[PATCH] PR 19339 - Parser confusion between lambda and designated initializer
Faisal Vali
faisalv at yahoo.com
Fri Apr 4 09:17:28 PDT 2014
Hi rsmith, krememek,
http://llvm.org/bugs/show_bug.cgi?id=19339
Here is a reduced test case:
struct Foo {
template<class T> Foo(T) {}
};
Foo f3 { [i{0}]{} };
The Parser gets confused in bool Parser::MayBeDesignationStart() in trying to distinguish between a lambda or a designated initializer when parsing
[i{0}]
FYI: Sema does not yet implement (recently voted into C++14? or is it C++17 I forget) the proper deduction for one element in a brace initializer yet.
This is a very cursory rough stab (includes a use of goto) at addressing this - just to draw attention to where i think the problem lies. I will leave it to those who grok the Parser better than me to either help me refine it - or conjure up their own fix.
http://llvm-reviews.chandlerc.com/D3290
Files:
lib/Parse/ParseInit.cpp
Index: lib/Parse/ParseInit.cpp
===================================================================
--- lib/Parse/ParseInit.cpp
+++ lib/Parse/ParseInit.cpp
@@ -75,11 +75,24 @@
case tok::amp:
case tok::identifier:
case tok::kw_this:
+ case tok::numeric_constant:
// These tokens can occur in a capture list or a constant-expression.
// Keep looking.
ConsumeToken();
continue;
-
+
+ case tok::l_brace:
+ case tok::r_brace:
+ if (PP.getLangOpts().CPlusPlus11) {
+ // These tokens can occur in a capture list or a constant-expression.
+ // Keep looking.
+ // Vec{[i{0}]() { }};
+ // int a[10] = { [Literal{}] = 3, 4 };
+ ConsumeBrace();
+ continue;
+ } else {
+ goto default_label;
+ }
case tok::comma:
// Since a comma cannot occur in a constant-expression, this must
// be a lambda.
@@ -99,11 +112,13 @@
}
default:
+ default_label:
// Anything else cannot occur in a lambda capture list, so it
// must be a designator.
Tentative.Revert();
return true;
}
+
}
}
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D3290.1.patch
Type: text/x-patch
Size: 1158 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20140404/16b436c4/attachment.bin>
More information about the cfe-commits
mailing list