Is there any guarantee on the order of the children of the AST nodes ?
Or is it just some fixed but unspecified order ?
To give some context I am working on shrinking the size of the Stmt/Expr
nodes (patches coming soon I hope!) and I have been hitting various pieces
of codes in Sema (in JumpScopeChecker to be precise) which depends on some
children being before some other children.
What I am afraid of is subtly breaking the code in some not immediately
visible way (i.e. not detected in tests) because of this.