf: service_name / message_name / enum_name children
     if (ts_node_is_null(name_node) && ctx->language == CBM_LANG_PROTOBUF) {
         name_node = cbm_find_child_by_kind(node, "service_name");
@@ -3744,6 +3783,16 @@ static void extract_class_def(CBMExtractCtx *ctx, TSNode node, const CBMLangSpec
             label = "Interface";
         }
     }
+    // ArkTS: `@Component struct Foo` parses to `component_declaration` (and
+    // `@Component export struct Foo` to `decorated_export_declaration`). These
+    // are UI components, not OOP classes — emit the precise "Component" label.
+    // Scoped to ArkTS only so templ's `component_declaration` (a Go templating
+    // function) keeps its established "Class" labeling and the golden snapshot
+    // for templ is unaffected.
+    if (ctx->language == CBM_LANG_ARKTS && (strcmp(kind, "component_declaration") == 0 ||
+                                            strcmp(kind, "decorated_export_declaration") == 0)) {
+        label = "Component";
+    }
     // Rust/Swift/D: a struct is a distinct kind from a class — emit the precise
     // "Struct" label rather than collapsing it to "Class". Scoped to these three
     // grammar/LSP languages. Rust's struct node is `struct_item`; D's is
@@ -3941,6 +3990,7 @@ static TSNode find_class_body(TSNode class_node, CBMLanguage lang) {
                                        "block",
                                        "closure",
                                        "implementation_definition",
+                                       "component_body",
                                        NULL};
     uint32_t count = ts_node_child_count(class_node);
     for (uint32_t i = 0; i < count; i++) {
@@ -4046,6 +4096,24 @@ static TSNode resolve_method_name(TSNode child, CBMLanguage lang) {
         return cbm_find_child_by_kind(child, "identifier");
     }
 
+    // ArkTS: method / build_method / decorated_function_declaration have no `name`
+    // field; the method name is a plain `identifier` child. For build_method,
+    // the name is always "build" (synthesized by the caller).
+    if (lang == CBM_LANG_ARKTS &&
+        (strcmp(ck, "method_declaration") == 0 || strcmp(ck, "function_declaration") == 0 ||
+         strcmp(ck, "decorated_function_declaration") == 0)) {
+        TSNode id = cbm_find_child_by_kind(child, "identifier");
+        if (!ts_node_is_null(id)) {
+            return id;
+        }
+    }
+    if (lang == CBM_LANG_ARKTS && strcmp(ck, "build_method") == 0) {
+        TSNode kw = cbm_find_child_by_kind(child, "build");
+        if (!ts_node_is_null(kw)) {
+            return kw;
+        }
+    }
+
     if (strcmp(ck, "arrow_function") == 0) {
         return resolve_arrow_func_name(child);
     }
@@ -4204,6 +4272,26 @@ static void extract_class_methods(CBMExtractCtx *ctx, TSNode class_node, const c
         }
 
