Hi, Thanks for your reply!
I think it's hard to say exactly which thing in the file is causing the problem--type objects often refer to each other. Could you try reducing the test case, removing as much of types.cc as possible while keeping the error? Hopefully then it will be clear whether the error is in Treehydra or Dehydra.
The test passes if I comment out the field "i" in "Boo" ("unsigned int i:31"). On the other hand, if I comment everything except the declaration of "struct Boo", the test still fails.
So it seems it's definitely the bit field.
This is a pretty minor error, by the way. It wouldn't affect most analyses, and if it doesn't affect yours, you might want to just ignore it.
Right.
On the subject of tags, I think the latest revision of Dehydra is always the best, for now.
Ah, probably. I just needed to give people a recipe which will reproduce exactly what I did, rather than pick the head version of the day.
Lassi _______________________________________________ dev-static-analysis mailing list [email protected] https://lists.mozilla.org/listinfo/dev-static-analysis
