Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[cdt-dev] RE: Question on Bug 237819

The problem node should cover everything that is not covered by another IASTNode. You should either get a problem declaration for the entire line: "int a = offsetof(...);" or a regular declaration for 'a' with the initializer containing a problem-_expression_ containing "offsetof(...);".
I guess this is not the case, which I consider to be a bug.

From: Josh Tessier [mailto:Josh_Tessier@xxxxxxxxxx]
Sent: Monday, June 23, 2008 4:20 PM
To: Schorn, Markus
Subject: Question on Bug 237819

Hi Markus,

Very quick question:

In the particular example I gave for the bug related to problem nodes, the information in the problem node was a small subset of the entire _expression_.

For example, if I gave
int a = offsetof(struct mystruct, variable);

IASTNode.getRawSignature() would only return struct and the location information would only point to the word struct. However, the rest of the line would be completely discarded. There is no evidence of a declaration or of the call to offsetof. Should the problem node contain the entire line/node or just that one word? Because as it stands, without grabbing the raw signature of the parent node, I haven't been able to obtain "int a = (...)" at all.


Back to the top