         TSNode name_node = resolve_method_name(method_node, ctx->language);
+        /* ArkTS: build_method has no name child — synthesize "build" as method name. */
+        if (ts_node_is_null(name_node) && ctx->language == CBM_LANG_ARKTS &&
+            strcmp(ts_node_type(method_node), "build_method") == 0) {
+            /* Use the method_node itself as name_node; push_method_def will
+             * get the text "build_method" — override below. */
+            char *build_name = cbm_arena_strdup(ctx->arena, "build");
+            const char *build_qn = cbm_arena_sprintf(ctx->arena, "%s.build", class_qn);
+            CBMDefinition mdef;
+            memset(&mdef, 0, sizeof(mdef));
+            mdef.name = build_name;
+            mdef.qualified_name = build_qn;
+            mdef.label = "Method";
+            mdef.file_path = ctx->rel_path;
+            mdef.start_line = ts_node_start_point(method_node).row + TS_LINE_OFFSET;
+            mdef.end_line = ts_node_end_point(method_node).row + TS_LINE_OFFSET;
+            mdef.lines = (int)(mdef.end_line - mdef.start_line + TS_LINE_OFFSET);
+            mdef.parent_class = class_qn;
+            cbm_defs_push(&ctx->result->defs, ctx->arena, mdef);
+            continue;
+        }
         if (ts_node_is_null(name_node)) {
             continue;
         }
@@ -5727,6 +5815,9 @@ static const char *compute_class_qn(CBMExtractCtx *ctx, TSNode node, const char
     if (ts_node_is_null(name_node) && ctx->language == CBM_LANG_SWIFT) {
         name_node = cbm_find_child_by_kind(node, "type_identifier");
     }
+    if (ts_node_is_null(name_node) && ctx->language == CBM_LANG_ARKTS) {
+        name_node = cbm_find_child_by_kind(node, "identifier");
+    }
     if (!ts_node_is_null(name_node)) {
         char *cname = cbm_node_text(ctx->arena, name_node, ctx->source);
         if (cname && cname[0]) {
@@ -5752,6 +5843,7 @@ static void push_class_body_children(TSNode node, const CBMLangSpec *spec, wd_st
         if (strcmp(ck, "field_declaration_list") == 0 || strcmp(ck, "class_body") == 0 ||
             strcmp(ck, "declaration_list") == 0 || strcmp(ck, "body") == 0 ||
             strcmp(ck, "block") == 0 || strcmp(ck, "suite") == 0 ||
+            strcmp(ck, "component_body") == 0 ||
             // Groovy class bodies are a `closure` node; routing through the
             // nested-class path keeps methods from being re-walked (and thus
             // double-extracted) as top-level functions. Gated to Groovy so other
@@ -6253,7 +6345,7 @@ static void walk_defs(CBMExtractCtx *ctx, TSNode root, const CBMLangSpec *spec,
                 bool descend_into_func =
                     (ctx->language == CBM_LANG_WOLFRAM || ctx->language == CBM_LANG_TYPESCRIPT ||
                      ctx->language == CBM_LANG_JAVASCRIPT || ctx->language == CBM_LANG_TSX ||
-                     ctx->language == CBM_LANG_ADA);
+                     ctx->language == CBM_LANG_ADA || ctx->language == CBM_LANG_ARKTS);
                 if (!descend_into_func) {
                     continue;
                 }

internal/cbm/grammar_arkts.c

@@ -0,0 +1,3 @@
+// Vendored tree-sitter grammar: arkts
+// Each grammar compiled as separate unit (conflicting static symbols).
+#include "vendored/grammars/arkts/parser.c"

internal/cbm/lang_specs.c

@@ -164,6 +164,7 @@ extern const TSLanguage *tree_sitter_apex(void);
 extern const TSLanguage *tree_sitter_soql(void);
 extern const TSLanguage *tree_sitter_sosl(void);
 extern const TSLanguage *tree_sitter_pine(void);
+extern const TSLanguage *tree_sitter_arkts(void);
 
 // -- Empty sentinel --
 static const char *empty_types[] = {NULL};
@@ -1596,6 +1597,43 @@ static const char *pine_var_types[] = {"variable_definition_statement",
 static const char *pine_branch_types[] = {"if_statement",     "switch_statement", "for_statement",
                                           "for_in_statement", "while_statement",  NULL};
 static const char *pine_assign_types[] = {"reassignment_statement", NULL};
+// ==================== ARKTS (HarmonyOS/OpenHarmony) ====================
+// ArkTS is a TypeScript superset with declarative UI extensions.
+// Grammar: tree-sitter-arkts (https://github.com/Million-mo/tree-sitter-arkts)
+// Reuses TS/JS node types where applicable, adds ArkTS-specific UI nodes.
+static const char *arkts_func_types[] = {"function_declaration",
+                                         "function_expression",
+                                         "arrow_function",
+                                         "method_declaration",
+                                         "constructor_declaration",
+                                         "build_method",
+                                         "decorated_function_declaration",
+                                         "ui_builder_arrow_function",
+                                         NULL};
+static const char *arkts_class_types[] = {"class_declaration",
+                                          "enum_declaration",
+                                          "interface_declaration",
+                                          "type_declaration",
+                                          "component_declaration",
+                                          "decorated_export_declaration",
+                                          NULL};
+static const char *arkts_field_types[] = {"property_declaration", "public_field_definition", NULL};
+static const char *arkts_module_types[] = {"source_file", NULL};
+static const char *arkts_call_types[] = {"call_expression", "new_expression", NULL};
+static const char *arkts_import_types[] = {"import_declaration", "import", NULL};
+static const char *arkts_branch_types[] = {"if_statement",       "for_statement",
+                                           "for_in_statement",   "while_statement",
+                                           "do_statement",       "switch_statement",
+                                           "switch_case",        "switch_default",
+                                           "try_statement",      "catch_clause",
+                                           "for_each_statement", "lazy_for_each_statement",
+                                           "ui_if_statement",    NULL};
+static const char *arkts_var_types[] = {"lexical_declaration", "variable_declaration", NULL};
+static const char *arkts_assign_types[] = {"assignment_expression",
+                                           "augmented_assignment_expression", NULL};
+static const char *arkts_throw_types[] = {"throw_statement", NULL};
+static const char *arkts_decorator_types[] = {"decorator", NULL};
+
 // ==================== SPEC TABLE ====================
 
