From: Scott Wolchok <swolchok@fb.com>
Date: Wed, 15 Sep 2021 16:55:02 +0000 (-0700)
Subject: [PyTorch] Avoid extra std::vector in parseSchemaOrName (#64678)
X-Git-Tag: accepted/tizen/8.0/unified/20231005.095509~197
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=bdbc622988ad9abb22622235bf3f4ad7a9e7ea30;p=platform%2Fupstream%2Fpytorch.git

[PyTorch] Avoid extra std::vector in parseSchemaOrName (#64678)

Summary:
Pull Request resolved: https://github.com/pytorch/pytorch/pull/64678

We know we only want one declaration, so let's not create an excess std::vector (and thus a heap allocation) for that.
ghstack-source-id: 138036978

Test Plan: CI

Reviewed By: dhruvbird, tugsbayasgalan

Differential Revision: D30813785

fbshipit-source-id: c67e0100cdef5d894282939fb6d39a57309bc240
---

diff --git a/torch/csrc/jit/frontend/function_schema_parser.cpp b/torch/csrc/jit/frontend/function_schema_parser.cpp
index 1656dda..927eae2 100644
--- a/torch/csrc/jit/frontend/function_schema_parser.cpp
+++ b/torch/csrc/jit/frontend/function_schema_parser.cpp
@@ -122,6 +122,13 @@ struct SchemaParser {
     return results;
   }
 
+  either<OperatorName, FunctionSchema> parseExactlyOneDeclaration() {
+    auto result = parseDeclaration();
+    L.nextIf(TK_NEWLINE);
+    L.expect(TK_EOF);
+    return result;
+  }
+
   Argument parseArgument(size_t idx, bool is_return, bool kwarg_only) {
     auto p = type_parser.parseType();
     auto type = std::move(p.first);
@@ -319,7 +326,7 @@ struct SchemaParser {
 
 C10_EXPORT either<OperatorName, FunctionSchema> parseSchemaOrName(
     const std::string& schemaOrName) {
-  return SchemaParser(schemaOrName).parseDeclarations().at(0);
+  return SchemaParser(schemaOrName).parseExactlyOneDeclaration();
 }
 
 C10_EXPORT FunctionSchema parseSchema(const std::string& schema) {