[cfe-dev] hasAncestor matching nothing

Farzad Sadeghi via cfe-dev cfe-dev at lists.llvm.org
Fri Nov 25 00:47:58 PST 2016


the two in this part of the AST:
|   | |-DeclRefExpr 0x4e52ae8 <col:2> 'short' lvalue Var 0x4e529e8 'sum' 'short'
|   | `-ImplicitCastExpr 0x4e52bb8 <col:8, col:12> 'short' <IntegralCast>
|   |   `-BinaryOperator 0x4e52b90 <col:8, col:12> 'int' '+'
|   |     |-ImplicitCastExpr 0x4e52b60 <col:8> 'int' <LValueToRValue>
|   |     | `-DeclRefExpr 0x4e52b10 <col:8> 'int' lvalue ParmVar
0x4e527d8 'a' 'int'

On 11/25/16, Manuel Klimek <klimek at google.com> wrote:
> which declRefExpr do you expect to be a child of which other declRefExpr?
>
> On Fri, Nov 25, 2016 at 9:40 AM Farzad Sadeghi <thabogre at gmail.com> wrote:
>
>> thank you for the explanation. I do realize that, but that is clearly
>> not the case here.
>>
>> On 11/25/16, Manuel Klimek <klimek at google.com> wrote:
>> > On Fri, Nov 25, 2016 at 9:24 AM Farzad Sadeghi via cfe-dev <
>> > cfe-dev at lists.llvm.org> wrote:
>> >
>> >> for the source code below:
>> >> short int test16(int a, int b)
>> >> {
>> >>         short int sum;
>> >>         short unsigned int sumus;
>> >>         sum = a + b;
>> >>         sumus = a + b;
>> >>
>> >>         return sum;
>> >> }
>> >> with AST:
>> >> |-FunctionDecl 0x4e52920 <line:821:1, line:829:1> line:821:11 test16
>> >> 'short (int, int)'
>> >> | |-ParmVarDecl 0x4e527d8 <col:18, col:22> col:22 used a 'int'
>> >> | |-ParmVarDecl 0x4e52848 <col:25, col:29> col:29 used b 'int'
>> >> | `-CompoundStmt 0x4e52d60 <line:822:1, line:829:1>
>> >> |   |-DeclStmt 0x4e52a48 <line:823:2, col:15>
>> >> |   | `-VarDecl 0x4e529e8 <col:2, col:12> col:12 used sum 'short'
>> >> |   |-DeclStmt 0x4e52ad0 <line:824:2, col:26>
>> >> |   | `-VarDecl 0x4e52a70 <col:2, col:21> col:21 used sumus 'unsigned
>> >> short'
>> >> |   |-BinaryOperator 0x4e52bd0 <line:825:2, col:12> 'short' '='
>> >> |   | |-DeclRefExpr 0x4e52ae8 <col:2> 'short' lvalue Var 0x4e529e8
>> >> 'sum'
>> >> 'short'
>> >> |   | `-ImplicitCastExpr 0x4e52bb8 <col:8, col:12> 'short'
>> <IntegralCast>
>> >> |   |   `-BinaryOperator 0x4e52b90 <col:8, col:12> 'int' '+'
>> >> |   |     |-ImplicitCastExpr 0x4e52b60 <col:8> 'int' <LValueToRValue>
>> >> |   |     | `-DeclRefExpr 0x4e52b10 <col:8> 'int' lvalue ParmVar
>> >> 0x4e527d8 'a' 'int'
>> >> |   |     `-ImplicitCastExpr 0x4e52b78 <col:12> 'int' <LValueToRValue>
>> >> |   |       `-DeclRefExpr 0x4e52b38 <col:12> 'int' lvalue ParmVar
>> >> 0x4e52848 'b' 'int'
>> >> |   |-BinaryOperator 0x4e52ce0 <line:826:2, col:14> 'unsigned short'
>> >> '='
>> >> |   | |-DeclRefExpr 0x4e52bf8 <col:2> 'unsigned short' lvalue Var
>> >> 0x4e52a70 'sumus' 'unsigned short'
>> >> |   | `-ImplicitCastExpr 0x4e52cc8 <col:10, col:14> 'unsigned short'
>> >> <IntegralCast>
>> >> |   |   `-BinaryOperator 0x4e52ca0 <col:10, col:14> 'int' '+'
>> >> |   |     |-ImplicitCastExpr 0x4e52c70 <col:10> 'int' <LValueToRValue>
>> >> |   |     | `-DeclRefExpr 0x4e52c20 <col:10> 'int' lvalue ParmVar
>> >> 0x4e527d8 'a' 'int'
>> >> |   |     `-ImplicitCastExpr 0x4e52c88 <col:14> 'int' <LValueToRValue>
>> >> |   |       `-DeclRefExpr 0x4e52c48 <col:14> 'int' lvalue ParmVar
>> >> 0x4e52848 'b' 'int'
>> >> |   `-ReturnStmt 0x4e52d48 <line:828:2, col:9>
>> >> |     `-ImplicitCastExpr 0x4e52d30 <col:9> 'short' <LValueToRValue>
>> >> |       `-DeclRefExpr 0x4e52d08 <col:9> 'short' lvalue Var 0x4e529e8
>> >> 'sum' 'short'
>> >> when i run the below matcher on clang-query:
>> >> match declRefExpr(hasAncestor(declRefExpr()))
>> >> i get no matches, while in the AST it clearly says that there are such
>> >
>> > nodes. so I was wondering what could be causing this.
>> >>
>> >
>> > Perhaps we can make the printing easier to read here:
>> > -a
>> > ||-b
>> > |`-c
>> > does not mean that c is a child of b, it means (if I'm not mistaken)
>> that c
>> > is the last child of a, but both b and a are equally children of a.
>> >
>> >
>> >> --
>> >> Farzad Sadeghi
>> >> _______________________________________________
>> >> cfe-dev mailing list
>> >> cfe-dev at lists.llvm.org
>> >> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>> >>
>> >
>>
>>
>> --
>> Farzad Sadeghi
>>
>


-- 
Farzad Sadeghi



More information about the cfe-dev mailing list