 static const CBMLangSpec lang_specs[CBM_LANG_COUNT] = {
@@ -2571,6 +2609,14 @@ static const CBMLangSpec lang_specs[CBM_LANG_COUNT] = {
                        pine_branch_types, pine_var_types, pine_assign_types, empty_types, NULL,
                        empty_types, NULL, NULL, tree_sitter_pine, NULL},
 
+    // CBM_LANG_ARKTS — ArkTS (HarmonyOS/OpenHarmony declarative UI language).
+    // TypeScript superset with UI extensions. Uses tree-sitter-arkts grammar.
+    [CBM_LANG_ARKTS] = {CBM_LANG_ARKTS, arkts_func_types, arkts_class_types, arkts_field_types,
+                        arkts_module_types, arkts_call_types, arkts_import_types,
+                        arkts_import_types, arkts_branch_types, arkts_var_types, arkts_assign_types,
+                        arkts_throw_types, NULL, arkts_decorator_types, NULL, NULL,
+                        tree_sitter_arkts, NULL},
+
 };
 
 _Static_assert(sizeof(lang_specs) / sizeof(lang_specs[0]) == CBM_LANG_COUNT,

internal/cbm/vendored/grammars/MANIFEST.md

@@ -9,12 +9,13 @@ The grammars were originally vendored as bare `parser.c`+`scanner.c` with **no r
 
