From 4676cd42004465bb22ab66efe4e582ca3317b74e Mon Sep 17 00:00:00 2001
From: Max Brunsfeld <maxbrunsfeld@gmail.com>
Date: Fri, 15 Mar 2024 16:55:21 -0700
Subject: [PATCH] Defer creation of external scanner until the beginning of
 parse

---
 lib/src/parser.c | 49 ++++++++++++++++++++++++++----------------------
 1 file changed, 27 insertions(+), 22 deletions(-)

diff --git a/lib/src/parser.c b/lib/src/parser.c
index 6421edd2..9d74c1ae 100644
--- a/lib/src/parser.c
+++ b/lib/src/parser.c
@@ -1827,6 +1827,7 @@ static unsigned ts_parser__condense_stack(TSParser *self) {
 
 static bool ts_parser_has_outstanding_parse(TSParser *self) {
   return (
+    self->external_scanner_payload ||
     ts_stack_state(self->stack, 0) != 1 ||
     ts_stack_node_count_since_error(self->stack, 0) != 0
   );
@@ -1889,7 +1890,7 @@ const TSLanguage *ts_parser_language(const TSParser *self) {
 }
 
 bool ts_parser_set_language(TSParser *self, const TSLanguage *language) {
-  ts_parser__external_scanner_destroy(self);
+  ts_parser_reset(self);
   ts_language_delete(self->language);
   self->language = NULL;
 
@@ -1908,8 +1909,6 @@ bool ts_parser_set_language(TSParser *self, const TSLanguage *language) {
   }
 
   self->language = ts_language_copy(language);
-  ts_parser__external_scanner_create(self);
-  ts_parser_reset(self);
   return true;
 }
 
@@ -1966,8 +1965,9 @@ const TSRange *ts_parser_included_ranges(const TSParser *self, uint32_t *count)
 }
 
 void ts_parser_reset(TSParser *self) {
-  if (self->language && self->language->external_scanner.deserialize) {
-    self->language->external_scanner.deserialize(self->external_scanner_payload, NULL, 0);
+  ts_parser__external_scanner_destroy(self);
+  if (self->wasm_store) {
+    ts_wasm_store_reset(self->wasm_store);
   }
 
   if (self->old_tree.ptr) {
@@ -2006,24 +2006,29 @@ TSTree *ts_parser_parse(
 
   if (ts_parser_has_outstanding_parse(self)) {
     LOG("resume_parsing");
-  } else if (old_tree) {
-    ts_subtree_retain(old_tree->root);
-    self->old_tree = old_tree->root;
-    ts_range_array_get_changed_ranges(
-      old_tree->included_ranges, old_tree->included_range_count,
-      self->lexer.included_ranges, self->lexer.included_range_count,
-      &self->included_range_differences
-    );
-    reusable_node_reset(&self->reusable_node, old_tree->root);
-    LOG("parse_after_edit");
-    LOG_TREE(self->old_tree);
-    for (unsigned i = 0; i < self->included_range_differences.size; i++) {
-      TSRange *range = &self->included_range_differences.contents[i];
-      LOG("different_included_range %u - %u", range->start_byte, range->end_byte);
-    }
   } else {
-    reusable_node_clear(&self->reusable_node);
-    LOG("new_parse");
+    ts_parser__external_scanner_create(self);
+    if (self->has_scanner_error) goto exit;
+
+    if (old_tree) {
+      ts_subtree_retain(old_tree->root);
+      self->old_tree = old_tree->root;
+      ts_range_array_get_changed_ranges(
+        old_tree->included_ranges, old_tree->included_range_count,
+        self->lexer.included_ranges, self->lexer.included_range_count,
+        &self->included_range_differences
+      );
+      reusable_node_reset(&self->reusable_node, old_tree->root);
+      LOG("parse_after_edit");
+      LOG_TREE(self->old_tree);
+      for (unsigned i = 0; i < self->included_range_differences.size; i++) {
+        TSRange *range = &self->included_range_differences.contents[i];
+        LOG("different_included_range %u - %u", range->start_byte, range->end_byte);
+      }
+    } else {
+      reusable_node_clear(&self->reusable_node);
+      LOG("new_parse");
+    }
   }
 
   self->operation_count = 0;