 ## Summary
 
-- Grammars: **159** — vendored-from-upstream: **142**, first-party/self-maintained: **12**, registry-disagreement: **5** (nim removed 2026-06-12; objectscript_udl + objectscript_routine added 2026-06-24; mojo added 2026-07-01 — see notes below)
+- Grammars: **160** — vendored-from-upstream: **143**, first-party/self-maintained: **12**, registry-disagreement: **5** (nim removed 2026-06-12; objectscript_udl + objectscript_routine added 2026-06-24; mojo added 2026-07-01; arkts added 2026-07-03 — see notes below)
 - ABI distribution: **7×** ABI-13 **85×** ABI-14 **64×** ABI-15 (runtime ceiling is ABI 15; never vendor ABI 16 without a runtime upgrade)
 - Vendored copies missing LICENSE: **0** — all upstream LICENSE files restored 2026-06-11 (first-party grammars carry the project MIT license; `move` uses the Helix-listed upstream tzakian/tree-sitter-move MIT text, `zsh` uses georgeharker/tree-sitter-zsh MIT)
 - `verdict`: VERIFIED-BOTH = our source matches *both* registries; VERIFIED-NVIM/HELIX = matches one; registry-disagreement = registries name a different repo (listed separately); `vendor-maintained` = the language vendor's own grammar, not in nvim/Helix.
 - **objectscript_udl / objectscript_routine** (added 2026-06-24): vendored from [intersystems/tree-sitter-objectscript](https://github.com/intersystems/tree-sitter-objectscript) @ `a7ffcdf` — MIT, the InterSystems-official grammars (a niche vendor language, hence `vendor-maintained`, not in nvim-treesitter/Helix). **Re-vendor note:** each `scanner.c`'s upstream `#include "../../common/scanner.h"` is repointed to a per-directory `objectscript_common.h` (a verbatim copy of upstream `common/scanner.h`), because this repo's shared `vendored/common/scanner.h` belongs to the cfml/fsharp grammars and differs. The generated `parser.c`/`scanner.c` are otherwise byte-for-byte upstream — on re-vendor, re-apply only that single include rename.
 - **mojo** (added 2026-07-01): vendored from [lsh/tree-sitter-mojo](https://github.com/lsh/tree-sitter-mojo) @ `33193a99afe6` — MIT, ABI 15. Helix tracks `lsh/tree-sitter-mojo` as its Mojo grammar source, but the Helix-pinned commit (`3d7c53b8038f`) no longer resolves in the upstream repository after a force-push, so this vendor uses current upstream `main` rather than the stale registry SHA. Security review covered only the vendored C surface (`parser.c`, `scanner.c`, `tree_sitter/*.h`) plus upstream license/provenance metadata; no package manager hooks, workflow files, prompt/agent instruction files, or generated lockfiles were vendored.
+- **arkts** (added 2026-07-03): vendored from [Million-mo/tree-sitter-arkts](https://github.com/Million-mo/tree-sitter-arkts) @ `2fd0ad75e2d8` — MIT (c) 2024 million, ABI 15. Community grammar for HarmonyOS ArkTS (`.ets` files), not tracked by nvim-treesitter or Helix. Upstream declares MIT in `grammar.js` header (`@license MIT`, `@author million`) and `package.json` (`"license": "MIT"`, `"author": {"name": "million"}`), but ships no standalone LICENSE file — the vendored `LICENSE` is reconstructed from this declaration. The grammar lacks `field('name', ...)` mappings on declaration nodes, so definition extraction uses `cbm_find_child_by_kind` fallbacks and `build_method` name synthesis (see Custom extraction handling table). Security review covered only the vendored C surface (`parser.c`, `tree_sitter/parser.h`) plus upstream license/provenance metadata; the grammar has no external scanner (`EXTERNAL_TOKEN_COUNT 0` in `parser.c`), so no `scanner.c` was vendored; no package manager hooks, workflow files, prompt/agent instruction files, or generated lockfiles were vendored.
 
 > ⚠️ **Pinned commit = the revision nvim-treesitter/Helix vendor** (battle-tested, canonical source), not bleeding-edge HEAD. When re-vendoring, update the pinned commit here.
 
@@ -35,6 +36,7 @@ Guarded by the `contract_all_grammars_in_graph` graph-breadth test in
 | grammar | custom handling |
 |---|---|
 | ada      | `resolve_func_name`: `subprogram_body`/`subprogram_declaration` → `procedure_specification`/`function_specification` child's `name` field |
+| arkts    | `resolve_func_name`: `function_declaration`/`decorated_function_declaration`/`method_declaration` → `identifier` child (no `name` field in grammar); `build_method` → synthesize `"build"` name; `extract_class_def`/`compute_class_qn`: `identifier` child fallback; `component_body` added to class body traversal; `Component` label for `component_declaration`/`decorated_export_declaration` |
 | cairo    | `resolve_func_name`: `function_definition`/`function_signature` → `identifier` child |
 | clojure  | `extract_lisp_def`: `(defn …)` / `(def …)` head-symbol forms in `list_lit` |
 | d        | `resolve_func_name`: `function_declaration` → `identifier` child |
@@ -70,6 +72,7 @@ Re-vendoring from upstream must re-apply these.
 | ada | 14 | briot/tree-sitter-ada | `6b58259a08b1` | VERIFIED-BOTH | ✅ |
 | agda | 14 | tree-sitter/tree-sitter-agda | `e8d47a6987ef` | VERIFIED-BOTH | ✅ |
 | apex | 14 | aheber/tree-sitter-sfapex | `3597575a4297` | VERIFIED-NVIM | ✅ |
+| arkts | 15 | Million-mo/tree-sitter-arkts | `2fd0ad75e2d8` | COMMUNITY | ✅ |
 | astro | 14 | virchau13/tree-sitter-astro | `213f6e6973d9` | VERIFIED-BOTH | ✅ |
 | awk | 14 | Beaglefoot/tree-sitter-awk | `34bbdc7cce8e` | VERIFIED-BOTH | ✅ |
 | bash | 15 | tree-sitter/tree-sitter-bash | `a06c2e4415e9` | VERIFIED-BOTH | ✅ |